Masaryk University Faculty of Economics and Administration Department of Finance and Institute for Financial Market European Financial Systems 2016 Proceedings of the 13th International Scientific Conference June 27-28,2016 Brno, Czech Republic Suggested citation: AUTHOR, A. Title of the paper. In: European Financial Systems 2016. Proceedings of the 13th International Scientific Conference, Brno: Masaryk University, 2016, pp. xx-xx. ISBN 978-80-210-8308-0, or ISBN 978-80-210-8309-7 (online : pdf). Editors: Jan Krajíček Josef Nešleha Karel Urbanovský © 2016 Masarykova univerzita ISBN 978-80-210-8308-0, or ISBN 978-80-210-8309-7 (online : pdf). All published papers have been reviewed before publishing. The proceedings have not been amended or proofread. Editors are not responsible for the language used in the papers. Program Committee of the Conference Chair: Ing. PetrValouch, Ph.D. (Masaryk University) Members: Prof. Dr. Ing. Dana Dluhošová (VŠB - Technical University of Ostrava) Prof. Ing. Eva Horvátová, CSc. (University of Economics in Bratislava) Prof. Ing. Juraj Němec, CSc. (Matej Bel University) Assoc. Prof. Ing. Petr Dvořák, Ph.D. (University of Economics, Prague) Assoc. Prof. Mag. Dr. Peter R. Haiss, MBA (Vienna University of Economics and Business) Assoc. Prof. Ing. Peter Krištofík, Ph.D. (Matej Bel University) Assoc. Prof. Ing. Danuše Nerudová, Ph.D. (Mendel University Brno) Assoc. Prof. Ing. Erika Pastoráková, Ph.D. (University of Economics in Bratislava) Assoc. Prof. Antonin Rusek, Ph.D. (Susquehanna University, USA) Assoc. Prof. Ing. Jaroslav Sedláček, CSc. (Masaryk University) Assoc. Prof. Ing. Jitka Veselá, Ph.D. (University of Economics, Prague) Assoc. Prof. Ing. Eva Vávrová, Ph.D. (Mendel University Brno) Assoc. Prof. Ing. Jana Vodákova, Ph.D. (University of Defense Brno) Dr. Piotr Tworek (University of Economics in Katowice) Organizing Committee of the Conference Chair: Dr. Jan Krajíček Members: Dr. Veronika Kajurová Ms. Bohuslava Doláková Ms. Hana Florianová Mr. Josef Nešleha Mr. Tomáš Plíhal Ms. Martina Sponerová Mr. Karel Urbanovský Conference website: www.efs.econ.muni.cz PREFACE Dear readers, It is my pleasure to introduce you a collection of papers from the 13th annual international scientific conference, European Financial Systems 2016, organized annually by the Department of Finance of the Faculty of Economics and Administration, Masaryk University in Brno, Czech Republic. This year has been famous for various areas of topics, ranging from most common topics, such as accounting, taxation, financial markets, financial literacy, financial education, corporate finance, public finance, banks and insurance companies to more specific areas, such as application of econometrics models, foreign investments and current trends related to economics, data processing and state policies. As the collection of papers presents the latest scientific knowledge in a number of areas, I believe you will get a number of new insights usable both for your scientific, and educational or practical activities. It is also my pleasant duty to invite you for the 14th year of this conference, held in Brno, Czech Republic, in 2017. I wish you a pleasant reading Petr Valouch Chairman of the Program Committee CONTENTS Luděk Benada, Dagmar Linnertová HEDGING OF NATURAL GAS ON SELECTED MARKETS...................................................15 Karina Benetti BANKRUPTCIES OF COMPANIES IN THE CZECH REPUBLIC AFTER NEW FINANCIAL CRISIS...........................................................................................................................................21 Miloš Bikár, Katarina Vávrová, Mariana Sedliačiková WOULD THE RUSSIAN ECONOMY TURN INTO A LOST DECADE?.................................26 Gábor Bota, Mihály Ormos OIL PRICE AND EUROPEAN STOCK MARKETS..................................................................34 Roman Brauner SPECIFIC FACTORS OF THE CONTEMPORARY DEVELOPMENT OF THE CZECH REAL ESTATE MARKET......................................................................................................................43 Zuzana Brokešová, Erika Pastoráková, Tomáš Ondruška TESTING FOR INFORMATION ASYMMETRY IN AUTOMOBILE INSURANCE: SAMPLE FROM SLOVAK REPUBLIC.......................................................................................................50 Emilia Brozyna, Grzegorz Michalski, Guenter Blendinger, Ahmed Ahmidat THE LIMITATIONS OF E-COMMERCE DEVELOPMENT IN FULL OPERATING CYCLE FIRMS: V4 COUNTRIES CASE..................................................................................................57 Emil Burak, Juraj Nemec OPTIMISING THE SLOVAK TAX POLICY AND TAX SYSTEM PERFORMANCE.............65 Liběna Černohorská COMPARISON OF THE EFFICIENCY OF SELECTED EUROPEAN BANKING SECTORS 72 Karolina Daszynska-Zygadlo, Tomasz Stonski, Magdalena Ligus INVESTMENT IN RENEWABLE ENERGY TECHNOLOGIES FROM THE PERSPECTIVE OF POLISH VENTURE CAPITAL FUNDS...............................................................................79 Karolina Daszynska-Zygadlo SUSTAINABLE VALUE CREATION - PERFORMANCE OF EUROPEAN MANUFACTURING COMPANIES.............................................................................................86 Oleg Deev, Martin Hodula SYSTEMIC RISK INDICATORS IN THE EUROZONE: AN EMPIRICAL EVALUATION ... 94 Oleg Deev, Vlad Morosan THE IMPACT OF CONTINGENT CONVERTIBLE BOND ISSUANCE ON BANK CREDIT RISK............................................................................................................................................102 Jaroslava Dittrichová, Libuše Svobodová, Ivan Soukal HEDGING CASE STUDY IN THE EXCHANGE RATE COMMITMENT REGIME ENVIRONMENT........................................................................................................................Ill Bohuslava Doláková, Jan Krajíček INFLUENCES ON CONSUMER RATIONALITY....................................................................119 Nadiya Dubrovina, Jana Peliova, Erika Neubauerova ANALYSIS OF THE RELATIONSHIP BETWEEN TAXES AND SOCIAL BENEFITS AND TRANSFERS IN THE EU..........................................................................................................125 Eva Ducháčková THE ROLE OF LIFE INSURANCE IN THE CONTEXT OF COVER THE NEEDS OF THE PEOPLE IN THE CZECH REPUBLIC......................................................................................133 Vlastimil Farkašovský, Ľubomír Pinter QUALITY AND EFFICIENCY OF BANK BRANCH SERVICES............................................141 Luboš Fleischmann THE ACCESS TO INSTRUMENT OF COUNTERCYCLICAL CAPITAL RESERVES IN THE EUROPEAN UNION AND THE USA.......................................................................................149 Hana Florianová, Tomáš Dráb HEDGING OF PORTFOLIOS AND TRANSACTION COSTS...............................................157 Frunza Irina COMPARATIVE STUDY OF BANKING SECTOR IN REPUBLIC OF MOLDOVA AND CZECH REPUBLIC....................................................................................................................166 Beáta Gavurová, Tatiana Vagašová, Viliam Kováč COMPETITIVENESS ASSESSMENT OF SLOVAK REPUBLIC REGIONS.........................175 Maria Ginzburg, Nadezhda Yashina, Elena Ivanova KEYNESIAN MODEL IN SMALL AND MEDIUM ENTERPRISES DEVELOPMENT: PUZZLING CASE OF RUSSIAN REGIONS.............................................................................183 Ján Gogola, Ondřej Slavíček PENSION-RELATED APPLICATION OF THE COHORT LIFE TABLE.............................191 Eva Grmanová, Peter Hošták INFLUENCE OF SELECTED ENVIRONMENTAL FACTORS ON THE EFFICIENCY OF COMMERCIAL INSURERS......................................................................................................199 Vladimír Gvozdják, Božena Chovancová HOLDINGS OF GOVERNMENT BONDS BY COMMERCIAL BANKS DURING THE FINANCIAL AND DEBT CRISIS IN EUROPE........................................................................207 Peter Haiss, Franz Binder, Kushtrim Hajzeri, Wadim Kalmykov FOREIGN TRADE FINANCE: WHAT IS THE IMPACT OF THE GLOBAL FINANCIAL CRISIS OF 2007-2009?...........................................................................................................214 Taťána Hajdíková THE INFLUENCE OF THE SIZE OF THE REGION ON THE FINANCIAL SITUATION OF HOSPITALS...............................................................................................................................222 Eva Hamplová, Jaroslav Kovárník, Pavel Jedlička ANALYSIS OF VARIOUS ENTREPRENEURIAL ACTIVITIES AND THEIR DEVELOPMENT IN THE CZECH REPUBLIC FROM 2008 TO 2015...............................227 Martina Hedvicakova, Libuše Svobodova DEVELOPMENT AND THE CURRENT SITUATION OF THE MORTGAGES FOR THE CZECH HOUSEHOLDS.............................................................................................................234 Irena Honková THE PROCESS OF HARMONIZATION OF ACCOUNTING IN THE CZECH REPUBLIC. 242 Juraj Hruška, Oleg Deev HIGH-FREQUENCY TRADING AND PRICE VOLATILITY IN THE PARIS EURONEXT STOCK MARKET......................................................................................................................249 Jana Hvozdenska THE PREDICTION OF ECONOMIC ACTIVITY GROWTH BY SOVEREIGN BOND SPREAD IN FRANCE, GERMANY AND GREAT BRITAIN.................................................256 Barbora Chmelíková, Martin Svoboda DIFFERENCE IN FINANCIAL KNOWLEDGE OF FINANCE STUDENTS IN THE CZECH REPUBLIC..................................................................................................................................262 Věra Jančurová, Petra Formánková THE USE OF FINANCIAL ADVISORY IN CZECH REPUBLIC: SELF-CONFIDENCE.......267 Magdalena Jasiniak FACE NOMINAL EFFECT ON CAPITAL MARKET TRANSACTIONS. THE CASE OF POLAND.....................................................................................................................................272 Anna J^drzychowska, Ewa Poprawska COMPENSATION FOR INCOME LOST - LONG-TERM EFFECTS ON THE VICTIMS PERSONAL FINANCE..............................................................................................................278 Monika Kaczata CROP INSURANCE AS THE INSTRUMENT FOR RISK FINANCING IN POLISH FARMS .....................................................................................................................................................286 Silvie Kafkova, Dagmar Linnertova THE INFLUENCE OF A LOW INTEREST RATE ON LIFE INSURANCE COMPANIES ...294 František Kalouda IMPACT OF THE REPO RATE ON COMMERCIAL RATES IN THE CZECH REPUBLIC ...........................................................................................................................................300 Ôzcan Karahan THE INTERACTION BETWEEN VENTURE CAPITAL AND INNOVATION IN EUROPE .....................................................................................................................................................306 Maria Klimiková, Martina Muchová PREVENTING CRISES IN THE BANKING SECTOR AND THE ROLE OF INTERNAL AUDIT IN CORPORATE GOVERNANCE......................314 Monika Klimontowicz, Karolina Derwisz MOBILE TECHNOLOGY ON THE RETAIL BANKING MARKET......................................322 Magdalena Kludacz-Alessandri THE ROLE OF ACCOUNTING POLICY IN MANAGEMENT OF POLISH HOSPITALS ...331 Kristina Kočišova REVENUE EFFICIENCY IN EUROPEAN BANKING.............................................................339 Michal Kolář IMPACT OF TRANSFER PRICING REGULATION ON MNES' BEHAVIOUR...................349 Eva Kolářová, Eva Podolská THE ELECTRONIC RECORD OF SALES AND IMPACT ON THE REDUCTION OF TAX EVASION....................................................................................................................................357 Jaroslav Kovárník, Pavel Jedlička, Eva Hamplová THE COMPARISON OF THE SELECTED ASPECTS OF TAXATION IN VISEGRÁD FOUR COUNTRIES...............................................................................................................................365 Patrycja Kowalczyk-Rólczyňska, Tomasz Rólczyňski ALTERNATIVE INVESTMENTS IN VOLUNTARY PENSION SECURITY........................373 Jan Krajíček, Bohuslava Doláková IMPACT OF THE INTEREST RATES IN THE ECONOMY, THE BANKING AND FINANCIAL SYSTEM................................................................................................................380 Miroslav Krč, Vladimír Golik, Jana Vodákova ACCRUAL ACCOUNTING IN THE ARMY OF GREAT BRITAIN - CLARITY, TRANSPARENCY, AND ACCOUNTING DATA ANALYSIS OPTIONS...............................385 Michaela Krejčová, Milena Otavová, Jana Gláserová THE IMPACTS OF THE DIRECTIVE NO. 2013/34 / EU TRANSPOSITION INTO NATIONAL ACCOUNTING MODIFICATIONS IN THE CZECH AND SLOVAK REPUBLIC .....................................................................................................................................................392 Zuzana Křížová CURRENT ISSUES OF ACCOUNTING FOR INTANGIBLES IN VARIOUS REPORTING SYSTEMS...................................................................................................................................400 Zuzana Kubaščíková, Zuzana Juhászová ANALYSIS OF FINANCIAL STATEMENTS FOCUSING ON DETECTION OF PONZI SCHEMES USING XBRL...........................................................................................................408 Michal Kuběnka THE SUCCESS OF BUSINESS FAILURE PREDICTION USING FINANCIAL CREDITWORTHY MODELS...................................................................................................413 Lumír Kulhánek STOCK MARKET VOLATILITY IN THE EUROPEAN EMERGING AND FRONTIER MARKETS..................................................................................................................................420 Dušan Litva PERSPECTIVE OF SUSTAINABILITY OF FISCAL POLICY IN CZECH REPUBLIC.........428 Marina Malkina INTERRELATION BETWEEN CONCENTRATION IN BANKING SECTOR AND ITS MAIN PERFORMANCE INDICATORS: CASE OF RUSSIA.............................................................436 Marina Malkina, Rodion Balakin RISKS AND EFFICIENCY OF TAX SYSTEM AT DIFFERENT BUDGET SYSTEM LEVELS: REVENUE FORMATION AND SHARING IN THE RUSSIAN FEDERATION REGIONS .444 Anton Marci, Zuzana Juhaszova ANALYSIS OF THE RELATIONSHIP BETWEEN ADOPTING AND USING XBRL AS A REPORTING LANGUAGE FOR SMALL AND MEDIUM SIZED ENTITIES AND THE DEVELOPMENT STATUS OF THE COUNTRY AND THE EXISTENCE OF XBRL JURISDICTION..........................................................................................................................451 Slavomíra Martinková, Jakub Danko CORPORATE TAX REVENUES OF SELECTED EUROPEAN COUNTRIES USING DYNAMIC CONDITIONAL CORRELATION APPROACH...................................................459 Peter Mokrička THE CHARACTERISTICS OF THE INVESTMENT THROUGH THE AIRBAG CERTIFICATES.........................................................................................................................467 László Nagy, Mihály Ormos FRIENDSHIP OF STOCK INDICES.........................................................................................473 Danuše Nerudová, Veronika Solilová CCTB AND CCCTB IMPLEMENTATION AND ITS IMPACT ON THE TAX BASES ALLOCATED IN THE SLOVAK REPUBLIC...........................................................................481 Josef Nesleha, Karel Urbanovsky STUDY OF FINANCIAL LITERACY IN THE FIELD OF INSURANCE PRODUCTS..........490 Tetyana Nestorenko, Nadiya Dubrovina, Jana Péliová LOCAL ECONOMIC IMPACT OF DOMESTIC AND INTERNATIONAL STUDENTS: CASE OF UNIVERSITY OF ECONOMICS IN BRATISLAVA..........................................................496 Inka Neumaierová, Ivan Neumaier THE PERFORMANCE RANKING OF CHOSEN MANUFACTURING DIVISION..............502 Thi Anh Nhu Nguyen THE IMPACT OF DEMOGRAPHIC CHARACTERISTICS ON FINANCIAL LITERACY: AN EMPIRICAL STUDY IN COMMERCIAL BANKS' CUSTOMERS.........................................508 Magdaléna Osak PRIVATE HEALTH INSURANCE AND MEDICAL SUBSCRIPTIONS - TWO FACES OF THE PRIVATE PRE-PAID FUNDING OF HEALTH CARE IN POLAND............................517 Gabriela Oškrdalová PAYMENT CARD FRAUDS WITH A HIDDEN CAMERA, TOUCH SENSORS AND A COUNTERFEIT PAYMENT CARD AND PROTECTION TECHNIQUES AGAINST THESE TYPES OF FRAUDS..................................................................................................................526 Viera Pacáková, Pavla Jindrová, David Zapletal COMPARISON OF HEALTH CARE RESULTS IN PUBLIC HEALTH SYSTEMS OF EUROPEAN COUNTRIES........................................................................................................534 Dalibor Pánek THE POLICY OF MONETARY EASING OF CENTRAL BANKS..........................................542 Mário Papik COMPOSITION OF PENSION FUNDS' INVESTMENT PORTFOLIO AND ITS IMPACT ON PROFIT......................................................................................................................................547 Erika Pastoráková, Zuzana Brokešová, Tomáš Ondruška, Eleonóra Zsapková CONTRIBUTION TO THE RESEARCH FOR ADEQUATE AND SUSTAINABLE PENSIONS - THE STUDY OF THE SLOVAK REPUBLIC AND THE CZECH REPUBLIC....................556 Jan Péta, Mária Režňáková ECONOMIES OF SCALE IN M&A IN THE MANUFACTURING INDUSTRY IN THE CZECH REPUBLIC..................................................................................................................................564 Sergey Petrov, Oksana Kashina, Roman Murashkin EXAMINATION OF STOCK MARKET "TEMPERATURE" USING PRICE-DIVIDEND DEPENDENCE FOR EUROPEAN SHARES............................................................................572 Marlena Piekut PERSONAL FINANCE IN TERMS OF INCOME AND EXPENDITURE ASPECTS............581 Sylwia Pieňkowska-Kamieniecka, Damian Walczak WILLINGNESS OF POLISH HOUSEHOLDS TO SAVE FOR RETIREMENT.....................588 Michal Plaček, Milan Puček, František Ochrana, Milan Křápek APPLICATION OF DEA METHODS FOR EVALUATING EFFICIENCY IN MUSEUMS, GALLERIES, AND MONUMENTS IN THE CZECH REPUBLIC...........................................596 Michal Plaček, Martin Schmidt, František Ochrana, Milan Puček THE IMPACT OF USING AN EXTERNAL AUTHORITY ON THE QUALITY OF PUBLIC PROCUREMENT.......................................................................................................................602 Tomáš Plíhal FORECASTING EXCHANGE RATE VOLATILITY: SUGGESTIONS FOR FURTHER RESEARCH................................................................................................................................609 Lenka Přečková FUNCTIONING OF BANCASSURANCE IN SELECTED COUNTRIES IN WHICH THE FINANCIAL GROUP ERSTE GROUP BANK OPERATES....................................................614 David Procházka FINANCIAL PERFORMANCE OF CZECH SUBSIDIARIES UNDER CONTROL OF THE EU LISTED COMPANIES...............................................................................................................623 Zuzana Rakovská, Martin Svoboda PRACTICAL APPLICATION OF SENTIMENT INDICATORS IN FINANCIAL ANALYSIS: BEHAVIORAL FINANCE APPROACH...................................................................................630 Monika Raulinajtys-Grzybek, Katarzyna Frankowska, Wojciech Matusewicz THE IMPACT OF THE FEATURES OF HEALTHCARE PROVIDERS IN POLAND ON THEIR COSTS IN THE ACCOUNTING SYSTEM..................................................................638 Oldřich Rejnuš PREDICTION OF FUTURE DEVELOPMENT OF THE WORLD ECONOMY UNDER CONDITIONS OF NEGATIVE INTEREST RATES................................................................646 Tomasz Rólczyňski, Tomasz Kopyšciaňski ECONOMIC CONDITION OF THE EUROPEAN UNION COUNTRIES AND LEVEL OF RATING......................................................................................................................................654 Katarína Rentková, Monika Roštárová USING OF VENTURE AND EQUITY CAPITAL IN FINANCING OF SMES IN THE SLOVAK REPUBLIC..................................................................................................................................661 Petr Sed'a, Juan Antonio Jimber del Rio TESTING THE WEAK FORM OF EFFICIENCY ON CHINESE STOCK MARKET............669 Jaroslav Sedláček FINANCIAL STATEMENTS IN THE FINANCIAL DECISION MAKING............................678 Kateřina Seinerová FINANCIAL LITERACY OF ELEMENTARY SCHOOL PUPILS IN PARDUBICE..............686 Eugenia Sch mitt STRESS-TESTING MODEL FOR STRUCTURAL LIQUIDITY RISK..................................692 Elena Širá, Katarína Radvanská, Zuzana Grančaiová FACTORS INFLUENCING CLIENTS IN SELECTION OF INSURANCE COMPANY.........700 Martin Širůček, Karel Šíma THE OPTIMIZED INDICATORS OF TECHNICAL ANALYSIS BY ANTICYCLIC ASSETS .....................................................................................................................................................708 Ladislav Šiška BUDGETARY GAMING BEHAVIOR AND ITS DETERMINANTS.....................................715 Ľudomír Šlahor, Daniela Majerčáková, Mária Barteková IMPLICATIONS OF LOW/NEGATIVE INTEREST RATES FOR BANKS' ASSET AND LIABILITY MANAGEMENT - AN EXAMPLE.......................................................................721 Veronika Solilová, Danuše Nerudová, Hana Bohušova, Patrik Svoboda THE RECOMMENDATION OF SAFE HAVEN INTEREST RATES IN THE BEPS CONTEXT .....................................................................................................................................................728 Miroslav Sponer, Martina Sponerovä FOREIGN EXCHANGE INTERVENTION BY THE CZECH NATIONAL BANK AND ITS CONSEQUENCES......................................................................................................................737 Martina Sponerová, Miroslav Sponer NEXUS OF BANK RISK-TAKING AND INTEREST RATES................................................745 Tomáš Štofa, Martin Zoričak SELECTED SUCCESS FACTORS OF CROWDFUNDING PROJECTS.................................752 Petr Suchánek, Maria Králová REFLECTION OF CUSTOMER SATISFACTION IN SELECTED PERFORMANCE INDICATORS OF FOOD ENTERPRISES...............................................................................760 Veronika Šuliková, Marianna Siničáková, Ľubica Štiblárová, Slavomíra Šuliková EVALUATION OF THE BUSINESS CYCLE SYNCHRONISATION IN EUROPE...............768 Veronika Svatošová, Zuzana Svobodová DETERMINATION OF FINANCIAL STRATEGY OF SELECTED COMPANY AND ITS IMPORTANCE FOR OTHER BUSINESS DEVELOPMENT - CASE STUDY.....................776 Milan Svoboda, Pavla Říhová ALGORITHMIC TRADING USING MARKOV CHAINS: COMPARING EMPIRICAL AND THEORETICAL YIELDS...........................................................................................................787 Libuše Svobodová, Martina Hedvičáková COMPARISON OF BUILDING SAVINGS BANKS ON THE CZECH MARKET..................794 tukasz Szewczyk CO-OPERATIVE BANKS IN POLAND, CURRENT ISSUES.................................................802 Aleksandra Szpulak, Tomasz Michael JOINT DETERMINISTIC AND STOCHASTIC APPROACH TO CASH BALANCE MODELLING: A CASH MODEL SPECIFICATION AND VERIFICATION..........................808 Dariusz Urban SOVEREIGN WEALTH FUND OWNERSHIP AND FINANCIAL PERFORMANCE OF COMPANIES LISTED ON THE WARSAW STOCK EXCHANGE........................................816 Tomáš Urbanovský CONNECTION BETWEEN EXCHANGE RATE AND BALANCE OF PAYMENTS ACCOUNTS: THE CASE OF THE CZECH REPUBLIC..........................................................824 Lukáš Vartiak COMPARING FINANCIAL PERFORMANCE OF SLOVAK EXCELLENT COMPANIES ...832 Eva Vávrová, Svatopluk Nečas THE EVALUATION OF FINANCIAL HEALTH OF THE INSURANCE SECTOR IN THE WORLD'S INSURANCE CENTERS.........................................................................................839 Jana Vodákova, Nela Sglundová EXPENSES AND REVENUE CLASSIFICATIONS FOR MANAGERIAL PURPOSES IN THE CZECH STATE ADMINISTRATION UNITS..........................................................................848 Radoslaw Witczak THE INCORRECTNESS OF ESTIMATING OF TAX BASE IN INCOME TAXES IN THE VERDICTS OF SUPREME ADMINISTRATIVE COURT IN 2014 IN POLAND................856 Agnieszka Wojtasiak-Terech TYPOLOGY OF THE MUNICIPAL BONDS RISK - APPLICATION FOR POLISH ORGANIZED BONDS MARKET..............................................................................................863 Nadezhda Yashina, Maria Ginzburg, Louisa Chesnokova FISCAL FEDERALISM AND REDISTRIBUTE POLITICS FOR INCOME TAX: CASE OF RUSSIA'S REGIONS..................................................................................................................872 Alexander Zureck, Julius Reiter, Martin Svoboda CROSS-GENERATIONAL INVESTMENT BEHAVIOR AND THE IMPACT ON PERSONAL FINANCE....................................................................................................................................881 Hedging of Natural Gas on Selected Markets Luděk Benada1, Dagmar Linnertová2 1 Masaryk University Faculty of Economics and Administration Lipova 41a, 602 00 Brno, Czech Republic E-mail: benada@mail.muni.cz 2 Masaryk University Faculty of Economics and Administration Lipova 41a, 602 00 Brno, Czech Republic E-mail: dagmar.linnertova@mail.muni.cz Abstract: The following study focuses on one of the key primary energy commodities in the modern society. Two independent and different regional markets for natural gas are compared from the hedging point of view. The market of considered commodity is unique in its characteristics. In the paper will be examined the possibility of price risk hedging on two different markets with distinct level of market maturity and characteristics. One of markets is represented by the system of endogenous price mechanism while the other one is based on price indexed mechanism. The analysis will be carried out on US and European market of natural gas. In both cases, the methodology will be focused the short hedge position. Keywords: hedging, hedge ratio, spot, futures, forward JEL codes: C58, Gil 1 Introduction The paper focuses on a hedging strategy of natural gas. The object of hedging is the spot price. Our research is restricted to financial hedging. Physical hedging is complex due to diverse duration of assets (portfolio), Grammatikos et a. (1986). Thus, we focus our analysis only on prices. We will examine short hedging, that means we will hedged the commodity of the owner perspective. Hydrocarbons as an energy source are part of everyday life in the western world. The lack of fuels and consequently the lack of energy would lead to a breakdown of the social order to which modern man is accustomed. Growing volume and importance of NG stimulate the importance of market development. The commodity is regionally bound. On that account the different processing costs and the completive environment create a heterogeneous market structure. From this perspective, only one price of NG does not exist. At the very beginning the transactions of natural gas were limited in bilateral contracts. Later on, the market of the commodity was found due to liberalization. The formation of market is inextricably linked to the USA, Prmia (2007). The price of natural gas is mostly based on two different price mechanisms. Either the price is derived from the price of crude oil or the price is mediated through the supply-demand interaction. The former mechanism is also known as Oil Indexation, Roye (1997). This price mechanism is typical for Asia and Europe. Second aforementioned mechanism is referred as Gas on Gas competition. The Gas on Gas trading has tradition in the United States, Victor et al. (2006). Europe is recently trying to "correct" prices toward Gas on Gas competition, Ekins (2005). A disproportionality of a demand and a supply has caused considerable price uncertainty. Substantial price volatility has stimulated the emergence of the derivatives products such as futures and forwards. The initial issue of term contracts was hedging, albeit nowadays there are other incentives for their use. We will use derivatives for the purpose of risk reduction in spot on the markets with different price mechanism. We assume that the 15 market with Gas on Gas competition will exhibit greater risk. Hence, the application of hedging on this market should be more beneficial in comparison to Oil Indexed market. Further, we assume that the hedge effectiveness based on the methodology of hedge ratio will provide a higher reduction of risk rather than a simple naive portfolio. 2 Methodology and Data The purpose of a hedging strategy is to lock the position, i.e. minimalize risk deviance from expected value. In the other words, fluctuations in held asset should be compensated with a reverse spike of hedging instrument. That is an optimal situation, but because the assets are random variables with no same characteristics, it could not be assumed that for hedging only naive portfolio will be sufficient. Thus, the number of assets in the long position would match the number of hedging instruments in short position, Alexander (2007). There are several optimization techniques according to the objective function. It is common to use principles of portfolio optimization. The pioneer stabling the hedging ratio was the minimum variance approach, Johnson (1960). The ratio could be found using the tangential portfolio based on Sharpe ratio, Chen et al. (2013). If there is a problem with risk free rate or the rate is zero, Sharpe ratio reduces to minimum variance approach. One of the implied ways is application of OLS on spot and futures, Benning et al. (1984). However, the problem with dynamic volatility could occur, Miffre (2004). For this reason more sophisticated models like GARCH are applied, Baillie et al. (1991). Further, Gini coefficient or extended Gini coefficient is applied, respectively, Lerman (1984). Futures are considered being suitable instruments for hedging purposes because of their features. In the case of an asset or a portfolio it is necessary to find an asset or a portfolio with extreme value of correlation coefficients. In our analysis the futures or forward prices correspond with that of underlying. Then, we expect high correlation between them. The prerequisite of applicability is the stable mutual coverage of percentage change of assets in portfolio. Since the analysis is based on historical data, this assumption cannot be guaranteed. We use daily closing prices of spot, future and forward for the analysis. Future contracts are set one month ahead. The same time conditions will be considered for the forward contracts. The validation of hedging will be set at 10, 20, 30, 40, 50 and 60 days. As a representative market of Gas on Gas competition we use Henry Hub (HH) and futures related to the terminal traded in NYMEX. Index Title Transfer Facility (TTF) was chosen as a representative of Oil Indexed price. Due to availability for TTF we use forward contract. The analysis is based on the period from September 2014 to December 2015. Subsequent validation is carried out on data from the following three month. Initially, we validate hedging effectiveness based on hedging ratio from optimization criteria. Subsequently, a comparison with the results of a naive portfolio will be provided. Model Specification In our study we focus solely on short hedging. Thus, we act as seller of natural gas. Some authors highlight the hedging asymmetry in the meaning of the short and long position, Cotter et al. (2012). We analyze only one optimization approach represented by the calculation of hedge ratio of futures/forward to the spot. The ratio will be determined on the basis of optimization criteria. The objective function is express as variance of portfolio consisting of two assets (spot and futu res/forward). Equation for portfolio variance is: °p = of + h2 * of + 2 * h * aSf (1) 0 (4) dhdh i v ' Hedging efficiency is a measure of whether it is worth or not to provide hedging. We will not consider the costs of hedging, but we will assess its effect. Specifically, we measure the reduction in variance. It will be utilized in accordance with established methodology, Ederington (1979): HE = °2Vun-°he (5) aVun HE stands for hedging effectiveness, agun is the variance of the spot (unhedged portfolio) and a£he is the variance of the hedged portfolio. The greater the compensation due to joint percentage change, the closer the result will be to one. Since futu res/forward will enter into short position we have to express the variance of portfolio in the following form: °p = as + h2 * of — 2 * h * aSj, (6) The second moments and the covariance are obtained by the difference of the natural logarithm of closing prices. Afterward we compare the effectiveness of hedge ratio with the naive portfolio. 3 Results and Discussion In accordance with the methodology used in literature, we have to determine the hedge ration based on percentage change in the closing price of spot and futu res/forward. The analysis results identify a very small proportion of hedging instruments in the situation of gas on gas competition. The ratio was 0.278 in HH to 0.513 in TTF. The measurement of hedging effectiveness was applied to six scenarios - 10, 20, 30, 40, 50 and 60 days. The results of risk reduction in percentage are express in the following table. Table 1 Hedging Effectiveness - Henry Hub, TTF (ln_returns) Market/Days 10 20 30 40 50 60 HH 0.066588 0.16188625 0.165092 0.174023 0.182594 0.180732 TTF 0.672532 0.770394749 0.782857 0.743612 0.743678 0.752222 Source: Based on data EIA, Bloomberg 17 Figure 1 Hedging Effectiveness with Hedging Ratio ■ 10 »20 |3Q m&0 50 i60 1 hh ttf Source: Based on data from EIA, Bloomberg As it was possible to estimate the suitability of hedging in both markets, the US market does not show perfect options for a reduction of systematic risk. Conversely, the European market reached good results. It is rather surprising finding, because in the case of US futures the market was more accurate to spot price. More details are shown in the Figure 2. Figure 2 Closing Prices of Spot and Futures/Forward in HH and TTF tt i ■ Hi »i* a ib to w vt vii vtu u i a Kicnm kii fi in » v viviivnin x id Source: Based on data from EIA, Bloomberg The correlation of the spot HH and futures was 0.97. In the case of spot TTF and forward was only 0.82. But if we use returns the results change significantly. Thus, the TTF market decreased in correlation to 0.71. But even more dramatic change has occurred in HH, where the correlation dropped to 0.24. The situation shows following charts. Figure 3 Returns of spot and futures (forward) HH and TTF. Source: Based on data from EIA, Bloomberg From the preceding figures could be identified an assumption for hedging results. The risk of the price change is for TTF higher compared with HH, but if we consider the returns it is just the opposite. Moreover, it is possible to identify different level of risk between the spot and the futures or the forward. The hedging instrument compared with spot is more risky on the TTF market. The standard deviation was 0.018325 for the forward and 0.013259 for the spot. On the HH the standard deviation was 0.0284 for the futures and 0.0328 for the spot. 18 Finally, we measure the hedging effectiveness using naive portfolio. This strategy better reflects the relation between the spot and the hedging instrument, which corresponds more to the tighter movement of closing prices. Following table and chart show the details. Table 2 Hedging Effectiveness - Henry Hub, TTF (Naive portfolio) Market/Days 10 20 30 40 50 60 HH -0.3442 0.342523643 0.300956 0.308332 0.319771 0.300736 TTF 0.910612 0.849081236 0.894785 0.922918 0.920413 0.925488 Source: Based on data EIA, Bloomberg Figure 4 Hedging Effectiveness with Naive Portfolio DAYSbIO i20 b30 «40 50 ■ 60 1 -0,4 Source: Base on data from EIA, Bloomberg The naive portfolio has worse outcome only in one case in comparison to the approach, which is based on hedge ratio. It is the hedging strategy for 10 days on the HH. In this period the hedging with the naive portfolio is inefficient, because the variance of the portfolio is higher than the variance of spot. The results of naive portfolio overcomes all remaining scenarios compared with the results of hedge ratio. However, it was not possible to reach the results obtained on the European market, where the naive portfolio produced better performance to the previous results, as well. There is a wide field for further investigation. It would be interesting to compare the pipeline markets with LNG. Furthermore, compare the homogeneity markets in Europe with each other. From the methodological point of view it would be worth to use a comparison of different models for time series analysis. The models based on dynamic volatility, cointegration, wavelet decomposition or copula approach could be appropriate. 4 Conclusions In our paper we investigated the problems of natural gas hedging. We examined two different markets. Each market has specific mechanism determining the price. The US market represented by Henry Hub was based on Gas on Gas competition. The European market was covered by Title Transfer Facility with the price derived from crude oil. The analyzed data for hedge ratio covering daily closing prices from September 2014 to December 2015. Afterwards, validation were applied on daily prices from January to March 2016. The assumption of higher risk on the US market was confirmed. However, two further assumptions were falsified. The hedging effectiveness was not higher on the Henry Hub market. The European market was more suitable for hedging purposes. The part of explanation could lie in the effect of financialization. Thus, a large part of the trade's 19 volume in finance derivatives is not used for hedging purposes. Moreover the results of naive portfolio led to unforeseen conclusion. The strategy of naive portfolio was more efficient on both markets then the methodology based on hedge ratio. Nevertheless, to determine the unambiguous conclusion it is necessary to examine the issue in more details. Acknowledgments The support of the Masaryk University internal grant MUNI/A/1025/2015 Risks and Challenges of the Low Interest Rates Environment to Financial Stability and Development is gratefully acknowledged. References Alexander, C, Barbosa, A. (2007). Effectiveness of minimum-variance hedging. Journal of Portfolio Management, vol. 33(2), pp. 46-59. Baillie, R. T., Myers, R. J. (1991). Bivariate GARCH estimation of the optimal commodity futures hedge. Journal of Applied Econometrics, vol. 6 (2), pp. 109-124. Benninga, S., Eldor, R., Zilcha, I. (1984). The optimal hedge ratio in unbiased futures markets. Journal of futures markets, vol. 4(2), pp. 155-159. Chen, S.-S., Lee, C. F., Shrestha, K. (2013). Futures hedge ratios: a review. Encyclopedia of Finance, pp. 871-890. Cotter, J., Hanly, J.(2012). Hedging effectiveness under conditions of asymmetry. The European Journal of Finance, vol. 18 (2), pp. 135-147. Ederington, L. H. (1979). The hedging performance of the new futures markets. The Journal of Finance, vol. 34(1), pp. 157-170. Ekins, P., Bradshaw, M., Watson, J. (2015). Global Energy: Issues, Potentials, and Policy Implications, 1 st ed. Oxford: Oxford University Press. Grammatikos, T., Saunders, A. (1986). Futures price variability: A test of maturity and volume effects, Journal of Business, pp. 319-330. Johnson, L. L. (1960). The theory of hedging and speculation in commodity futures. The Review of Economic Studies, vol. 27(3), pp. 139-151. Miffre, J. (2004). Conditional OLS minimum variance hedge ratios, Journal of Futures Markets, vol. 24(10), pp. 945-964. Prmia, P. (2007). Guide to the Energy Markets: Introduction to Natural Gas Trading, 1st ed. McGraw-Hill Education. Rojey A., J., C. (1997). Natural Gas: Production, Process, 1st ed. Paris: Transport Editions Technip. Roncoroni, A., Fusai, G., Cummins, M. (2015). Handbook of Multi-Commodity Markets and Products: Structuring, Trading and Risk Management,1st ed. Wiley. Taverne, B. (2008). Petroleum, industry, and governments: a study of the involvement of industry and governments in the production, and use of petroleum, 1st ed. Kluwer Law International. TechReport (2015). Wholesale Gas Price Survey: A global review of price Forforma mechanism 2005 - 2014, International Gas Union. Victor, D., Jaffe, A., Hayes, M. (2006). Natural Gas and Geopolitics: From 1970 to 2040, 1st ed. Cambridge: Cambridge University Press. 20 Bankruptcies of Companies in the Czech Republic after New Financial Crisis Karina Benetti1 1 ŠKODA AUTO VYSOKÁ ŠKOLA O.p.S. Department of Finance and Accounting Na Karmeli 1457, 293 01 Mladá Boleslav, Czech Republic E-mail: karina.benetti@savs.cz Abstract: In 2006 was published a law no. 182/2006 Coll., On Bankruptcy and Its Resolution (Insolvency Act), which came into effect from January 1st, 2008. This law came into effect just at a time when the new financial crisis started getting stronger. This crisis meant for bankruptcies of companies is an important milestone, not only with regard to the number of bankrupt companies, but especially to the emergence of new ISO standards dealing with much needed risk management in business practice. The value of bankrupt companies was according to statistics since the beginning of the recent financial crisis until the end of 2008 in the amount of US $ 14.5 trillion, which is more than 145 times the amount for the Marshall Plan to rebuild Europe after World War II. In the Czech Republic from 2008 to 2012 to increase the number of corporate bankruptcies by 288%, but their numbers began from 2012 to decline. The aim of this paper is to analyze the number of corporate insolvency proposals and corporate auditions in the Czech Republic since the beginning of the recent financial crisis until the present (i.e. the last eight years). Keywords: corporate insolvency proposals, corporate auditions, dependence analysis, financial health, ISO standards JEL codes: CIO, C39, G01, G31, G33 1 Introduction The effects of the recent financial crisis can be divided into positive and negative effects. The negative effects have already been published many scientific but also non-sicentific works. One of the negative effects of the recent financial crisis, the number of bankruptcies, not only personal, but also corporate. In the Czech Republic published analysis results of personal bankruptcies in recent years Bokšová et al (2014), Hospodka et al (2015) and Maixner et al (2014). According to a study Pittman & Ivry (2009) the total value of bankrupt companies at the end of 2008, in amount of 14.5 trillion USD. The predominant reason for these bankruptcies Barbulescu et al (2015) was the absence of risk management in interim management or its poor quality. Application of quality risk management into internal enterprise policy can mean early detection of financial problems of these companies. In this view are very important bankruptcy models (Čámská, 2012) and basic characteristics of enterprises which are in insolvency (Čámská, 2013). Detailed analysis of corporate insolvency during the crisis years (with data analysis from 2008 to 2013) published Kislingerová and Schoenfeld (2014) and forecasts of corporate insolvencies for the period 2013-2017 published in 2013 Kislingerová. Absence of quality risk management was one of the main reasons to the creation of ISO 31000 (2009) for risk management in business practice, which can be considered a positive effect of the recent financial crisis. With regard to the reasons for the changes in risk management in corporate practice, this paper will focus precisely analyzing the number of corporate insolvency proposals and corporate auditions in the Czech Republic after the crisis period. The research questions are: "How are monitored variables develops over time?" and "Is there dependence between these two variables?". To answer these questions, it is necessary to analyze: the development of proposals of corporate insolvencies and corporate auditions in the Czech Republic for the period January 2008 to December 2015 and do the dependence analysis. 21 2 Methodology and Data In the research were particular used scientific methods: induction, comparative analysis, synthesis of partial knowledge, elementary statistical analysis and dependence analysis. For elementary statistical analysis was used the following selected three indicators (Hindis, etal, 2000): • the first difference (absolute gain, iAt iN - for corporate insolvencies and iAt Au -for corporate auditions) (1) A=a,-Vi- (!) • the average absolute gain (2) and n El A, ^ _ t=2 _yn~ y\ n-l n-l (2) • the average growth coefficient (3) k ■ (3) where n is the number of values (in this paper n = 96). For the dependence analysis was used software STATGRAPHICS Centurion XVI. For the analysis was used secondary data from Creditreform (2016). 3 Results and Discussion The results of elementary statistical analysis, by selected three characteristics, of development of number of corporate insolvency proposals and corporate auditions are given below. The basic development of number of corporate insolvency proposals and corporate auditions with development of its first difference illustrated Figure 1. Figure 1 Development of Number of Corporate Insolvency Proposals and Corporate Auditions with Development of its First Difference 1/2006 1/2009 1/2010 1/2011 1/2012 1/2013 1/2014 1/2015 1000 250 950 * The number of corporate Insolvency proposals -•— lit IN The number of corporate auditions lit AU Source: Own from Creditreform (2016) 22 According to the development of the values specified in Figure 1 can be deduced that the observed characteristics were examined over a period of very fluctuating development. For this reason, does not make sense to describe the examined values other statistical characteristics (such as e.g. coefficient growth, growth rate and increase rate). For a basic overview of the development of the examined values sufficient to indicate the results of absolute average gain and average growth coefficient. The result of average absolute gain is for corporate insolvencies proposals 1.37 and for corporate auditions 1.42. The results of average growth coefficient are for corporate insolvencies proposals 1.008416 (which corresponds to 0.841 %). and for corporate auditions 1.03379 (which corresponds to 3.379 %). For dependence analysis will be used monthly data. Firstly, was conducted multiple variable analysis, summary statistics illustrated Table 1, results from correlations shows Table 2 and Figure 2. Table 1 The Summary Statistics from Multiple Variable Analysis The Number of The Number of Corporate Insolvencies Corporate Proposals Auditions Count 96 96 Average 437.198 152.354 Standard deviation 158.31 40.95 Coefficient of 36.2102% 26.8782% variation Minimum 106.0 6.0 Maximum 903.0 228.0 Range 797.0 222.0 Standard skewness 1.71433 -3.20719 Standard kurtosis -0.474946 2.47394 Source: Own elaboration The table 1 shows summary statistics for each of the selected data variables. It includes measures of central tendency, measures of variability, and measures of shape. Of particular interest here are the standardized skewness and standardized kurtosis, which can be used to determine whether the sample comes from a normal distribution. Values of these statistics outside the range of -2 to +2 indicate significant departures from normality, which would tend to invalidate many of the statistical procedures normally applied to this data. Table 2 Correlation The Number of The Number of Corporate Insolvencies Corporate Auditions Proposals The Number of 0.1002 (Correlation) Corporate Insolvencies (96) (Sample Size) Proposals 0.3312 (P-Value) The Number of 0.1002 (Correlation) Corporate Auditions (96) 0.3312 Source: Own elaboration The table 2 shows Pearson product moment correlations between each pair of variables. In this case is Pearson product moment correlation 0.1002. Pearson product moment correlation coefficients range between -1 and +1 and measure the strength of the linear relationship between the variables. Also shown in parentheses is the number of pairs 23 of data values used to compute each coefficient. The third number in each location of the table is a P-value which tests the statistical significance of the estimated correlations. P-values below 0.05 indicate statistically significant non-zero correlations at the 95.0% confidence level. None form the analyzed pairs of variables have P-values below 0.05. This is the reason, why is for this case the Pearson product moment correlations indicator not correct, we must use for example Spearman correlation coefficient. However, the results shown in Table 2 cannot be properly assessed without visualization - see Figure 2. Figure 2 Scatterplot Matrix The Number of Corporate □ o Inotvincit* Proposal cP □ trj ödü I- -1 a £°S-8p° Bä$&B The Number of Corporate AudCnra □ □ I- a D Source: Own elaboration From the results in the Figure 1 it is evident that between variables is no correlation. Certainty as to whether between variables is or is not correlation will bring a result of the Spearman rank correlation, the results are given in Table 3. Table 3 Spearman Rank Correlations The Number of The Number of Corporate Insolvencies Corporate Auditions Proposals The Number of -0.0259 (Correlation) Corporate Insolvencies Proposals (96) (Sample Size) 0.8006 (P-Value) The Number of -0.0259 (Correlation) Corporate Auditions (96) (Sample Size) 0.8006 (P-Value) Source: Own elaboration This table shows Spearman rank correlations between each pair of variables. These correlation coefficients range between -1 and +1 and measure the strength of the association between the variables. In contrast to the more common Pearson correlations, the Spearman coefficients are computed from the ranks of the data values rather than from the values themselves. Consequently, they are less sensitive to outliers than the Pearson coefficients. Also shown in parentheses is the number of pairs of data values used to compute each coefficient. The third number in each location of the table is a P-value which tests the statistical significance of the estimated correlations. P-values below 0.05 indicate statistically significant non-zero correlations at the 95.0% confidence level. None from the analyzed pairs of variables have P-values below 0.05. From the above analysis results, it is clear that it has not been proved dependence between the number of corporate insolvencies proposals and the number of corporate auditions. 24 4 Conclusions This paper had as objective: firstly, to describe the development of the number of corporate insolvencies proposals and the number of corporate auditions in the Czech Republic since the beginning of the recent financial crisis until the present. Secondly, analyze the dependence between these examinees variables. From the results of the analysis, it is clear that the development of the examined variables during the monitored period was highly variable (fluctuating). Dependence between examined variables could not be prove. For further research is recommended detailed analysis of corporate insolvencies proposals and corporate auditions by region and then also in terms of business sectors. Acknowledgments This paper was created with the support of specific research in ŠAVŠ o.p.s. References Barbulescu, M., Hagiu, A., Vasilica, R. (2015). The Insolvent Company: Bankruptcy Difficulties. Euro and the European Banking System: Evolutions and Challenges, pp. 591-601. Bokšová, J., Randaková, M., Hospodka, J., Maixner, J. (2014). Personal bankruptcies of individuals in the Czech Republic in relation to different groups of creditors. Managing and Modelling of Financial Risks: 7th International Scientific Conference, PTS I-III, pp. 80-86. Čámská, D. (2012). National View of Bankruptcy Models. 6th International Days of Statistics and Economics, pp. 268-278. Čámská, D. (2013). Basic Characteristics of Enterprises in Insolvency. Mezinárodni Vědecká Konference: Hradecké Ekonomické Dny 2013 - Ekonomicky Rozvoj A Management Regionu, Part I, pp. 83-88. Creditreform.cz (2016). Vývoj insolvencí v České republice. Hindis, R., Hronová, S., Novák, I. (2000). Metody statistické analýzy pro ekonomy. Praha: Management Press. Hospodka, 1, Buben, O., Randaková, M., Bokšová, J. (n.d.). Personal Bankruptcy in the Capital City Region and South Bohemian Region in the Czech Republic. 16th Annual Conference on Finance and Accounting, ACFA Prague 2015, 25, pp. 41-52. ISO 31000:2009 Risk Management - Principles and guidelines (with supporting standard IEC 31010:2009 - Risk Management - Risk assessment techniques) Kislingerová, E. (2013). Estimated development of the number of filings for insolvency and declared bankruptcies in the Czech Republic between 2013 and 2017. Financial Management of Firms and Financial Institutions: 9th International Scientific Conference Proceedings, PTS I-III, pp. 356-366. Kislingerová, E., Schoenfeld, J. (2014). The development of insolvency in the entrepreneurial sphere in the Czech Republic during the crisis years. Managing and Modelling of Financial Risks: 7th International Scientific Conference, PTS I-III, pp. 367-378. Maixner, 1, Hospodka, 1, Randaková, M., Bokšová, J. (2014). Personal Bankruptcy in The South Bohemian Region in The Czech Republic. International Conference on Accounting, Auditing, and Taxation (ICAAT 2014), pp. 105-111. Pittman, M., Ivry, B. (2009). U.S. Taxpayers Risk $9.7 Trillion on Bailout Programs. 25 Would the Russian Economy Turn into a Lost Decade? Miloš Bikár1, Katarína Vávrová2, Mariana Sedliačiková3 1 University of Economics in Bratislava Faculty of Business Management, Corporate Finance Department Dolnozemská cesta 1, 852 35 Bratislava, Slovak Republic E-mail: milos.bikar@gmail.com 1 University of Economics in Bratislava Faculty of Business Management, Corporate Finance Department Dolnozemská cesta 1, 852 35 Bratislava, Slovak Republic E-mail: katarina.vavrovlO@gmail.com technical University in Zvolen Faculty of Wood Sciences and Technology, Department of Business Economics Ul. T.G.Masaryka 24, 960 53 Zvolen, Slovak Republic E-mail: sedliacikova@tuzvo.sk Abstract: During the last couple of years the Russian economy was marked by a severe problems, those could turn into a quite long-lasting recession. The economic sanctions and low commodity prices caused significant damage and economic turmoil. The corporate and state-controlled investment into Russia significantly fell jointly with sharp rubble depreciation against major currencies. The paper deals with the analyses and predictions of the Russian economy development and the main factors influencing the current and future economy and the structure of GDP. The objective of this article is to analyse the time-varying nature among Russian GDP, oil prices and some major macroeconomic variables. The results of the carried out research reflected that the oil prices has played a key role and has indeed changed the nature of correlation relationship between analysed variables. The paper highlights the monetary policy including currency prediction, public debts, inflation levels, as well as commodity export dependency. Keywords: fiscal and monetary policy, commodity prices, exchange rate volatility, currency movements, public finances JEL codes: F34, G15, G18, H60 1 Introduction More than ten years ago, the Russia economy was on a good track, driven by foreign direct investment inflow, growing consumption, productivity and economic output, low public debt, raising commodity prices and stable public finances. The global financial crises starting in 2008 in USA, expanded to the rest of the world in 2009-2010, caused significant revision of potential growth in most advanced economies and has been particularly sharp in countries most exposed to commodity price developments. As stated in the World Bank (2014) report, frail domestic demand with zero investment growth dragged the Russian economy close to stagnation. The economic sanctions in Russia pulled down the potential growth further and jointly with low oil prices and persistently weak demand, resulted to a recession, which could turn in a pessimistic scenario to a lost decade. The Russia economy have drawn the attention of investors, policy makers and research academics across the world. Several studies have focused on the correlation between the commodity prices and the Russia economic output, dependency of oil prices and state budget development, the general economic environment and foreign direct investment and currency volatility and reactions to different shock scenarios. Conceptual views on the Russia economic development are similar and point out on the one hand, an inevitable structural reforms across all economy and industry sectors and on the second hand, strong correlation between input commodity prices and achieved 26 economic results (Bikár, 2013). As confirmed by the study of Gaddy and Ickes (2010), Russia's oil and gas give an unmatched source for generating wealth. The problem is, how those source are used, mainly how inefficient sector and companies as addictive are laying claims to an inordinate share of the rents, and how to restrain the addiction to the ensuing rents. The problem for Russia is how to move away from addiction within the confines of the rent management system that Putin has created. Kudrin and Gurvich (2015) found out that commodity markets substantially accelerated output growth allowed record increases in incomes (wages for all sectors of the economy, including the public sector, pensions, etc.) and improved macroeconomic stability. However, significant resources aimed at modernizing the economy failed to produce tangible results, as Russia's international competitiveness has not fundamentally improved. This casts doubt on the possibility that a resource approach can foster conditions for long-term economic growth. They came to a conclusion that even if oil prices suddenly recover rapidly, the Russia model based on imported growth would still fail to ensure economic growth. Bogetic and Olusi (2013) carried out the research study analysing the divers of firm-level productivity in Russia's manufacturing sector using Cobb-Douglas production function. They found out, that productivity grew steadily between 2003 and 2008, with an annual growth rate averaging 4% over the period, showing no signs of a slowdown from the previous period after the 1998 crisis. Gonzalez et al. (2013) applied a model searching up factors influencing the Russian volatility. They proved that Russian manufacturing output growth is characterized by a higher volatility than other comparator countries. Higher volatility is mostly driven by the presence of more numerous, deeper and longer slumps and is mostly associated with aggregate slumps with yearly effects. In addition, as shown by study of Sharafutdinova and Kisunko (2014) and Reisinger and Moraski (2013), the Russia is characterized by close informal ties between the authorities and business. In means, that the regional diversity in business climate - regional location was found to be significantly correlated with firms' perceptions of administrative burden, corruption, and state capture indicators. In a more recent study, Deevand Hodula (2016) investigated the interdependence of the sovereign default risk and banking system fragility in Russia, using credit default swaps as a proxy for default risk. The typical feature is that the biggest state-owned universal banks in emerging markets are closely managed by the government. But the fragility of those banks does not directly affect the state of public finances. However, in cases where state-owned banks directly participate in large governmental projects, banking fragility may result in the deterioration of state funds, while raising the risk of sovereign default. Mau (2016) analysed an anti-crisis measures and effect of the external shocks for the key Russian export products and claims, that there were serious structural problems that have reduced growth potential since the middle of the past decade, and have caused stagnation in the Russian economy. Another aspect of the economy is connected with the Russia's defense spending. According to study of Oxenstierna (2016), behind the rise in the defense budget was the new state armament program 2011-2020, with the ambition, that by 2020, 70% of the Armed Forces' arms were to be modern. However, the failure to modernize the economy and make it less dependent on hydrocarbons and more innovative led to weak growth after 2009, which means that rising defense spending has become more costly to the economy. The substance of the Russian economy in a near future, would still be heavily dependent on the oil prices (Kudrin, 2013). As confirmed by the study of Jacks (2013), greater volatility is a slightly more certain prospect for real commodity prices in the future. He assumes, that anything can be done to mitigate this volatility in a coordinated fashion either through market or policy mechanisms, this volatility will certainly continue to affect the growth prospects of nations, particularly those which are commodity exporters and which have relatively low levels of financial development. The paper deals with the analyses and predictions of the Russian economy development and the main factors influencing the current and future economy and the structure of GDP. The objective of this article is to analyse the time-varying nature among Russian GDP, oil prices and some major macroeconomic variables. 27 2 Methodology and Data To suggest the regression model, it was required a use of methods of summary, synthesis and analogy of the knowledge and creation of a short literature review. Second, it was done a data collection. In our model there were used yearly data and our time series range from 1996 to 2016 (21 observations). While similar studies generally use quarterly data, we decided to work on a yearly frequency due to the difficulty of capturing the precise quarterly data. It were used the time series from the QUANDL DataStream. To capture the dynamics of our model, it was used as a dependant variable GDP output, as a measure of economic activity. Next, it was used Brent crude oil prices (Europe quotes) and consumer price index to include the inflation trend (CPI including regulated prices). Further, the real effective RUB/EUR exchange rate was used to incorporate the monetary policy shock (currency depreciation) into our framework. Last, we incorporated into model the government yearly budget debt and the cumulative debt to GDP to analyse the influence on public finances. All variables are in percentage change over the same period in the previous year except an exchange rates. The selected parts of data series are plotted in a Figures 1-3. Regarding the methodology, we were used a method of multiple regression in order to explain the relationship among the independent variables to the dependent variable, according the formula (Hair et all., 2010): Y = a+ b1x1+ b2x2 + ~- + bxxx (1) where Y is the value of the Dependent variable (Y), a (Alpha) is the Constant or intercept, and bi is the slope (Beta coefficient) for xi , xi first independent variable that is explaining the variance in Y, b2 is the slope (Beta coefficient) for x2 , x2 second independent variable that is explaining the variance in Y, and so on. The computations were completed in Eviews. Table 1 reports the data, mnemonics, descriptions, sources and specifications. Table 1 Dataset Description Variable Description Source Specification BRENT Crude Oil Price: Brent Europe Quandl annualized, SA INFLATION Consumer price index Quandl percent, SA RUB_EUR Real effective exchange rate RUB/EUR Quandl annualized, SA BUDGET_DEBT Government Yearly Budget Debt to GDP Quandl percent, SA CUMM_DEBT Cumulative Debt to GDP Quandl percent, SA Note: SA=seasonally adjusted Source: Own production, 2016 3 Results and Discussion In this section, the multiple regression estimates for the Russia GDP and chosen independent variables - Brent crude oil, consumer price index, real effective exchange rate RUB/EUR, Government Yearly Budget Debt to GDP and finally Cumulative Debt to GDP.The output from the model confirmed the negative correlation among three selective variables (crude oil, exchange rate and cumulative debt), while two other variables (inflation and government yearly budget debt) show the positive contribution to the economic output. Experienced have showed that external shocks should be addressed with monetary and fiscal consolidation. Cash injection in such a situation would lead to increasing inflation and undermine, rather than stimulate, investment activity. This was confirmed by Russia as well, when the inflation jumped to a double digit level in 2015, followed by reduction trend in this fiscal year (see Fig.l). A healthy expansion of budget funding is also very 28 difficult due to the sharp declines in budget revenues, as the demand for military funding is increasing. Figure 1 Russia GDP Growth Annual Rate and Monthly Inflation Rate 18 < w G. / I lllll.l). 1 2011 v 2012 2013 201> 2 \ i V 16 14 12 10 8 6 4 2 0 Rate s 2 Russia GDP Annual Growth Rate — — Russia Montly Inflation Source: Quandl, 2016. Russia tend to have a surplus of government budget up to 2009, being able to generate extra sovereign fund, driven mainly by high export commodity prices. But since the global financial crises, the Russia GDP annual growth significantly suffer, and after rapid decline of crude oil prices (since end of 2014 reaching a bottom in February 2016) turned to a negative numbers. The generally weak global demand supported by lower China GDP growth cause low levels of commodity prices, what was immediately reflected in a gap of export revenues and deficit public finances (see Fig. 2). Figure 2 Russia Government Budget as a % of the GDP 10,00 8,00 - 6,00 G. II 2,00 | ° > % 0,00 o 5 .2 2 -2,00 Ifi o 2j G, -4,00 6,00 -8,00 I 2004 2006 2008 0 2012 I 2014 ™ 2^5 Source: Quandl, 2016 29 The government debt to GDP in Russia averaged 26.1% from 1999 until 2014, reaching an all time high of 99% in 1999 and a record low of 7.9% in 2008. The trend changed rapidly in a last couple of years, as the Russia recorded a cumulative government debt to GDP of 17.9% in 2014, respectively 25.3% in 2015 and forecasted 35% in 2016 (The Central Bank of the Russian Federation, 2016). The rubble fell rapidly following the imposition of external sanctions and the decline of oil prices (see Fig.3), and monetary authorities opted to hold the reserves, rather than spending them to maintain the national currency. The dynamics of the real exchange rate were largely shaped by changes in the terms of trade for the Russian economy. Figure 3 Correlation between Crude Oil Prices Brent and Exchange Rate RUB/EUR ■Z 165 100 Crude Oil Prices: Brent Europe--RUB/EUR Source: Quandl, 2016 The next table below summarises the model output including the regression equation. Table 2 Model Output Variable Coefficient Std. t-Statistic Prob. Error C 2226.29 2340.27 -0.9881 0.3490 Brent -0.1079 0.0904 -1.1907 0.2535 Inflation 0.0248 0.0623 0.3979 0.6966 Rub_EUR -0.4591 0.3125 -1.4690 0.1639 Budget Debt 0.4767 0.2662 1.7907 0.0949 Cummulat Debt -0.0412 0.0580 -0.7101 0.4892 R-squared 0.5969 Mean dependent var. 2.9704 Adjusted R-squa red 0.4242 S.D.dependent var. 4.9607 S.E. of regression 3.7641 Akaike info criterion 5.7501 Sum squared resid 198.3645 Schwarz criterion 6.0983 Log likelihood -53.3763 Hannan-Quinn crit. 5.8256 F-statistic 3.4561 Durbin- Watson stat 2.1890 Source: Own calculation (Eviews), 2016 30 Regression equation: GDP = -2267.29269162 - 0.107649652183*BRENT + 0.0248209780198*INFLATION -0.459163220786*RUB_EUR + 0.476754797483*BUDGET_DEBT 0.0412225505577*CUMMULATIVE_DEBT Further we analyse covariance coefficient and correlations among selective variables. There are positive relations among crude oil prices, inflation and an exchange rate movement, while negative correlations are among budget debt and cumulative debt against all other variables (Table 3). The result confirm the strong dependence of the Russia economy on oil prices. The lower the oil prices, the higher budged debt and cumulative debt as well. Table 3 Coefficient Covariance Matrix BRENT INFLATION RUB EUR BUDGET DEBT CUMM DEBT BRENT 0.0081 0.0006 0.0254 -0.0108 -0.0026 INFLATION 0.0006 0.0038 0.0024 -0.0024 -0.0012 RUB EUR 0.0254 0.0021 0.9760 -0.0403 -0.0117 BUDGET_DEBT - -0.0024 -0.0403 0.0708 0.0107 0.0108 CUMM_DEBT - -0.0012 -0.0117 0.0107 0.0033 0.0026 Source: Own calculation (Eviews), 2016 The results confirmed the strong dependence of the Russia economy on oil prices. The lower the oil prices, the higher budged debt and cumulative debt as well (Tuzova and Quayum, 2016). Second, the Russian government debt is less sensitive to FX shocks, and in spite of the Russia cut off from the international markets, the balance of payment gap was primary finance through sufficient FX reserves. The Central Bank of Russia is still able to help the Russian businesses with liquidity in order to roll over their debt. The Russian economy need to solve the structural problem via the diversification from the heavy reliance on the export of oil and hydrocarbons, improve the business climate for small and medium-sized businesses and suppress the corruption (Gurvich, 2013). The modernization program launched in 2009 addressed mentioned problems, but it was too much of a challenge for the ruling political elites and never materialized. It is evident that firms' geographical location matters and that regional variation in business environment might potentially be an important factor for explaining differences in firms' entry, growth and productivity and, consequently, regional economic development and prosperity (Frijters, Nemanja, 2016). Regarding the prediction of the Russia economy the GDP in 2016 should amount -1.8% with a return to a positive growth of 0.8% in 2017 (The Central Bank of the Russian Federation, 2016). Weak domestic demand will be the main factor behind the inflation decline in 2016-2017. Slower consumer price growth will also be based on cuts in producer costs, moderate global food price dynamics and tentative decline in inflation expectations. The Bank of Russia's monetary policy may remain moderately tight over a longer period of time than expected. Besides, the key rate level will be determined given the influence of decreasing structural liquidity deficit and possible transition to a structural liquidity surplus as a result of massive Reserve fund expenditures in order to cover the budget deficit. 4 Conclusions This paper investigated the link between the Russia GDP and selected independent variables (Brent crude oil, consumer price index, real effective exchange rate RUB/EUR, government yearly budget debt to GDP and cumulative debt to GDP). We relied on the multiple regression method to establish whether the correlation among dependent and independent variables would capture their mutual relation. Our main findings can be 31 summarized as follows. The selected variables imbedded into model describe the dependent variable for almost 60%. Generally this number should be higher, but in a case of Russia, the other, mainly non-measurable qualitative factors amounts for remaining 40% of the GDP development. The output from the model confirmed, that the key role in a direction of Russia GDP play oil and commodity prices, their recovery would immediately help to stabilize the state budget and export revenues. The other aspect of GDP growth (not captured by our model) is connected with the specific feature of the Russia economy. Despite the change of system from a command economy to a market economy, the institutions that normally support market allocation are weak, and in many ways they are overruled by the informal institutions surviving from the Soviet era. That Russia's institutions are deficient is reflected in the Worldwide Governance Indicators (WGI, 2014). The WGI project constructs aggregate indicators of six broad dimensions of governance: political stability and absence of violence/terrorism; voice and accountability; government effectiveness; regulatory quality; rule of law; and control of corruption. When these indicators are studied over time, it is found that in Russia they have generally been low, that they improved up to the early 2000s, but that since 2004 there has been a marked deterioration in vital institutions (Oxenstierna, 2016). Weak institutions create scope for manual management of economic matters, which is also a reason why institutions need to be kept weak - so that political goals rather than economic goals can be pursued. Business activity indicators in the global economy remained mixed in the absence of clear points of growth. As before, the greatest concerns were evoked by the growth prospects of the Chinese economy and other EMEs. The economic activity indicators for several large developed countries in particular in the euro area, which is Russia's key trading partner, were more stable and remained relatively high in comparison with recent years, what gives a chance for soon recovery of the Russian economy and return to stabilize economic environment. Acknowledgments The presented working paper is the output of the scientific grants VEGA n. 1/0007/16 The impact of the global economic developments and trends in the direction of the euro area economy on financial management of business entities in Slovakia. References Bikar, M. (2013). The trends of the development of the Russian economy. International journal of Science, Commerce and Humanities, Vol. 1, No. 7, pp. 102-111. Bogetic, Z., Olusi, O. (2013). Drivers of firm-level productivity in Russia's manufacturing sector. World Bank Policy Research Working Paper 6572, pp. 29. Deed, O., Hodula, M. (2016). Sovereign default risk and state-owned bank fragility in emerging markets: evidence from China and Russia, Post-Communist Economies, Vol. 28 (2), pp. 232-248. Frijters, P., Nemanja, A. (2016). Can collapsing business network explain economic downturns? Economic modelling, Vol. 54, pp. 289-308. Gaddy, C, G., W. Ickes, B., W. (2010). Russia after the Global Financial Crisis. Eurasian Geography and Economics, Vol. 51, No. 3, pp. 281-311. Gonzalez, A., Iacovone, L, Subhash, H. (2013). Russian volatility: Obstacle to firm survival and diversification. World Bank Policy Research Working Paper 6605, pp. 36. Gurvich, E. (2013). flo/irocpoHHbie nepcneKTi/iBbi poccmmckom skohommkm. 3kohomi/inecKafl no/iMTMKa N? 3, C. 7-32. Long-term development prospects for the Russian economy. Ekonomicheskaya Politika No. 3, pp. 7-32. Hair, IF., Black, W.C, Babin, B.J., Anderson, R.E. (2010). Multivariate Data Analysis, 7th Edition. New Jersey: Pearson Education, USA. ISBN: 9780138132637. 32 Jacks, D. (2013). From Boom to Bust: A Typology of Real Commodity Prices in the Long Run. NBER Working Paper No. 18874. pp. 62. Retrieved from: http://www.sfu.ca/~djacks/pa pers/workingpapers/wl8874%20(typology).pdf. Kudrin, A. (2013). B/ii/iflHi/ie aoxoaob ot 3KcnopTa Hect>Tera30Bbix pecypcoB Ha fleHexHO-KpeflMTHyHD no/iMTMKy Poccmm. Bonpocbi 3kohommkm. N? 3. C. 4-19. The Influence of Oil and Gas Exports on Russia's Monetary Policy. Voprosy Ekonomiki No. 3, pp. 4-19. Kudrin, A., Gurvich, E. (2015). A new growth model for the Russian economy. BOFIT Policy Brief, No. 1, Bank of Finland, Institute for Economies in Transition, pp. 1-35. Retrieved from: http://www.eeg.ru/files/lib/Kudrin-Gurvich%20(BOFIT-2015).pdf. Mau, V. (2016). Anti-crisis measures or structural reforms: Russian economic policy in 2015. The Russian Journal of Economics, Vol. 2, pp. 1-22. Oxenstierna, S. (2016). Russia's defense spending and the economic decline. Journal of Eurasian Studies, Vol. 7, Issue 1, pp. 60-70. Reisinger, W., Moraski, B. (2013). Deference or Governance? A Survival Analysis of Russia's Governors under Presidential Control. In: Russia's Regions and Comparative Subnational Politics, W. Reisinger (ed.), London: Routledge, pp. 40-62. Sharafutdinova, G., Kisunko, G. (2014). Governors and governing institutions: A comparative study of state-business relations in Russia's regions. World Bank Policy Research Working Paper 7038, pp. 46. The Central Bank of the Russian Federation (2016). Monetary policy report. Moscow, March 2016, pp-58. Retrieved from: http://www.cbr.ru/Eng/publ/ddcp/2016_01_ddcp_e.pdf. Tuzova, Y., Quayum, F. (2016). Global oil glut and sanctions: The impact on Putin's Russia. Energy Policy, Vol. 90, pp. 140-151. World Bank (2014). Confidence Crisis Exposes Economic Weakness. Russia Economic Report No. 31. Washington D.C., pp. 60. World Bank (2014). Worldwide Governance Indicators. Country Data Report for Russian Federation, 1996-2014. Retrieved from: http://info.worldbank.Org/governance/wgi/index.aspx#countryReports. Quandl (2016). Russia GDP Growth Annual Rate and Monthly Inflation Rate. Russia Government Budget as a % of the GDP. Crude Oil Prices Brent and Exchange rate RUB/EUR. Retrieved from: https://www.quandl.com/data/OECD/NAAG_RUS_GDPG-Russia-Gross-Domestic-Product-Gdp-Volume-Annual-Growth-Rates-Percentage, https://www.quandl.com/data/RATEINF/INFLATION_RUS-Inflation-YOY-Russia, https://www.quandl.com/data/FRED/DCOILBRENTEU-Crude-Oil-Prices-Brent-Europe, https://www.quandl.com/data/ECB/EURRUB-EUR-vs-RUB-Foreign-Exchange-Reference-Rate. 33 Oil Price and European Stock Markets Gabor Bota1, Mihaly Ormos2 1 Budapest University of Technology and Economics, Department of Finance Magyartudosok krt. 2., 1117 Budapest, Hungary E-mail: bota@finance.bme.hu 2 Budapest University of Technology and Economics, Department of Finance Magyar tudosok krt. 2., 1117 Budapest, Hungary E-mail: ormos@finance.bme.hu Abstract: In this paper we investigate the effects of oil price changes on European stock returns. We run regressions for extended versions of different market equilibrium models: (standard CAPM, Fama-French 3 factor model, Carhart four-factor model) incorporating oil price changes. We also separate different market situations based on the oil price. Our results suggest that from among the investigated European industry indexes oil and gas companies have higher exposure to oil price changes than other companies from other industries. When examining the broad oil and gas industry on a sub-sector level we can detect some significant differences. Keywords: asset pricing, oil price J EL codes: G12, G15 1 Introduction The economic effects of oil price changes have been investigated extensively in the past decades. One direction of research (e.g.: Hamilton 1983, Mork 1989, Jones et al. 2004) concentrates on the macroeconomic consequences of oil price shocks. Other papers (e.g.: Chen, Roll and Ross 1986, Basher and Sadorsky 2006, Nandha and Hammoudeh 2007, Fang and You 2014) focus the relationship between oil prices and stock market returns and the results are highly dependent on the countries, regions, industries and even periods examined. Aloui et al. (2013) show positive (however varying) dependence between oil price changes and returns of Central and Eastern European stock market indexes. Asteriou and Bashmakova (2013) find that the reaction of CEE stock returns to oil price changes is more significant when oil prices are low. Articles examining stock returns on sectoral level find that oil and gas industry of Australia (Faff and Brailsford 1999), Canada (Boyerand Filion 2007), Europe (Arouri and Nguyen 2010) and the UK (El Sharif et al 2005) has all significant sensitivity to oil price changes. Nandha and Faff (2008) detect a negative impact of oil price increases on returns for all of the examined 35 global sectors except mining, and oil and gas industries. Ramos and Veiga (2011) show that oil price has a positive impact on global oil and gas industry returns, however oil price is a more important factor in developed countries than in emerging markets. Nandha and Brooks (2009) also document substantial differences in the role of oil price changes in determining transport sector returns between developed and emerging countries. Oberndorfer (2009) shows that oil price changes positively related to returns of oil and gas stock returns in the Eurozone. Mohanty et al. (2010) find no significant relation between oil prices and returns of CEE oil and gas companies. Narayan and Sharma (2011) not only find positive relation between oil price changes and returns of US energy and transportation companies but they report adverse effects for stock returns of companies representing other sectors. We examine the effects of oil price changes on returns of European sectoral stock indexes. As the examined period ends in April 2016 we can detect the effects of latest developments in oil prices, especially the decline from mid-2014 to early-2016. By distinguishing the bullish and bearish oil market sub-periods our results help to understand not just the general effects of oil price changes on the return generating process of various European industries (and sub-sectors of the broad oil and gas industry), but to detect the differences in distinct oil market conditions. 34 2 Methodology and Data We apply different equilibrium models to capture the excess returns of the examined indexes and to calculate the explanatory power of the different models. We run ordinary least squares regressions with different set of explanatory variables. The first equilibrium model we use is the standard Capital Asset Pricing Model (CAPM) proposed by Sharpe (1964), Treynor (1961), Lintner (1965), and Mossin (1966), and is in the following form, where r, represents the return of the index; a represents the constant term of the regression, i.e., the abnormal return; represents /3 a relevant risk parameter that is estimated as the independent variable of the regression; rM represents the market return; and e represents the error term of the regression: rt = a+prM+e (1) The second equilibrium model is the Fama and French (1992, 1993, 1996) three-factor model. The authors extend the explanatory variable using the SMB (small minus big) and HML (high minus low) factors respectively, to capture the size premium and the value over growth premium. The model is written as follows, where the /3 variables represent the regression coefficients and rM, SMB and HML are the market, size, and value premiums, respectively: rt = a+ (3MrM + (3SMBSMB + /3HMLHML + e (2) Carhart (1997) extends the three-factor model using a momentum (MOM) parameter that measures the tendency for the share price to continue increasing if it was previously increasing and its tendency to continue decreasing if it was previously decreasing. Therefore, the model can be written in the following form, where /3MOm captures the excess return gained by the persistency of the previous month's return and MOM stands for the momentum factor: rt = a+ (lMrM + pSMBSMB + (1HMLHML + (lMOMMOM + e (3) Table 1 Descriptive Statistics of the Indexes and Oil and Market Proxies Name Code Obs. Avg. return SD Skew. Kurt. JB-prob. CRSP Europe Mkt-rf 310 0.005 0.050 -0.598 1.591 0.000 Oil price Oil 310 0.003 0.106 -0.258 1.414 0.000 Oil and Gas (103) OILGS 310 0.007 0.062 -0.434 1.148 0.000 Oil and Gas Producers (62) OILGP 310 0.007 0.062 -0.353 0.896 0.000 Exploration and Production (27) 01 LEP 310 0.009 0.081 -0.471 3.041 0.003 Integrated Oil and Gas (35) 01 LIN 310 0.007 0.061 -0.329 0.802 0.000 Oil Equip. Services and Distrib. (30) 01 LES 310 0.002 0.093 -0.891 3.050 0.000 Oil Equipment and Services (25) OILSV 310 0.002 0.093 -0.880 2.976 0.000 Pipelines (5) PI PEL 254 0.006 0.085 -1.177 4.611 0.000 Alternative Energy (11) ALTEN 216 0.013 0.123 -0.727 2.131 0.000 Renewable Energy Equipment (9) RENEE 216 0.012 0.125 -0.730 2.110 0.000 Alternative Fuels (2) ALTFL 115 -0.005 0.107 0.234 0.282 0.000 Basic Materials (154) BMATR 310 0.006 0.070 -1.028 4.591 0.000 Industrials (566) INDUS 310 0.006 0.063 -0.870 2.292 0.000 Consumer Goods (330) CNSMG 310 0.007 0.058 -0.752 1.525 0.000 Health Care (149) HLTHC 310 0.008 0.041 -0.454 1.326 0.000 Consumer Services (318) CNSMS 310 0.006 0.054 -0.617 1.574 0.000 Telecommunications (54) TELCM 310 0.006 0.063 -0.397 1.471 0.000 Utilities (87) UTILS 310 0.007 0.053 -0.586 1.410 0.000 Financials (556) FINAN 310 0.005 0.067 -0.846 4.240 0.000 Technology (142) TECNO 310 0.007 0.086 -0.344 1.605 0.000 Source: Based on data from Thomson Reuters 35 We use monthly total return data (in USD) of European Datastream equity indexes of different industries for the period July 1990 - April 2016 provided by Thomson Reuters. There are ten top-level sector Europe-Datastream indexes taken into consideration as well nine sub-sector indexes of the oil and gas industry. The descriptive statistics of the monthly returns of the market proxy, oil price and the indexes are summarized in Table 1. (The number of constituents of the examined indexes are in brackets). The oil price is represented by the Brent USD per barrel price (also from Thomson Reuters). The market (CRSP Europe Value Weighted Return Premium), size, value and momentum factors are the European factors from Kenneth R. French's data library. In order to separate different oil market conditions we use an oil price dummy variable which equals 1 when the oil price return of the given month is above the mean return of the whole investigated period and equals 0 when it is below the mean. 3 Results We run the equilibrium linear regression models for the monthly returns of the sample period of 1990-2015. The main focus of our investigation is the oil and gas sector; thus we collect the different models in two distinct array: Table 2 presents the parameter estimations for the Oil and Gas sector with its constituent subsectors and Table 3 presents the results of the broad industry sectors. For this full period the standard CAPM beta, thus the sensitivity for the market proxy is significant for all industrial segments and for the oil and gas sub-segments as well. The average determination coefficient is 0.547. The two smallest indexes: Pipelines and Alternative fuel sub-industries represent the lowest R2s with 0.135 and 0.245 with five and two constituents respectively. We find the two broadest industry sectors with R2s above 0.8; the Industrials and Financials with 0.844 and 0.813 respectively. The average R2s of the Oil and Gas sector is 0.406 that is significantly lower than for the remaining broad indexes which represents an average 0.684 determination coefficient in the standard CAPM frame work. The average parameter estimations for the oil and gas sector and for the other broad indexes are not significantly different with 0.98 and 1.03 respectively. The oil and gas industry subsectors bearing the highest risk are Alternative Energy and Renewable Energy Equipment with a beta value of 1.51 and 1.50 respectively. This result suggests that the most advanced and sustainable energy sectors are still are handled with caution by the investors and they require higher risk premium to invest into these companies. On the other hand Pipelines bearing the lowest risk with 0.62 beta that suggest investors expects pipeline industry is not heavily exposed to exogenous shocks affect the capital markets. The standard oil and gas production (OILGP) exhibit 0.936 of beta; thus, it follows the market evenly, while Oil Equipment, Services and Distribution has higher risk of 1.28. Adding the oil returns to the standard CAPM generates 10.1% higher R2s for the Oil and gas index; however, it gives only a slight increase (0.3% on average) in the case of the broader indexes. This result shows that the oil and gas companies have much higher exposure to oil prices than the other industries. The oil beta is significant at 1% in the case of all the oil and gas sub-industry indexes but the pipelines only at 5% and it is an interesting fact that the alternative fuel industry exhibit no significant sensitivity to oil prices. The other broader industries exhibit low sensitivity to oil prices as only the Basic materials index has a 1%, Healthcare has a 5% and Telecommunication a 10% significant sensitivity and to oil and all remaining industries are not significant on any usual levels. The market betas are still significant in all cases, and they are decreased by 0.115 in all oil industry segment. In the Fama and French (1996) equilibrium framework we find only a slight increase in the R2s compared to the standard CAPM and these are remarkably lower than that of the CAPM+oil model. However, one can find significant SMB and HML parameters in many cases. It is an interesting fact that on the one hand the oil and gas sector as a whole exhibits significant HML and non-significant SMB parameter. On the other hand most of the oil and gas subsectors are significantly sensitive to the SMB factor suggesting positive size premium. HML is positive and significant in all cases but for Alternative Energy (Renewable Energy Equipment and 36 Alternative Fuels). Thus; on the one hand most of the oil companies behaves like small firms but exhibit high book value compared to their market value. We measure mixed results in the cases of the remaining indexes; both the sign of the parameter estimation and the significance level is changing from sector to sector. Table 2 Equilibrium Models and Their Oil Extended Version for the Oil and Gas Industry and Its Sub-sector Indexes. Alpha Beta M Beta SMB Beta HML Beta MOM Beta Oil R2 -0.0289 0.9633*** 0.5752 0.0130 0.8484*** 0.2014*** 0.6765 #1TI f— C -0.1103 0.9426*** 0.1003 0.2840*** 0.5849 UlLus -0.0600 0.8045*** -0.1410 0.2834*** 0.2102*** 0.6887 -0.0871 0.9383*** 0.1013 0.2763** -0.0200 0.5837 -0.0253 0.7980*** -0.1398 0.2719** -0.0299 0.2104*** 0.6880 -0.0105 0.9361*** 0.5542 0.0309 0.8224*** 0 1992*** 0.6553 -0.0911 0.9136*** 0.0768 0.2833*** 0.5635 OILGP -0.0410 0.7762*** -0.1634* 0.2827*** 0.2093*** 0.6685 -0.0685 0.9094*** 0.0778 0.2758** -0.0196 0.5622 -0.0069 0.7698*** -0.1622* 0.2714** -0.0294 0.2095*** 0.6677 0.1391 1.0399*** 0.3947 0.1965 0.8825*** 0.2758*** 0.5065 0.0141 1.0625*** 0.7967*** 0.3750** 0.4490 OILEP 0.0732 0.9004*** 0.5134*** 0.3743*** 0.2468*** 0.5328 0.0838 1.0496*** 0.7996*** 0.3518** -0.0602 0.4479 0.1565 0.8847*** 0.5162*** 0.3466** -0.0719 0.2473*** 0.5323 -0.0378 0.9236*** 0.5448 0.0029 0.8120*** 0.1955*** 0.6432 ATI TM -0.1164 0.8976*** 0.0268 0.2812*** 0.5535 U1L1IN -0.0665 0.7607*** -0.2125** 0.2806*** 0.2085*** 0.6587 -0.0995 0.8945*** 0.0275 0.2755** -0.0146 0.5521 -0.0381 0.7554*** -0.2115** 0.2711** -0.0245 0.2086*** 0.6577 -0.6385 1.2849*** 0.4660 -0.5849 1.1380*** 0.2574*** 0.5408 r\Ti cc -0.7546** 1.3078*** 0.7638*** 0.3460** 0.5032 U1LE9 -0.6997** 1.1574*** 0.5007*** 0.3454** 0 2291*** 0.5586 -0.5659 1.2731*** 0.7717*** 0.2833* -0.1629 0.5059 -0.4982 1.1195*** 0.5077*** 0.2784* -0.1738* 0.2304*** 0.5621 -0.6596* 1 2924*** 0.4624 -0.6046* 1.1415*** 0.2643*** 0.5397 -0.7794** 1.3136*** 0.7594*** 0.3598** 0.4990 OILSV -0.7228** 1.1581*** 0.4876*** 0.3592** 0.2368*** 0.5570 -0.5913 1.2790*** 0.7673*** 0.2973* -0.1624 0.5015 -0.5214 1.1202*** 0.4945*** 0.2923* -0.1737* 0.2381*** 0.5604 0.0800 0.6161*** 0.1349 0.0736 0.5531*** 0.1046** 0.1473 -0.1305 0.5980*** 0.5420** 0.5969*** 0.1762 PIPEL -0.1286 0.5399*** 0.4465** 0.5993*** 0.0838* 0.1824 -0.0348 0.5789*** 0.5469** 0.5673** -0.0779 0.1742 -0.0339 0.5212*** 0.4515** 0.5701** -0.0770 0.0836* 0.1804 0.5400 1.4968*** 0.4310 0.4906 1.3936*** 0.1657*** 0.4486 A1 TT IVI 0.3962 1.5371*** 0.7674*** 0.0282 0.4459 AL 1 CIN 0.3866 1.4467*** 0.5983** 0.0314 0.1294** 0.4547 0.4701 1.5188*** 0.7787*** 0.0021 -0.0676 0.4438 0.4537 1.4304*** 0.6091** 0.0077 -0.0614 0.1289** 0.4525 0.5252 1.5069*** 0.4221 0.4757 1.4037*** 0.1657*** 0.4390 DC MCE 0.3733 1.5464*** 0.7844*** 0.0456 0.4371 KtlNtt 0.3637 1.4567*** 0.6166** 0.0487 0.1284** 0.4453 0.4460 1.5285*** 0.7955*** 0.0199 -0.0665 0.4349 0.4297 1.4407*** 0.6272** 0.0254 -0.0603 0.1280** 0.4431 -0.8494 0.8953*** 0.2447 -0.8046 0.8163*** 0.0899 0.2437 -0.8392 0.8559*** 0.5161 0.2160 0.2416 ALTFL -0.7858 0.7862*** 0.4328 0.2466 0.0699 0.2379 -1.0198 0.9028*** 0.5597 0.4057 0.2702 0.2421 -0.9664 0.8331*** 0.4766 0.4361 0.2700 0.0698 0.2383 Source: Based on data from Thomson Reuters 37 Table 3 Equilibrium Models and Their Oil Extended Versions for the Broader Industry indexes. Alpha Beta M Beta SMB Beta HML Beta MOM Beta Oil R2 -0.0289 0.9633*** 0.5752 0.0130 0.8484*** 0.2014*** 0.6765 OILGS -0.1103 0.9426*** 0.1003 0.2840*** 0.5849 -0.0600 0.8045*** -0.1410 0.2834*** 0.2102*** 0.6887 -0.0871 0.9383*** 0.1013 0.2763*** -0.0200 0.5837 -0.0253 0.7980*** -0.1398 0.2719*** -0.0299 0.2104*** 0.6880 -0.2057 1.2180*** 0.7399 -0.1870 1.1668*** 0.0898*** 0.7555 BMATR -0.2902 1.2020*** 0.1692 * 0.2886*** 0.7504 -0.2696 1.1453*** 0.0701 0.2884*** 0.0864*** 0.7638 -0.2480 1 1942*** 0.1710 * 0.2746*** -0.0364 0.7499 -0.2226 1.1365*** 0.0717 0.2728*** -0.0405 0.0866*** 0.7635 -0.1398 1.1638*** 0.8443 -0.1366 1.1550*** 0.0154 0.8444 INDUS -0.1270 1 1779*** 0.1116 * -0.0571 0.8455 -0.1247 1.1717*** 0.1008 -0.0571 0.0094 0.8451 -0.0249 1.1591*** 0.1159 * -0.0910 -0.0881** 0.8477 -0.0219 1.1524*** 0.1043 -0.0912 -0.0886** 0.0101 0.8475 0.0429 0.9592*** 0.6933 0.0429 0.9592*** 0.0000 0.6923 CNSMG 0.0285 0.9455*** -0.1004 0.0613 0.6936 0.0300 0.9413*** -0.1077 0.0613 0.0064 0.6927 0.1355 0.9258*** -0.0960 0.0257 -0.0924* 0.6962 0.1376 0.9211*** -0.1040 0.0256 -0.0927* 0.0070 0.6954 0.3270** 0.5948*** 0.5163 0.3186** 0.6177*** -0.0402** 0.5244 HLTHC 0.3123** 0.5624*** -0.3196 *** 0.0834 0.5470 0.3068* 0.5776*** -0.2932 *** 0.0834 -0.0231 0.5485 0.1596 0.5906*** -0.3260 *** 0.1342** 0.1319*** 0.5602 0.1525 0.6066*** -0.2985 *** 0.1347** 0.1330*** -0.0240 0.5620 -0.0972 0.9851*** 0.8147 -0.0964 0.9830*** 0.0038 0.8142 CNSMS -0.0266 1.0195*** 0.1060 * -0.2648*** 0.8295 -0.0273 1.0216*** 0.1096 * -0.2648*** -0.0031 0.8289 0.0102 1.0127*** 0.1075 * -0.2771*** -0.0317 0.8294 0.0093 1.0147*** 0.1109 * -0.2770*** -0.0316 -0.0029 0.8288 -0.0172 0 9229*** 0.5183 -0.0261 0 9474*** -0.0429* 0.5212 TELCM 0.1901 0.9582*** -0.4628 *** -0.7039*** 0.6086 0.1856 0.9706*** -0.4412 *** -0.7039*** -0.0188 0.6081 0.1171 0.9717*** -0.4658 *** -0.6796*** 0.0631 0.6087 0.1114 0.9845*** -0.4437 *** -0.6792*** 0.0640 -0.0193 0.6083 0.0404 0.8542*** 0.6379 0.0404 0.8544*** -0.0004 0.6367 UTILS -0.0507 0.8207*** -0.0080 0.3296*** 0.6575 -0.0505 0.8202*** -0.0090 0.3296*** 0.0008 0.6564 -0.1630 0.8414*** -0.0127 0.3670*** 0.0970** 0.6612 -0.1630 0.8413*** -0.0129 0.3670*** 0.0970** 0.0001 0.6600 -0.3540** 1.2226*** 0.8131 -0.3594** 1.2375*** -0.0261 0.8140 FINAN -0.4852** 1.1550*** -0.2417 *** 0.4963*** 0.8505 -0.4880** 1.1627*** -0.2282 *** 0.4963*** -0.0118 0.8503 -0.2960* 1.1202*** -0.2338 *** 0.4333*** -0.1634*** 0.8584 -0.2991* 1 1272*** -0.2216 *** 0.4336*** -0.1629*** -0.0106 0.8582 -0.2030 1.3353*** 0.5804 -0.2077 1.3483*** -0.0228 0.5797 TECNO 0.0883 1.4107*** -0.3458 *** -1.0180*** 0.6618 0.0870 1 4142*** -0.3396 ** -1.0180*** -0.0054 0.6608 0.3410 1.3641*** -0.3353 *** -1.1020*** -0.2182*** 0.6697 0.3398 1.3667*** -0.3309 ** -1.1019*** -0.2180*** -0.0038 0.6686 Source: Based on data from Thomson Reuters In the next step we add the oil factor to the Fama and French three-factor model. Again we find only a slight increase compared to the CAPM+oil setup and a large increase compared to the standard three-factor model. The increase was moderate in the case of 38 Pipelines and Alternative Energy sector and was considerably high in the case of the broader Oil and gas index with the Oil Equipment and Services, Oil and Gas Producers and Exploration and Production. The alternative fuel industry still exhibits no significant sensitivity to oil prices. Concerning the broader indexes we find that the oil significance disappears in the case of Healthcare and it plays a significant role only on Basic materials. We also add the momentum factor to the standard and oil extended Fa ma and French three-factor model to get the Carhart (1998) model and its extended version. The momentum factors are not significant in any of the oil and gas sector constituent indexes and thus adds no increase to the explanatory power of the model in the standard Carhart framework. By adding the oil to the Carhart model the Oil Equipment, Services and Distribution sector shows 0.174 significant MOM beta factor at 10%, and we find no more change. The broader other industry indexes exhibit mixed sensitivity to the momentum factor: the momentum factor cannot explain the Consumer Services, Telecommunications and Utilities indexes, it has positive and significant effect on Healthcare and Utilities, while significant negative effect on Industrials, Consumer Goods Financials and Technology. Altogether we find that the momentum factor does not add any information to the previous models. Table 4 Equilibrium Models Extended with Oil Prices and Oil Trend Dummy for the Oil _and Gas Industry and Its Sub-sector Indexes._ Oil Beta Beta Beta Beta Oil D x Oil D Oil D x Oil D x _alpha dummy Beta M SMB HML mom oil Oil DxM SMB x HML MOM Oil_R2 -0.352-0.007 0.881_0.141 -0.065_0.194*** 0.684 OILGS -0.466-0.007 0.861 0.017 0.234_0.136 -0.140 -0.358* 0.112_0.223"'* 0.699 -0.430-0.007 0.851 0.020 0.223-0.052 0.136 -0.126 -0.361* 0.131 0.067 0.225"'" 0.698 -0.383-0.007 0.845_0.135 -0.043_0.198*" 0.663 OILGP -0.492-0.006 0.822-0.019 0.237_0.133 -0.115 -0.324* 0.104_0.225*** 0.678 -0.448-0.007 0.810-0.016 0.224-0.064 0.132 -0.096 -0.327* 0.130 0.087 0.228*** 0.677 0.927 -0.016 0.947_0.318 -0.128_0.043 0.507 OILEP 0.674 -0.014 0.956 0.605 0.304_0.275 -0.125 -0.234 0.132_0.060 0.530 _0.896-0.019 0.893 0.620 0.239-0.320 0.271 -0.034 -0.252 0.256 0.426** 0.075 0.537 -0.476-0.007 0.832_0.125 -0.036_0.211*** 0.652 OILIN -0.579-0.006 0.807-0.057 0.239_0.125 -0.115 -0.346* 0.096_0.240*** 0.670 -0.553-0.007 0.799-0.055 0.232-0.037 0.125 -0.104 -0.348* 0.113 0.054 0.242*** 0.668 1.027-0.018 1.141_0.405 0.008_-0.182 0.545 OILES 0.804-0.016 1.164 0.698 0.206_0.357 -0.024 -0.428 0.253_-0.154 0.563 _1.073-0.018 1.087 0.716 0.127-0.388 0.353 0.046 -0.445 0.317 0.360* -0.152 0.569 1.009-0.017 1.141_0.413 0.015_-0.187* 0.544 OILSV Q.774-0.016 1.161 0.681 0.236_0.365 -0.014 -0.418 0.221_-0.159 0.561 _1.047-0.018 1.083 0.699 0.156-0.393 0.361 0.057 -0.435 0.286 0.366* -0.157 0.567 0.368-0.011 0.669_0.099 -0.232_0.106 0.145 PIPEL Q.242-0.011 0.654 0.106 0.213_0.090 -0.197 0.687 0.793**_0.058 0.191 _0.210-0.008 0.664 0.102 0.223 0.047 0.092 -0.237 0.693 0.731* -0.143 0.045 0.186 0.656-0.010 1.655_0.119 -0.533**_0.181 0.455 ALTEN Q.352-0.008 1.628 0.417 0.386_0.090 -0.370 0.184 _0.175 0.461 _0.435-0.008 1.590 0.446 0.357-0.151 0.086 -0.352 0.162 - 0.090 0.172 0.456 0.666-0.010 1.674_0.121 -0.551**_0.177 0.446 RENEE Q.356-0.008 1.646 0.430 0.390_0.091 -0.388 0.190 _0.170 0.450 _0.438-0.008 1.609 0.458 0.362-0.149 0.087 -0.370 0.168 - 0.089 0.168 0.446 -2.175 0.003 0.923_-0.078 -0.215_0.392 0.237 ALTFL -2.139 0.003 0.783 0.362 0.560 -0.085 0.043 0.103 _0.340 0.220 -2.695 0.008 0.850 0.355 1.138 0.936-0.051 -0.006 0.140 - -0.884 0.305 0.227 Source: Based on data from Thomson Reuters In order to separate different market conditions based on the monthly oil price change we also run the equilibrium models extended with oil price trend dummy. If we compare the determination coefficients of the models using the oil price dummy with the previous ones there is only a slight increase for most of the oil and gas sub-sectors and model settings (and even a minimum decrease in some cases), so the separation of the oil 39 market conditions does not result in models with higher explaining power. In order to detect the differences of the distinct oil market conditions we run the regressions using the product of the oil price dummy and the other factors. For market, SMB, HML and MOM factors there are very few cases when the product of the given factor and the oil price dummy is significant, thus these factors do not behave differently when oil prices increasing or decreasing. The product of oil price dummy and oil price factor is significant (at 1%) for the broad oil and gas industry index, for oil and gas producers and for integrated oil and gas companies for all (CAPM+oil, 3 factor+oil and 4 factor+oil) model settings so the oil price changes have a significant effect on the return generating process of these companies. For other sub-sectors of the oil and gas industry no significant effects can be detected for any model settings. 4 Conclusions The regression results for the standard CAPM-model and its extended version with the oil price factor show that for the oil and gas industry and all of its sub-sectors the explanatory power significantly increases when the oil price is taken into consideration, while for other sectors there is only a slight difference in the R2s. Oil beta is significant at 1% for all but two of the oil and gas sub-sector indexes, for pipelines oil price is significant only at 5%, while alternative fuel industry exhibit no significant sensitivity to oil prices. From among the other top-level industry indexes only Basic materials (at 1%), Healthcare (at 5%) and Telecommunication (at 10%) exhibit sensitivity to oil prices. These results suggest that not surprisingly oil and gas companies have higher exposure to oil price changes than other industries. In the three-factor model settings we can measure higher R2s than in the standard CAPM, however lower than in the CAPM+oil model. The oil and gas industry as a whole exhibits non-significant SMB parameter, while its sub-sectors are significantly sensitive to the size factor. HML is significant for the broad oil and gas industry index and for all its sub-sector indexes (except Alternative Energy and its components). These results suggest that most of the oil companies behave like small firms but exhibit high book value compared to their market value. By adding the oil factor to the three-factor model we get a substantial increase in the explaining power of our regressions for the oil and gas sector and most of its sub-sectors (but the R2s are only slightly higher than in the CAPM+oil model settings). The only exception is the alternative fuel industry exhibits no significant sensitivity to oil prices. From among other industries oil is a significant factor only for Basic materials. Momentum factor is significant only for one sub-sector (Oil Equipment, Services and Distribution) at 10%, and there is no change in the in the explanatory power of the models compared to the previous versions. After separating the oil market conditions by adding a dummy variable we can see that the effects of the market, SMB, HML and MOM factors is the same with no regard of the oil price change of the given month, while the impact of the oil price factor is different based on the oil market conditions for the companies of the broad oil and gas industry, for oil and gas producers and for integrated oil and gas companies. Acknowledgments The authors would like to thank Dusán Timotity for his research assistance. Mihály Ormos acknowledges the support by the Jánoš Bolyai Research Scholarship of the Hungarian Academy of Sciences. References Aloui, R., Hammoudeh, S., Nguyen, D.K. (2013). A time-varying copula approach to oil and stock market dependence: The case of transition economies, Energy Economics, vol. 39, pp. 208-221. Arouri, H.M.E., Nguyen, K.D. (2010). Oil prices, stock markets and portfolio investment: Evidence from sector analysis in Europe over the last decade. Energy Policy, vol. 38(8), pp. 4528-4539. 40 Asteriou, D., Bashmakova, Y. (2013). Assessing the impact of oil returns on emerging stock markets: A panel data approach for ten Central and Eastern European Countries. Energy Economics, vol. 38, pp. 204-211. Basher, S. A., Sadorsky, P. (2006). Oil price risk and emerging stock markets, Global Finance Journal, vol. 17(2), pp. 224-251. Boyer, M. M., Filion, D. (2007). Common and fundamental factors in stock returns of Canadian oil and gas companies. Energy Economics, vol. 29(3), pp. 428-453. Carhart, M. M. (1997). On persistence in mutual fund performance. Journal of finance, vol. 52(1), pp. 57-82. Chen, N. F., Roll, R., Ross, S.A. (1986). Economic forces and the stock market, Journal of Business, vol. 59, pp. 383-403. El-Sharif, I., Brown, D. Burton, B. Nixon, B. Russell, A. (2005). Evidence on the nature and extent of the relationship between oil prices and equity values in the UK, Energy Economics, vol. 27(6), pp. 819-830. Faff, R., Brailsford, T. (1999). Oil price risk and the Australian stock market, Journal of Energy Finance & Development, vol. 4, pp. 69-87. Fama, E. F., French, K. R. (1992). The cross-section of expected stock returns. Journal of Finance, vol. 47(2), pp. 427-465. Fama, E. F., French, K. R. (1993). Common risk factors in the returns on stocks and bonds. Journal of Financial Economics, vol. 33(1), pp. 3-56. Fama, E. F., French, K. R. (1996). Multifactor explanations of asset pricing anomalies. Journal of Finance, vol. 51(1), pp. 55-84. Fang, C. R., You, S. Y. (2014). The impact of oil price shocks on the large emerging countries' stock prices: Evidence from China, India and Russia. International Review of Economics and Finance, vol. 29, pp. 330-338. Hamilton, J. D. (1983). Oil and the macroeconomy since World War II. Journal of Political Economy, vol. 91, pp. 228-248. Jones, D. W., Leiby, P. N., Paik, I. K. (2004). Oil Price Shocks and the Macroeconomy: What Has Been Learned Since 1996, The Energy Journal, vol. 25(2), pp. 1-32. Lintner, J. (1965). The valuation of risk assets and the selection of risky investments in stock portfolios and capital budgets. The review of economics and statistics, 13-37. Mohanty, S., Nandha, M., Bota, G. (2010). Oil shocks and stock returns: The case of the Central and Eastern European (CEE) oil and gas sectors, Emerging Markets Review, vol. 11(4), pp. 358-372. Mork, K. A. (1989). Oil and the macroeconomy when prices go up and down: an extension of Hamilton's results, Journal of Political Economy, vol. 97, pp. 740-744. Mossin, J. (1966). Equilibrium in a Capital Asset Market, Econometrica, vol. 34(4), pp. 468-483. Nandha, M, Brooks, R. (2009). Oil prices and transport sector returns: an international analysis. Review of Quantitative Finance and Accounting, vol. 33(4), pp. 393-409 Nandha, M., Faff, R. (2008). Does oil move equity prices? A global view, Energy Economics, vol. 30(3), pp. 986-997. Nandha, M., Hammoudeh, S. (2007). Systematic risk, and oil price and exchange rate sensitivities in Asia-Pacific stock markets, Research in International Business and Finance, vol. 21(2), pp. 326-341. Narayan, P.K, Sharma, S.S. (2011). New evidence on oil price and firm returns, Journal of Banking and Finance, vol. 35(12), pp. 3253-3262. 41 Oberndorfer, U. (2009). Energy prices, volatility, and the stock market: Evidence from the Eurozone, Energy Policy, vol. 37(12), pp. 5787-5795. Ramos, S.B., Veiga, H. (2011). Risk factors in oil and gas industry returns: International evidence. Energy Economics, vol. 33(3), pp. 525-542. Sharpe, W. F. (1964). Capital asset prices: A theory of market equilibrium under conditions of risk. Journal of Finance, vol. 19(3), pp. 425-442. Treynor, 1 L. (1962). Toward a Theory of Market Value of Risky Assets. Unpublished manuscript. Subsequently published in Korajczyk, R. A. (1999). Asset Pricing and Portfolio Performance: Models, Strategy and Performance Metrics, Risk Books pp. 15-22. 42 Specific Factors of the Contemporary Development of the Czech Real Estate Market Roman Brauner University of Technology Faculty of Business and Management, Department of Economics Kolejní 2906/4, 612 00 Brno, Czech Republic E-mail: brauner@fbm.vutbr.cz Abstract: The paper deals with the problems of the present development of the Czech real estate market in the conditions of long-term artificial under-valuation of the Czech koruna by the Czech National Bank. Primarily it analyzes the Czech real estate market development during the past 10 years, and investigates specific factors which influenced it in the course of that period. An integral part of the subject is also comparison of the Czech real estate market with other markets in the Czech Republic, mainly the stock and bond markets. Further, dependence of interest rates is analysed. In conclusion the paper predicts a future development based on the reached results, which would occur in case the artificial under-valuation of the Czech koruna continues over the long term. Keywords: financial system, interest rate, real estate market, central banks, inflation JEL codes: E42, E43, E58, G15 1 Introduction The effort to support economic growth changes certain factors which then may impact standard values of different markets depending on them. Already since 2008, when the whole world was hit by an economic crisis, all economies have been trying to encourage growth through standard as well as non-standard instruments. Their implementation leads to the change in consumers' behaviour and consequently to price adjustment. One of the instruments of influencing consumption growth is the reduction of interest rates, or possibly foreign exchange market intervention towards own currency. If consumers are pushed by different interventions to not save their assets for later use, they look for alternatives of their utilization. One of them is purchase of properties, for purposes of both getting a home for oneself and as possible investment of financial assets. 2 Methodologies and Data Mainly the secondary data from the Czech National Bank database and the Czech Statistical Office were used in the analyses and examined with a focus on interdependencies and mutual relations. The analysis results are then used to predict a possible future development of prices on the real estate market. The data was processed within a MS Excel 2007 programme. The examined data was from the period starting at the end of 2006 until the first quarter of 2016. The used methodology combines the qualitative and quantitative analysis. The qualitative component of the completed research draws on the findings of economic theory and their subsequent comparison with the current way of management of the world's largest economies. Regarding economics the following theoretical approaches are used: "Classical theory of interest rates", "Liquidity preference theory of interest rates", "The loanable funds theory", and "The rational expectations theory of interest rates", while monetary economics is also represented, namely by "Fisher's quantity theory of money". As for the quantitative component, it is based on economic data taken over from the Eurostat database, US department of the Treasury and Trading Economics. 3 Results and Discussion Investment in real estate is a long-term form of financial capital placement. With regard to many consumers undertaking this form of investment once or twice in a lifetime, it is 43 appropriate to consider relevant parameters affecting its changes. To enable the prediction of future development we will analyze the relations between the values of real estate market prices and the development of inflation rates, PRIBOR interbank interest rate announced by the Czech National Bank, which impact the interest rates of commercial mortgage banks, and the term account interest rates. Or possibly further important factors with an impact on the other variables such as ČNB foreign exchange intervention. Analysis of development of real estate prices Investments in real estate are made by a whole range of subjects. First of all they may be end users (households), who buy a property for their own needs, i.e. housing. An integral part of the real estate market are small and big investors or companies who place their assets in real estate with regard to both preserving their value and income from rent. Investment in real estate should be a type of long-term stable investment, however also in this segment significant and sometimes even unexpected fluctuations in terms of its immediate value can be encountered. Figure 1 Price Index of Real Estate 2010=100 no Source: Czech national bank Based on the analysis and Chart 1 it can be stated that if extreme situations such as the year 2008 and the current period of 2015-2016 are omitted, the real estate value with regard to certain fluctuations is constant. When examining the causes of these extreme situations we can subsequently predict whether this "constant" level will move to a different level to where it is now. Up until 2007 we could observe slow though continuous growth in the real estate market. However the beginning of 2008 saw a sudden increase in real estate prices. In the third quarter of 2008 this rapid increase changed into a deep slump which lasted until the beginning of 2010. After this hectic period the real estate market somewhat stabilized and until the beginning of 2014 it rather slowly decreased. Since 2014 the real estate prices have continued to increase and they tend to approach the levels existing in the extreme situation in 2008, while this increase is slower but it has continued for three years now. Real estate prices are to a large extent influenced by demand. Although the investment in real estate requires large amounts of financial resources, in large part these operations are financed from loans, particularly from mortgage loans. Therefore interest rates are among the major factors influencing the mood in the real estate market. The interest rate levels, especially those for mortgage loans, are in large part impacted by an interbank reference interest rate announced by ČNB. Analysis of development of interbank interest rates and inflation The level of Prague Interbank Offered Rate (PRIBOR) which is used as a price source for establishing interest rates and yields from different financial products such as bonds, financial derivatives, mortgage loans and suchlike, is also used as an instrument for 44 inflation control, therefore the examined area analyzes the relation between the development of inflation and interest rate adjustments responding to it. Figure 2 Development of Inflation and PRIBOR Interest Rate in the Period of 2006-2016 ■INFLATION ■PRIBOR 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 n Dioioi^i^i^cacacaaiaiaioooTHTHTHfNfNfNmmm'í'í'ímmmio PERIOD The Czech National Bank establishes a level between 2-4% as a safe inflation target over a long period of time. If this level is exceeded it starts to employ available methods to reduce it. As the chart shows, inflation began to rise disproportionately in 2008, when it surpassed all predictions by both domestic and foreign analysts. This rise was due to more than one factor, e.g. the expected increase of value added tax from 5% to 9% that traders "included" in their prices even before its introduction, and primarily the direct and undirect effects of the past surge in energy and food prices at the global level. Moreover, the labour cost increase sped up in the past quarters as indicated by some signals, in contrast to the slowing growth of productivity of work in the same period, and resulted in the sharp increase in unit labourcosts. (Tomšík 2016) Czech national bank responded by employing one of the most widely used instruments, i.e. by reducing interest rates which it started applying in November 2008. Through this regulator ČNB intends to attain increased consumption of all subjects and thereby the rise of inflation. This increase occurred at the end of 2012, when the trend turned around and inflation plummeted to the very bottom. Since ČNB had already "exhausted" to some extent the room for regulation by means of interest rate reduction, which in that period neared a zero level, on 7 November 2013 the Bank Board decided to start using the foreign exchange rate as another instrument for easing currency conditions. This decision said that "in the case of need" the Bank would intervene in the foreign exchange market with the aim of weakening the Czech koruna to maintain the exchange rate of the Czech koruna against the Euro "close" to 27 CZK/EUR. ČNB adopted this commitment unilaterally, justifying it as the measure to prevent excessive strengthening of the koruna exchange rate below the set level through its interventions i.e. selling and purchasing of foreign currencies. On the other hand, the weakening above this level leaves the offer and demand in the foreign exchange market to development (Tomšík 2016). The measure resulted in the accelerated weakening of the exchange rate of the Czech koruna above the set limit, which then remained slightly above it and approached the limit of 27 CZK/EUR as late as July 2015. That is why ČNB did not directly intervene into the exchange rate development during this interval. Subsequently, after the exchange rate value settled at the set limit, ČNB launched an automatic process of interventions from 17 July 2015 to January 2016, in a volume of over EUR 11 bn. As for economic growth in that period, it lagged behind expectations according to ČNB at the end of 2015, because of a slower growth of investments (both public and private) and the government consumption. For this year, the slowing of growth is expected due to decline in governmental investments with only a slow start of the new period of EU funds. (Tomšík 2016). 45 Figure 3 Inflation with and without Foregin Exchange Rate Source Czech national bank According to ČNB, inflation would likely fall into negative territory and economic growth could slow down even more without the exchange rate commitment. This measure further had an impact on the growth of wages in the business sphere: in the 4th quarter of 2015 it lagged behind the CNB prediction, but since 2013 it has increasingly sped up, which trend is expected to continue by ČNB also in 2016. The volume of wages has increased in real terms by almost 10% in Czech crowns since the adoption of the exchange rate commitment and by approximately 5% in Euros as well. Analysis of development of interest rates for mortgage loans and term accounts Given the fact that the Prague Interbank Offered Rate PRIBOR is one of the key factors influencing the interest rates of commercial banks providing both consumer or mortgage loans, as well as possibilities of depositing surplus resources e.g. in savings and term accounts, the decrease was also reflected in these. Figure 4 Development of Interest Rate of Saving Accounts up to 3 Years (Light Lines) and PRIBOR Interest Rate (Dark Lines) in the Period of 2006-2016 6, 5, I— 4, z LU O 1, 0, T—I T—I T—I C\l C\l C\l CO CO CO LT) LT) LT) irjirjirjt-vt---t---oooooocncncnooo oooooooooooo*-h<-h|t| 95% Coef. Interval Insurance sum Claims in MTPL .3102 .4186 0.74 0.459 -.5101 1.1305 Age .0106 .0146 0.72 0.469 -.0181 .0393 Gender .8468 .3639 2.33 0.020 .1336 1.5610 Region -.1091 .0659 -1.66 0.098 -.2383 .0201 Brand .03895 .0189 2.6 0.040 .0019 .07604 Construction year -.1075 .0535 -2.01 0.044 -.2124 -.0027 Engine volume -1.8307 .4060 -4.51 0.000 -2.6265 -1.0348 _cons 219.39 107.13 2.5 0.041 9.43 429.35 Claims in Casco Claims in MTPL .6116 .3174 1.93 0.054 -.0106 1.2337 Age -.0086 .0114 -0.76 0.449 -.0310 .0137 Gender .2325 .2957 0.79 0.432 -.3469 .8120 Region -.0124 .0570 -0.22 0.828 -.1241 .0993 Brand -.0127 .0148 -0.86 0.389 -.0417 .0162 Construction year -.0098 .0404 -0.24 0.809 -.0888 .0693 Engine volume .1076 .1961 0.55 0.583 -.2767 .4919 cons 19.60 80.99 0.24 0.809 -139.15 178.34 /athrho -.1132 .1951 -0.58 0.562 -.4955 .2691 rho -.1127 .1926 -.4586 .2628 Source: Authors'own calculations We failed to reject hypothesis that correlation between insurance sum and number of claims in Casco insurance equals 0 (A2 = .3369, df = 1, p-value =.5616). In the 54 regression, we controlled for two groups of characteristics used for risk type categorization as well as pricing by insurer, i.e. age, gender, region, automobile year of construction, automobile brand engine volume and MTPL claims. Our result is in line with a stream of previous research concluding that the relation between coverage and the number (as well as size) of claims is significantly influenced by the driver's and car's characteristics (Zavadil, 2015). According to Spindler (2012), the absence of a correlation between coverage and risk can be consistent with the presence of asymmetric information. This reasoning is applicable when individuals have private information not only about their risk but also about their risk aversion (DeMeza and Webb, 2001). Risk aversion is not simply observable for insurer, as this concept is complex and risk aversion could vary under different state of the art (Péliová, 2014). On the other hand, our result is also in line with Chiappori and Sálanie (2000), Dionne et al. (2001), and Chiappori et al. (2006) who found that insurance coverage and insurance risk occurrence are not correlated. 4 Conclusions The paper focuses on the information asymmetry in the insurance market. We use the micro-data about automobile insurance from Slovak insurance company to verify presence of information asymmetry in insurance relation. Based on the results of the empirical data analysis, we conclude that we have not find any correlation between coverage and loss in our dataset of Casco insurance. This result does not prove absence of information asymmetry in the insurance market and it could imply that private information of individuals affects this analysis. Our study has few limitations. The most important is the size of our sample that could affect the results of our analysis. However, we understand this study as a first draft and we would like to replicate our estimations on bigger sample. Also application of Test with Unused Observables could increase reliability of our results. Test with Unused Observables predict existence of a characteristic that is known to insured, but unknown or not used by the insurer and this characteristic is positively correlated with coverage and loss (Finkelstein and Poterba, 2006). Further research is needed in this area. Acknowledgments This paper is part of a research grant No. 1/0849/15 entitled 'Economic and social aspects of the information asymmetry in the insurance market' and research grant No. 1/0431/14 entitled 'Insurance Relationship as a Key Element of the Insurance Sector Operation in the Context of Socio-economic Changes' provided by the Ministry of Education, Science, Research and Sport of the Slovak Republic. References Cohen, A., Einav, L. (2007). Estimating Risk Preferences from Deductible Choice. The American Economic Review, vol. 97(3), pp. 745-88. Arrow, K. J. (1963). Uncertainty and the welfare economics of medical care. The American economic review, vol. 53(5), pp. 941-973. Chiappori, P. A., Sálanie, B. (2000). Testing for asymmetric information in insurance markets. Journal of political Economy, vol. 108(1), pp. 56-78. Chiappori, P. A., Sálanie, B. (2003). Testing Contract Theory: A Survey of Some Recent Work. In: Dewatripont, M., Hansen, L, and Turnovsky, S. eds., Advances in Economics and Econometrics, Cambridge University Press. Chiappori, P. A., Jullien, B., Sálanie, B., Sálanie, F. (2006). Asymmetric information in insurance: General testable implications. The RAND Journal of Economics, vol. 37(4), pp. 783-798. 55 Cohen, A., Siegelman, P. (2010). Testing for adverse selection in insurance markets. Journal of Risk and Insurance, vol. 77(1), pp. 39-84. Cutler, D. M., Zeckhauser, R. J. (2000). The anatomy of health insurance. In: Arrow, K. J. ed., Handbook of health economics, vol. 1, Elsevier, pp. 563-643. De Meza, D., Webb, D. C. (2001). Advantageous selection in insurance markets. RAND Journal of Economics, vol. 32(2), pp. 249-262. Daňhel, J. (2002). K problému asymetrie informace v pojišťovnictví. Politická ekonomie, vol. 50(6), pp. 809-813. Dionne, G., Fombaron, N., Doherty, N. (2013). Adverse selection in insurance contracting. In: Dionne, C. ed., Handbook of Insurance. New York: Springer, pp. 231-280. Dionne, G., Gouriéroux, C, Vanasse, C. (2001). Testing for evidence of adverse selection in the automobile insurance market: A comment. Journal of Political Economy, vol. 109(2), pp. 444-453. Einav, L, Finkelstein, A., Schrimpf, P. (2010). Optimal mandates and the welfare cost of asymmetric information: Evidence from the UK annuity market. Econometrica, vol. 78(3), pp. 1031-1092. Finkelstein, A., Poterba, 1 (2014). Testing for asymmetric information using "unused observables" in insurance markets: Evidence from the UK annuity market. Journal of Risk and Insurance, vol. 81(4), pp. 709-734. Pauly, M. V. (1968). The economics of moral hazard: comment. The American Economic Review, vol. 58(3), pp. 531-537. Pauly, M. V. (1974). Overinsurance and public provision of insurance: The roles of moral hazard and adverse selection. The Quarterly Journal of Economics, vol. 88(1), pp. 44-62. Péliová, 1 (2014). How do the risk free investment options change our decisions under risk? In: Deev, O., Kajurová, V., Krajíček, 1, eds., European Financial Systems 2014: proceedings of the 11th International Scientific Conference. Brno, Czech Republic: Masaryk University, pp. 469-474. Rothschild, M., Stiglitz, J. (1976). Equilibrium in competitive insurance markets: An essay on the economics of imperfect information. Netherlands: Springer. Spindler, M. (2012). Asymmetric information and unobserved heterogeneity in the accident insurance. MEA Working paper. Spinnewijn, J. (2012). Heterogeneity, demand for insurance and adverse selection. Working paper. Tumpach, M., Baštincová, A. (2014). Cost and Benefit of Accounting Information in Slovakia: Do We Need to Redefine Relevance? In: Deev, O., Kajurová, V., Krajíček, J., eds., European Financial Systems 2014: proceedings of the 11th International Scientific Conference. Brno, Czech Republic: Masaryk University, pp. 655-676. Weisburd, S. (2015). Identifying Moral Hazard in Car Insurance Contracts. Review of Economics and Statistics, vol. 97(2), pp. 301-313. Zavadil, T. (2015). Do the better insured cause more damage? Testing for asymmetric information in car insurance. Journal of Risk and Insurance, vol. 82(4), pp. 865-889. 56 The Limitations of E-commerce Development in Full Operating Cycle Firms: V4 Countries Case Emilia Brozyna1, Grzegorz Michalski2, Guenter Blendinger3, Ahmed Ahmidat4 1,2 Wroclaw University of Economics Faculty of Egineering and Economics, Department of Corporate Finance Management Komandorska 118, PL53345 Wroclaw, Poland E-mail: Grzegorz.Michalski@ue.wroc.pl, Emilia.Brozyna@ue.wroc.pl 3,4Technical University of Kosice Faculty of Economics Bozeny Nemcovej 26/32, 040 01 Kosice-Sever, Slovakia E-mail: guenter.blendinger@hs-heilbronn.de, ahmed.ahmidat@gmail.com, Abstract: Financial efficiency of an entities operating in full operating cycle conditions is influenced by environmental characteristics. That paper analyzes one of the financial results measured by cash tied in inventories in e-commerce full operating cycle financially constrained V4 entities. In recent decades, the countries of Central and Eastern Europe there are significant changes in national economies. Many of them joined the European Union in 2004. Among the EU Member States noteworthy is the largest group of countries of Central and Eastern Europe: Czechia, Poland, Slovakia and Hungary. On 15 February 1991 they have concluded an agreement on the formation of the Visegrad Group. It was aimed at expanding cooperation between these countries and in the initial phase of accession to the European Union and NATO. Analysis of the changes that have taken place in companies using online sales in the countries of the Visegrad Group is the subject of this article. Keywords: financing e-commerce, financial analysis, financial performance, V4 financial reality, financial liquidity management JEL codes: C58, G02, G32, 016, P43 1 Introduction Financial treasury and cash management influence financial performance of e-commerce entities. The main purpose of the paper is a scientific discussion about the influence of inventories levels in enterprises selling online. The research includes companies from the countries of the Visegrad Group, called in brief V4. In the initial stage, these countries were of a similar economic conditions (Jasova etal., 2016), (Tkacova, Sinicakova, 2015), (Soltes, Gavurova, 2014), (Michalski, 2016b), (Reznakova, Karas, 2015), (Raisova et. al., 2014). Name of the Visegrad Group was formed during the meeting of the presidents of Czechoslovakia and the Polish and Hungarian Prime Minister. This meeting took place at the castle in the Hungarian town of Visegrad. It has been planned exactly in this group, because these countries have not only convergent main foreign policy goals, but also enjoy the possibility of its implementation (Novotna, Luhan, 2012). Visegrad Group established International Visegrad Fund. These countries from the beginning of political transformation build their competitive potential. Giving priorities for investment in the industrial sector industries that rely on local raw materials, and private industries that are based on the production of the chemical industry and petrochemical, As well as expansion in some industries based on agriculture products and marine resources sector in order to reduce dependence on the raw material sector. V4 countries were concentrated on the possibility of creating some of the devices and the institutions that support export activity physically and morally. They focused on domestic production to cover domestic demand and reduce dependence on imports and restrict the goods and products that the local economy is unable to produce them locally (Ahmidat, 2016), (Michalski, 2016a), (Michalski, 2015c), (Merickova et. al., 2015), (Horvatova et al., 2014), (Cheben et al., 2015), (Gavurova, Soltes, 2016), 57 (Gavurova, Soltes, 2014), (Brozyna et. al., 2015), (Bern et al., 2015b), (Michalski, 2015b), (Soltes, Gavurova, 2015). They do this in order to meet the competitive forces in quite the single European market and the global markets (Zielinska-Glebocka and Gawlikowska-Hueckel, 2013). Even for rather small economies, such type of cooperation can be of great significance (Pavlicek, Kristoufek, 2015), (Michalski, 2015a), (Bern et al., 2014b), (Bern, Michalski, 2016), (Bartak, Gavurova, 2015), (Szczygiel et. al., 2015). Visegrad countries were different from the other countries of the former communist bloc. In other countries, internal changes were generally much less advanced, and strive for EU membership much longer or as was the case in Slovenia, the road was much shorter. At present, all countries of the Visegrad Group are the members of the European Union (Krajicek, 2013), (Michalski, 2008), (Galas et al., 2015), (Michalski, 2010), (Bern et al., 2015c), (Bern et al., 2014a). Electronic commerce (e-commerce, e-commerce) as defined by the Polish Central Statistical Office includes the transactions carried out by the network. They can be based on IP or other computer networks. Goods and services are ordered over those networks, but the payment and the ultimate delivery of the ordered goods can be made in or outside the network. Literature describes the internet and e-commerce as an indispensable element of the development process (Lawrence, 2013), (Michalski, 2008). This is some of the attributes associated with the e-commerce revolution (Brozyna etal., 2015), (Bern et al., 2015a) that has brought about a fundamental change in the conceptualization of commercial transactions: economic cost, convenience, sustainable value creation and product diversity (Michalski, 2008). Definition of an objective and measurable financial criteria of what is considered sustainable value creation for corporations is key to understand full operating cycle firm position and direct firms in the right strategic direction for financial success. Analyzing corporate annual reports unveil different approaches to measure the financial value creation of a firm (Blendinger, 2016), (Michalski, Brozyna, Soroczynska, 2015). For this paper we deliberately looked at the so called Value Added. It's supposed to quantify the financial value which is added during a given fiscal year taking into account the capital invested to provide the operational assets. For fiscal year 2015 Daimler for instance stated the value add to be 5675 T€ increasing from 4416 T€ in 2014. Looking at the factors in the formula used to calculate Value Added , it becomes obvious that inventory levels are an important element which shows the high relevance of it for this paper. The formula takes net profit and deducts net assets multiplied by cost of capital as percentage. In a long lasting manufacturing firm, property, plant and equipment besides inventory is the key factor of the net asset value. In a typical young e-commerce firm, however, property, plant and equipment are often low, hence, inventory levels become the crucial component in this Value Added consideration. It also shows that this value add calculation and the logic behind makes fully sense, especially when looking at it from the key objective of good corporate governance which is to create sustainable value for a company. The formula and calculation can be seen as a basis to compare value adds of different companies allowing to benchmark and subsequently develop the best in class approach for financially proper comparisons (Blendinger, 2016). E-commerce is associated production and sale of goods by modern information and communication technologies (Sedlacek, Valouch, 2009), (Szopinski, 2013). Each transaction conducted over the internet is the result of steps: search, order, payment and delivery. According to Kraska e-commerce is commercial transactions via telecommunication networks, coupled with making payments for goods and services. This takes place without direct contact between the parties (Kraska, 2004). During the past years steadily increasing the number of internet users has created a possibility to get to know the advantages and benefits offered by electronic commerce (Kim et al., 2011), (lo Storto, 2013). 58 2 Methodology and Data The test procedure is based on the method of multiple case studies. The research will be companies implementing online sales in countries belongs to the Visegrad Group. The project will be mainly used methods of descriptive statistics and financial analysis. Empirical data are derived from the financial statements of companies operating in V4 countries. Selected research units will differ from each other in terms of market offer and the number of employees. An in-depth financial analysis reports will focus in particular on measuring the effectiveness of activities. In addition, literature studies and analysis of extensive statistical data will allow for the emergence of strategic factors affecting the competitiveness of enterprises operating in the field of internet commerce. This will allow to assess the impact of factors specific to the investigated company on the relationship between the use of modern forms of sales and competitive advantage. The study began by calculating the index Inventories rotation in days. It indicates how many days the company renews its levels of inventories to realize sales. The formula to calculate the index can be presented as follows: Inventories rotation in days = (((Si+S0H2) x 365) -r- SA (1) where: INVi - levels of inventories in current year; INV0 - levels of inventories in previous year; SA - sale in current year. 3 Results and Discussion Modern concepts of inventory management are focused on maximum reduction of differences between the intensity of use of the levels of inventories and the rate of supply, in order to obtain the continuity of material flows with minimal inventories. The most important goal for the organization becomes an increase in the worth of the business successfully applying competitive means. Modern organizations are forced to seek alternative means for resolving business problems. Companies can no longer afford to lose their money in e-business initiatives without developing and using suitable means to support the appropriate level of inventories. Inventories rotation in days for each of the countries of the V4 are presented in Table 1. Table 1 Inventories Rotation in Days increase increase increase increase 2009 2010 2010 2011 2011 2012 2012 2013 2013 Czechia 331 54 62% 102 91% 42 -59% 46 8% Hungary 158 91 -42% 70 -24% 493 606% 99 -80% Poland 189 108 -43% 1175 985% 35 -97% 137 293% Slovakia 1964 1476 -25% 120 -92% 481 300% 506 5% Source: Own study based on data from e-commerce firms reported in Database Amadeus product of Bureau van Dijk, [date: 2016 MAY 10] Inventories rotation in days in Czechia rise between year 2009 and 2011. In 2012 it fell from 102,36 to 42,33. In last examined year in Czechia inventories rotation increase by 8,06%. In Hungary we can see decrease in every year, beside 2012. In this year inventories rotation reach the value of 492,94 days. In Poland the value rise sharply in 2011 by 984,82%. Slovakia reach a peak in 2009 year with value 1964,28 days. 59 Country with the biggest money levels tied in inventories is Poland. Average value of inventories in Poland is 1600 thousand euro. Other countries have the money levels tied in inventories between 600 and 1200 thousand euro. The next stage of the study was examined the money levels tied in inventories in comparison to each country. There was conducted Student t-test to examine whether these values differ from each other statistically. The results for Czechia compared to each country are presented in a separate table. Figure 1 Levels of Inventories in E-commerce Companies in V4 Countries 1650 -| 1550 - 1450 -1350 - -a 1250 650 -j— 2009 2010 2011 2012 2013 Source: Own study based on data from e-commerce firms reported in Database Amadeus product of Bureau van Dijk, [date: 2016 MAY 10] Table 2 T-student Test for E-commerce Companies in Czechia and Hungary Average Average X Df St. Dev. St. Dev. cz HU i P CZ HU Inv. rotation 2013 45,7 98,6 0,922 1734 0,357 81,3 935 Inv. rotation 2012 42,3 492,9 0,481 1588 0,631 63 15348 Inv. rotation 102,4 69,8 0,894 1463 0,371 1059 320 2011 Inv. rotation 2010 53,7 91,4 0,801 1440 0,423 280 725 Inv. rotation 2009 33,1 157,6 0,701 1112 0,483 40 2582 Source: Own study based on data from e-commerce firms reported in Database Amadeus product of Bureau van Dijk, [date: 2016 MAY 10] 60 In any of the tested years p-value is lower than 0,05. In enterprises selling online in Czechia we cannot find the relationship between enterprises selling online in Hungary. The same as in Hungary case in any of the tested years p-value is lower than 0,05. We cannot find the relationship between enterprises selling online in Hungary and enterprises selling online in Czechia. Table 3 T-student Test for E-commerce Companies in Czechia and Poland Average cz Average PL T df P St. Dev. CZ St. Dev. PL Inventories rotation 2013 48 137 -0,806 656 0,421 81,3 1857 Inventories rotation 2012 42 35 1,342 639 0,180 63 73 Inventories rotation 2011 102 1175 -0,831 607 0,406 1059 20795 Inventories rotation 2010 54 108 -0,584 559 0,560 280 1444 Inventories rotation 2009 33 158 -0,701 1112 0,483 40 2582 Source: Own study based on data from e-commerce firms reported in Database Amadeus product of Bureau van Dijk, [date: 2016 MAY 10] Table 4 T-student Test for E-commerce Companies in Czechia and Slovakia Average CZ Average SK T df P St. Dev. CZ St. Dev. SK Inventories rotation 2013 46 505 -1,346 490 0,179 81,3 5583 Inventories rotation 2012 42 481 -1,676 488 0,094 63 4287 Inventories rotation 2011 102 120 -0,217 476 0,828 1059 669 Inventories rotation 2010 54 1476 -1,159 451 0,247 280 19218 Inventories rotation 2009 33 1964 -1,156 372 0,249 40 24341 Source: Own study based on data from e-commerce firms reported in Database Amadeus product of Bureau van Dijk, [date: 2016 MAY 10] The last table shows that there is no relationship between inventories rotation in e-commerce companies in Czechia and Slovakia. 61 4 Conclusions Problems in the development of e-commerce may be due to the characteristics typical of the post-socialist backwardness constraints. The economic reforms associated with the transition to a market economy resulted in rising unemployment and general impoverishment of the population. The European Union currently consists of many countries, and some of them are countries of the former Eastern bloc. Europe's center of gravity shifts. The process of enlargement of the European Union not only drives the changes in the new countries, but also leads to a change of the whole of Europe. Acknowledgments The presented work and results is part of monothematic cycle realized as part of grant titled: Cash management in small and medium enterprises that use full operating cycle. The work is supported by National Science Centre, and financed from the Polish budget resources in the years 2015-2018 as the research project DEC-2014/13/B/HS4/00192. The presented work and results is also part of monothematic cycle realized as part of grant titled: Determinants of capital structure in nonprofit organizations. The work is supported by National Science Centre, and financed from the Polish budget resources in the years 2016-2019 as the research project DEC-DEC-2015/19/B/HS4/01686. References Ahmidat, A. (2016). Oil impact on production rates in the Libyan economy (analytical study during the period 1995-2008), unpublished material. Bartak, M., Gavurova, B. (2015). Economics and social aspects of long-term care in the context of the Czech republic and the Slovak republic EU membership, Proceedings of 12th International Scientific Conference: Economic Policy in The European Union Member Countries, pp. 35-44. Bern, A., Michalskí, G. (2016). Hospital profitability vs. selected healthcare system indicators. CEFE 2015 - Central European Conference in Finance and Economics. Technical University of Kosice, Kosice, pp. 52-61. Bern, A., Predkiewicz, K., Predkiewicz, P., Ucieklak-Jez, P. (2014b). Hospital's Size as the Determinant of Financial Liquidity. Proceedings of the 11th International Scientific Conference European Financial Systems 2014, pp. 41-48. Bern, A., Predkiewicz, K., Predkiewicz, P., Ucieklak-Jez, P. (2014a). Determinants of Hospital's Financial Liquidity. Procedia Economics and Finance, no. 12, pp. 27-36. Bern, A., Predkiewicz, P., Ucieklak-Jez, P., Siedlecki, R. (2015a). Profitability versus Debt in Hospital Industry. European Financial Systems 2015. Proceedings of the 12th International Scientific Conference, pp. 20-27. Bern, A., Predkiewicz, P., Ucieklak-Jez, P., Siedlecki, R. (2015c). Impact of hospital's profitability on structure of its liabilities, Strategica 2015 Proceedings, pp. 657-665. Bern, A., Siedlecki, R., Predkiewicz, P., Ucieklak-Jez, P., Hajdikova, T. (2015b). Hospital's financial health assessment. Gradient method's application. Enterprise and Competitive Environment. Conference Proceedings, pp. 76-85. Bledinger, G. (2016). Best in class approach and finetune the formula to allow financially proper comparisons, unpublished material. Brozyna E., Michalskí, G., Soroczynska, J. (2015). E-commerce as a Factor Supporting the Competitiveness of Small and Medium-Sized Manufacturing Enterprises. Central European Conference in Finance and Economics (CEFE2015), Technical University of Kosice, 2015, pp. 80-90. Cheben, J., Lancaric, D., Savov, R., Toth, M., Tluchor, J. (2015). Towards Sustainable Marketing: Strategy in Slovak Companies. Amfiteátru Economic, 17(40), pp. 855. 62 Galas, S., Galas, A., Zelenakova, M., Zvijakova, L, Fialova, 1, Kubickova, H. (2015). Environmental impact assessment in the visegrad group countries. Environmental Impact Assessment Review, 55, pp. 11-20. Gavurova, B., Soltes, M. (2016). System of day surgery in Slovakia: analysis of pediatric day surgery discrepancies in the regions and their importance in strategy of its development, E & M EKONÓMIE A MANAGEMENT, vol. 19, no. 1. Gavurova, B., Soltes, V. (2014). Importance of day surgery clinics specialization to the financing on health care. In SGEM2014 Conference on Psychology and Psychiatry, Sociology and Healthcare, Education, vol. 2, pp. 399-408. Horvatova, E., Schwarzova, M., Horvat, J. (2014). Theoretical and practical aspects of liquidity measurement and Basel III. Proceedings of the 8-th International Conference on Currency, Banking and International Finance, pp. 111-120. Jasova, E., Cermakova, K., Kaderabkova, B., Prochazka, P. (2016). Influence of institutional factors on structural and cyclical unemployment in the countries of the visegrad group. Politická Ekonómie, vol. 64(1), pp. 34-50. Kim, M., Chung, N., Lee, C. (2011). The effect of perceived trust on electronic commerce: Shopping online for tourism products and services in South Korea. Tourism Management, vol. 32(2), pp. 256-265. Krajicek, J. (2013). Cash Management in Practice. European Financial Systems 2013, pp. 176-180. Kraska, M. (2004). Elektroniczna Gospodarka w Polsce - Raport 2004, Instytut Logistyki i Magazynowania, Poznan, pp. 1-10. Lawrence, J. (2013). Barriers hindering ecommerce adoption: A case study of kurdistan region of Iraq. Technology diffusion and adoption: Global complexity, global innovation, pp. 152-165. lo Storto, C. (2013). Evaluating ecommerce websites cognitive efficiency: An integrative framework based on data envelopment analysis. Applied ergonomics, vol. 44(6), pp. 1004-1014. Merickova, B. M., Nemec, J., Svidronova, M. (2015). Co-creation in Local Public Services Delivery Innovation: Slovak Experience. Lex Local is, vol. 13(3), pp. 521. Merkevicius, J., Davidaviciene, V., Raudeliuniene, 1, Buleca, 1 (2015). Virtual organization: specifics of creation of personnel management system. E+ M Ekonómie a Management, vol. 4, pp. 200-211. Michalskí, G. (2010). Planning optimal from the firm value creation perspective, levels of operating cash investments, Romanian Journal of Economic Forecasting, vol. 13(1). Michalskí, G. (2015a). Agency Costs in Small and Medium Wood Industry Enterprises with Full Operating Cycle and Cash Levels. Procedia Economics and Finance, vol. 34, pp. 461-468. Michalskí, G. (2015b). Full operating cycle and its influence on enterprise characteristics: V4 food and beverages processing firms case. 18th International Scientific Conference Enterprise and the Competitive Environment, Brno, pp. 559-568. Michalskí, G. (2015c). Relation Between Cash Levels and Debt in Small and Medium Wood and Furniture Industry Enterprises with Full Operating Cycle. Procedia Economics and Finance, vol. 34, pp. 469-476. Michalskí, G. (2016). Risk pressure and inventories levels. Influence of risk sensitivity on working capital levels. Economic Computation and Economic Cybernetics Studies and Research, vol. 50(1). 63 Michalski, G. (2016b). Full operating cycle influence on the food and beverages processing firms characteristics, Agricultural Economics-Zemedelska Ekonomika, vol. 62(2). Novotna, V., Luhan, J. (2012). Dynamical model of e-commerce using differential equations with delay, Innovation Vision 2020: Sustainable growth, Entrepreneurship, and Economic Development, Proceedings of the 19th International Business Information Management Association Conference, vol. 1, 2012, pp. 473-481. Pavlicek, 1, Kristoufek, L. (2015). Nowcasting unemployment rates with google searches: Evidence from the visegrad group countries. PLoS ONE, vol. 10(5). Raisova, M., Buleca, 1, Michalski, G. (2014). Food processing firms inventory levels in hard times. 2004-2012 Slovak, Czech and Polish enterprises case. Procedia Economics and Finance, vol. 12, pp. 557-564. Reznakova, M., Karas, M. (2015). The Prediction Capabilities of Bankruptcy Models in a Different Environment: An example of the Altman Model under the Conditions in the Visegrad Group Countries. Ekonomicky Casopis, vol. 63(6), pp. 617-633. Sedlacek, J., Valouch, P. (2009). Fair Value in the Price Regulation of Natural Monopoly. E&M EKONOMIEA MANAGEMENT, vol. 12(2), pp. 6-14. Soltes, M., Gavurova, B. (2015). Quantification and comparison of avoidable mortality-causal relations and modification of concepts. Technological and Economic Development of Economy, vol. 21(6), pp. 917-938. Soltes, V., Gavurova, B. (2014). Innovation policy as the main accelerator of increasing the competitiveness of small and medium-sized enterprises in Slovakia. Procedia Economics and Finance, vol. 15, pp. 1478-1485. Szczygiel, N., Rutkowska-Podolska, M., Michalski, G. (2015). Information and Communication Technologies in Healthcare: Still Innovation or Reality? Innovative and Entrepreneurial Value - creating Approach in Healthcare Management, 5th Central European Conference in Regional Science Conference Proceedings, pp. 1020-1029. Szopinski, T. (2013). Czynniki determinujace korzystanie z handlu elektronicznego przez konsumentow, Handel Wewnetrzny, vol. 6(347), pp. 32-42. Tkacova, A., Sinicakova, M. (2015). New composite leading indicator of the Hungarian business cycle. Acta Oeconomica, vol. 65(3), pp. 479-501. Zielinska-Glebocka, A., Gawlikowska-Hueckel, K. (2013). Konkurencyjnosc miedzynarodowa i regionalna panstw Grupy Wyszehradzkiej. Wydawnictwo Uniwersytetu Gdanskiego, pp. 10-62. 64 Optimising the Slovak Tax Policy and Tax System Performance Emil Burak1, Juraj Nemec2 1 School of Economics and Management of Public Administration Furdekova 3240/16, 851 04 Bratislava, Slovak Republic E-mail: burak@nextra.sk 2 Masaryk University 2 Faculty of Economics and Administration, Department of Public Economics Lipova 41a, 602 00 Brno, Czech Republic E-mail: juraj.nemec@econ.muni.cz Abstract: There is a rich body of general scientific research dealing with issues such as optimum tax systems and optimum tax policies; however, there is limited scientific research focused on these issues specifically in Slovakia. To help fill this gap, our paper defines and discusses the most significant elements of the future optimisation of the Slovak tax system. The paper is based on primary and secondary data; it combines qualitative and quantitative research methods. The paper's main basis is the comparison of perceived needs by tax officials, collected in the course of our research, with existing evidence. The most frequent responses (simplify tax collection, decrease tax bureaucracy, provide better information about the tax system to businesses and citizens, increase the level of risk connected with tax evasion, and improve tax administration services) are very much in line with the positions of existing scientific and professional literature. However, the second most frequent answer, to decrease the tax burden, indicates a level of tax illusion even at the level of tax administration professionals. Keywords: tax administration, Slovakia, optimizing, opinion of tax officials JEL codes: H20, H26,D04 1 Introduction There is a rich body of general scientific research dealing with issues such as optimum tax systems and optimum tax policies; however, there is limited scientific research focused on these issues specifically in Slovakia. Considering that there are several university programmes dealing with tax policy, the fact of recent massive tax administration reforms in Slovakia, and the very negative findings by recent academic studies concerning the levels of administrative and compliance costs of taxation in Slovakia (Pompura, 2012 and Cizmarik, 2013), this situation is rather surprising. The goal of our paper is to define and to discuss the most important questions connected with the future optimisation of the Slovak tax system. The paper is based on primary and secondary data; it combines qualitative and quantitative research methods. It is based mainly on a comparison between the perceived needs by tax officials and the existing evidence. This comparison serves as the basis for the paper's conclusions and policy proposals. Literature review The first integrated concept of how to construct tax systems was presented by Smith (2005), whose principles of taxation, involved in the canons of taxation, formed the starting point for the study of the theory and practice of tax administration. Justice, certainty, convenience, and efficiency are principles that informed the development of contemporary taxation theory. The most frequently investigated issue is efficiency, something that appears to be a constant problem in tax administration. The high numbers and low productivity of tax officers and the huge costs for taxpayers to comply with tax system requirements have always been a real problem in tax administration. 65 The costs of taxation may vary over time and place, and they may be analysed in either of two ways. The first group of authors uses the term 'administrative costs of taxation' only to cover the expenses of the public sector (Sandford et al., 1989). The second group of authors, most notably Stiglitz (1989), divides the costs into administrative costs of taxation and the indirect expenses of the private sector (the incurred compliance expenses of taxation). Authors adhering to this theory of administrative costs in Slovakia include Hamernikova and Kubátova (2000) and Pekova (2002). There have been many important international studies about this topic (Aim, 1996, Evans, 2003, Chittenden et al, 2005, Lignier and Evans, 2012, Maimer, 1995, Mirrlees, 1971, Sandford, 1989 and 1995, Slemrod and Sorum, 1984, Susila and Pope, 2012, Tran Nam et al, 2000, Vaillancourt, 1987). There have also been some studies in the Central European region (Bayer, 2013, J i I ková and Pavel, 2006, Klun 2004, Klun and Blazic, 2005, Pavel and Vitek, 2012, Vitek, 2008, Teperová and Kubantová, 2013). Other dimensions of the tax system are less frequently investigated, even in international literature, and there is a near-absence of scientific articles in Central Europe. The few books on taxation including issues connected with optimisation of the tax system include Kubátova, Vybihal et al., 2004 and Kubátova, 2009. 2 Methodology and Data Our primary research was conducted between February 2013 and February 2016. During this period, we interviewed 282 executive tax officials participating in training at the tax school of the Slovak Financial Office (the requirement for inclusion in the sample was a minimum of four years of practice in tax administration). Our request was rather simple: Try to define the most significant elements of the possible optimisation of the Slovak tax system.' All written responses were processed and summarised by the authors. In the following text, we first highlight the responses with the highest frequency (Table 1). We next compare the most frequent suggestions provided by the respondents to existing data about the Slovak tax system. This comparison serves as the basis for the final part of this paper: conclusions and recommendations. Table 1: Selected Responses of Tax Officials Most frequent responses - suggestions Frequency Simplify tax collection, decrease tax bureaucracy 158 Decrease the tax burden 145 Provide better information about the tax system to businesses and citizens 110 Increase the level of risk connected with tax evasion 86 It is difficult to optimise the tax policy and the tax system, because there is 80 no optimum model available Prevent the transfer of Slovak firms to tax havens by lowering the direct 43 and indirect tax burden Utilise existing international good practices 33 Educate taxpayers - taxes are not the worst issue in the world 31 Be administratively simple, using low and stable tax rates and providing 27 effective tax administration services Improve tax administration services 26 Important but infrequent response Frequency Create a whistle-blowing system for reporting tax evasion 4 Source: Own research 3 Results and Discussion In this section, we comment on the main expressed opinions of the interviewed tax officials. We support or criticise their perceived priorities using existing data and research. 66 Simplify tax collection, decrease tax bureaucracy (+ improve tax administration services) The fact that tax officials perceive the Slovak tax system as complicated and not customer friendly, generating significant tax administration and tax compliance costs, can be evaluated as a positive finding. Current academic studies have documented a critical situation. Pompura (2012) calculated the administrative costs of taxation in Slovakia and estimated them using standard approaches employed by other scientists (see: Vitek, 2008, Vitkova & Vitek, 2012). The results are shown in Table 2. Table 2: Administrative Costs as a Percentage of Tax Revenues, by Specified Tax % 2004 2005 2006 2007 2008 2009 2010 2011 Income tax of individuals - X 1.77 1.96 1.64 1.48 1.62 1.81 1.65 Employees Income tax of individuals - 1.98 5.86 7.64 7.92 7.04 7.92 30.76 25.51 Entrepreneurs Corporate income tax 2.99 1.62 1.52 1.37 1.23 1.18 2.11 1.65 Income tax - lump 1.33 2.43 2.01 1.19 1.45 1.61 2.04 2.25 sum form Property tax 0.53 1.82 1.81 19.32 14.80 14.61 13.42 31.80 VAT 3.63 1.32 1.28 1.41 1.47 1.52 1.52 1.59 Road Tax 4.10 1.97 1.72 1.16 1.52 1.00 1.26 1.12 Source: Pompura, 2012 These data document that tax administration in Slovakia is among the most expensive in the world - see Table 3. Table 3: Taxation Level and Administrative Costs of Taxation: Selected Countries Countries according to their administrative costs of taxation (%) Countries according to their tax revenues to GDP < 20% 20-30% 30-40% Over 40% - 0.60 USA Sweden Ireland, Austria, Denmark, 0.61 - 0.80 Korea Spain, New Finland, Zealand Germany, Norway 0.81 - 1.00 Mexico Turkey France Hungary, 1.01 - 1.20 Netherlands, Luxembourg UK 1.21 - 1.40 Canada Belgium, Czech Republic Poland, 1.40 + Japan Portugal, Slova kia Source: OECD, 2011. The existence of high administrative costs has been reflected in the actions of the Slovak government, which in 2012 undertook the large scale tax system reform called UNITAS. The main goals of UNITAS are to improve the flow and use of information and to merge the collection of all taxes and social contributions under one administration. According to the official data collected on the basis of OECD country reports, the administrative costs 67 of taxation were decreased through UNITAS reform by about 50%; however, such data are hard to believe. Cizmarik (2013) estimated the compliance costs of taxation. Because his first calculations were really negative, he tried to provide other possible alternative calculations. However, for even the most optimistic calculation, the costs incurred to taxpayers are really high; see Table 4. Table 4: Alternative Calculations of Compliance Costs of Income Taxation in Slovakia Alternative CC to tax revenues total CC to tax revenues physical persons CC to tax revenues legal persons Original results 73.37 % 839.02 % 47.13 % Alternative A 53.11 % 242.29 % 35.98 % Alternative B 62.36 % 713.17 % 40.06 % Alternative C 40.12 % 637.04 % 19.67 % Alternative D 61.36 % 734.61 % 38.29 % Alternative E 62.99 % 599.71 % 44.59 % Alternative A+B+C 24.69 % 156.37 % 12.76 % Source: Cizmarik, 2013 Customer service is an element that is almost missing from Slovak tax administration. One specific step forward would be creating a 'customer friendly' tax administration system that would provide taxpayers with better information and increase their trust in the tax system. This would include several necessary and important improvements. For example, it should reflect the fact that there is no system for really effective and binding tax advice from tax offices. The 'tax case law precedents list' may be, but is not necessarily, fully respected by tax authorities and especially court decisions. Because of this, the decisions of tax officials exerting tax control may differ even with similar cases (Burak and Mazary, 2012). This situation creates important levels of critical uncertainty. Another issue is too-frequent changes in tax legislation, complicating life and increasing tax compliance costs. Decrease the tax burden This suggested improvement is connected with a frequent criticism of the Slovak tax system as a system creating excessive tax burdens on taxpayers. However, it is only partially compatible with reality. When looking only at income taxes, Slovakia is one of the European countries with really low tax rates. The revenues from income taxation in Slovakia are among the lowest in the European Union. According to official Eurostat data, Slovak income (direct) tax revenues in 2012 represented only 5.6 % of the GDP (one of the three lowest), even though more countries have lower implicit tax rates for income taxation (we will return to this finding later in the text). The issue of a high tax burden can be associated with social contributions, increasing labour costs. Provide better information about the tax system to businesses and citizens This response occurred rather frequently, but we do not feel that the situation is so critical in terms of the scope and scale of information provided by the tax system to taxpayers. Unfortunately, Slovakia did not provide data for benchmarking the communication strategies and channels of the national tax administration for the report produced by the Intra-European Organisation of Tax Administrations (IOTA, 2013), thus we are not able to provide exact data. However, this opinion by tax officials might be connected with the fact that the level of fiscal and tax literacy of taxpayers is frequently evaluated as low - as documented for example by Cizmarik (2013) and Solilova and Nerudova (2013). Their data confirm the 68 existence of tax illusions. During the research on compliance costs, Cizmarik (2013) asked respondents their opinion about the level of the compliance costs of taxation. The responses were rather surprising - 8% of respondents felt that compliance costs were marginal, and 31% felt that their level was fully acceptable. Their views of the payroll system were not so positive, but even so 46% of respondents felt that its costs were acceptable. Increase the level of risk connected with tax evasion Taking into the account information provided above - that tax rates are moderate, but tax revenues are very 'small' - the issue of tax evasion and its costs and benefits should be more frequently mentioned by tax officials. Tax evasion in Slovakia is estimated to be really high (official data estimate approximately three billion EUR yearly). Orviska and Hudson (2003) clearly indicate that tax evasion is a common approach in Slovak business, in part perhaps because the risk of punishment is low. For example, Slovakia is the only EU country to apply the principle of'effective regret'. Even taxpayers caught by the tax office for evasion can retrospectively pay their tax assessments, plus a 10% surcharge, and remain 'clean', provided they pay up before the final court decision. The tax officials who responded to our survey did not see tax evasion as a core issue, and very few of them provided suggestions for how to cope with it, such as education or whistle blowing. 4 Conclusions The main source for this paper is the responses of tax officials regarding their opinions about the most significant possibilities for optimising the Slovak tax policy and administration. The answers indicate that these tax officials have an imperfect picture about the pros and cons of the current situation. Some of the most frequent responses, especially to simplify tax collection, to decrease tax bureaucracy, to provide better information about the tax system to businesses and citizens, and to improve tax administration services, are very much in line with the positions of existing scientific and professional literature. However, the second most frequent answer, to decrease the tax burden, indicates some level of tax illusion even at the level of tax administration professionals. The specific issue is the relatively low frequency of the response 'increase the level of risk connected with tax evasion'. All of the existing data indicate that the level of tax evasion in Slovakia is rather high, with several important factors behind this situation. One of the most significant factors is low level of risk connected with tax evasion - few cases are discovered and penalised, and moreover Slovakia seems to be the only EU country to apply the principle of 'effective regret'. Even taxpayers caught by the tax office for evasion can retrospectively pay their tax assessments, plus a 10% surcharge, and remain 'clean', provided they pay up before the final court decision. The responses of the tax officials and the critical data-based evaluation of those responses provide simple tax policy advice. Future tax reforms should focus on structural changes (UNITAS reform contents) and especially on decreasing the scope of tax bureaucracy, providing better tax administration services to taxpayers (this is now accomplished mainly by the gradual improvement of e-government services). The ultimate goal should include increasing the tax fraud risk level. Acknowledgments This research was completed with support from the research project 'Performance Management in Public Administration: Theory and Practice in the Czech Republic and Other CEE Countries', ID (CEP) GA16-13119S. 69 References Alm, J. (1996). What is an "Optimal" Tax System? National Tax Journal, vol. 49(1), pp. 117-133. Bayer, O. (2013). Research of Estimates of Tax Revenue: An Overview. European Financial and Accounting Journal, vol. 2013(3), pp. 59-73. IEOTA (2013). Benchmarking: Pilot Exercise. Retrieved from: https://www.iota-tax.org/publication/iota-booklet-2012-2013. Burak, M., Mazary, M. (2012). Tax Optimisation: Selected Issues. Tax Tribune, issue 28, pp. 207-211. Cizmarik, P. (2013). Vyvolané náklady zdanenia. Brno: ESF MU. Evans, C. (2003). Studying the Studies: An overview of recent research into taxation operating costs. eJournal of Tax Research, vol. 1(1), pp. 64-82. Hamernikova, B., Kubátova, K. (2000). Verejné finance. Praha: Eurolex Bohemia. Hasseldine, J., Hansford, A. (2002). The Compliance Burden of the VAT. Further Evidence from the UK. Australian Tax Forum, issue 17, pp. 369-388. Chittenden, F., Kauser, S., Poutziouris, P. (2005). PAYE-NIC compliance costs: empirical evidence from the UK SME economy. International Small Business Journal, vol. 23(6), pp. 635-656. Jilkova, 1, Pavel, J. (2006). Hodnoceni efektivnosti verejných vydajú na ochranu životního prostredí. Praha: IREAS. Klun, M. (2004). Compliance costs for personal income tax in a transition country: the case of Slovenia. Fiscal Studies vol. 25(1), pp. 93-104. Klun, M., Blazic, H. (2005). Tax compliance costs for companies in Slovenia and Croatia. Finanzarchiv, vol. 61(3), pp. 418-437. Kubátova, K. (2009). Danova teorie. Praha: ASPI Publishing. Kubátova, K., Vybihal, V. (2004). Optimalizace daňového systému CR. Praha: EUROLEX. Lignier, P., Evans, C. (2012). The Rise and Rise of Tax Compliance Costs for the Small Business Sector in Australia. Australian Tax Forum, vol. 27(3), pp. 615-672. Malmer, H. (1995). The Swedish Tax Reform in 1990-91 and Tax Compliance Costs in Sweden. In: Sandford, C. (ed.) Tax Compliance Costs. Measurement and policy. Bath: Fiscal Publications, pp. 226-262. Mirrlees, 1 A. (1971). An Exploration in the Theory of Optimum Income Taxation. The Review of Economic Studies, vol. 38(2), pp. 175-208. OECD (2011). Tax Administration in OECD and Selected non-OECD countries: Comparative Information Series. Paris: Organization for Economic Co-operation and Development. Pavel, J., Vítek, L. (2012). Transaction Costs of Environmental Taxation: the Administrative Burden. In: Milne, J. E. - Andersen, M. S. (eds.). Handbook of Research on Environmental Taxation. Cheltenham: Edward Elgar, pp. 273-282. Pavel, J., Vitek, L. (2015) Vyvolané náklady daňového systému v ČR. Politická ekonomie, vol. 63(3), pp. 317-330. Pekova, J. (2002). Verejne finance - uvod do problematiky. Praha: ASPI. Pompura, L. (2012). Hodnotenie a meranie výkonnosti daňovej správy: administratívne náklady zdanenia. Brno: ESF MU. Sandford, C. (1989). Administrative and Compliance Costs of Taxation. London: Fiscal Publications. 70 Sandford, C. (1995). Tax Compliance Costs. Measurement and policy. Bath: Fiscal Publications. Slemrod, 1 R., Sorum, N. (1984). The Compliance Cost of the U.S. Individual Income Tax System. The National Tax Journal, vol. 37(4), pp. 461-474. Smith, A. (2005). Inquiry into the Nature and Causes of the Wealth of Nations. Philadelphia: The Pennsylvania State University. Stiglitz, J. E. (1989). Economics of the Public Sector. New York: Norton. Susila, B., Pope, 1 (2012). The Tax Compliance Costs of Large Corporate Taxpayers in Indonesia. Australian Tax Forum, vol. 27(4), pp. 719-772. Solilova, V., Nerudova, D. (2013). Transfer pricing: General Model for Tax Planning. Ekonomicky časopis, vol. 61(6), pp. 597-617. Teperova, J., Kubantova, K. (2013). Omezeni a možnosti jednoho inkasního mista v České republice. Česky finanční a ucetni časopis, vol. 2013(1), pp. 61-76. Tran-Nam, B., Evans, C, Walpole, M., Ritchie, K. (2000). Tax Compliance Costs: Research Methodology and Empirical Evidence from Australia. National Tax Journal, vol. 53(2), pp. 229-252. Vaillancourt, F. (1987). The Compliance Cost of Taxes on Business and Individuals: A Review of the Evidence. Public Finance, vol. 42(3), pp. 395-414. Vitek, L. (2008). Ekonomická analýza zdaněni přijmu. Praha: IREAS. Vitkova, 1, Vitek, L. (2012). Spolocenske vyvolané náklady zdaněni. Acta Oeconomica Pragensia, vol. 2012(1), pp. 15-30. 71 Comparison of the Efficiency of Selected European Banking Sectors Liběna Černohorská1 1 University of Pardubice Faculty of Economics and Administration, Institute of Economics Science Studentská 95, 532 10 Pardubice, Czech Republic E-mail: libena.cernohorska@upce.cz Abstract: There is no generally accepted concept of efficiency nor is there a uniform system of indicators for measuring bank efficiency. It is even possible to use the method of financial analysis to measure bank efficiency. In this paper, the following ratios are used for measuring bank efficiency: ROA, ROE, total assets, nonperforming loans/total loans, quick liquid assets/total assets, quick liquid assets/short-term liabilities, loans/deposits, and capital adequacy. The goal of this paper is to assess the efficiency of Czech banks using cluster analysis on the basis of selected ratios and to conduct a comparison with bank efficiency in Poland, Austria, Greece, Portugal, France, and Slovakia. The collective ratios for the entire banking sector will be compared for the selected countries for the years 2010-2014. The cluster analysis demonstrates that the Czech banking sector is the most similar to the Slovakian sector. According to a combination of selected ratios, it is possible to designate the cluster composed of the Czech and Slovak banking sectors as being the cluster with the highest banking sector efficiency. It differs extensively from the cluster of Greece and Portugal. Keywords: bank, banking sector, banks efficiency, cluster analysis, the Principal Components Analysis JEL codes: G14, G21, C38 1 Introduction The banking system has become an important component in the economic sector of each country. Like other industries, the banking industry has its own unique characteristics and specifics that adapt by internal and external influences economic sector. Each state is required for the proper functioning of the economy needs a reliable a stable banking system, because the problems in the banking sector may have an impact on the entire financial sector. Each banking system of each country has its own specifics that influence global globalization. It operates on banking systems around the world. Each state receives it but in different ways. Some states retain more of their traditional banking features that arose during the development of the system, in turn, take some elements of the globalized economy. Banks are an inseparable part of life for all economical subjects (Hedvičáková and Svobodová, 2015). The bank stability and efficiencyjs an important assumption for function the financial markets (Teplý et al., 2010 or Černohorský, 2014). For qualified analysis, it is necessary to work with a time series of ratios and monitor the trends of their development over past periods of time (Tokarčíková et. al., 2014 or Svobodová, 2013). The aim of the article is to undertake a cluster analysis of efficiency of chosen banking sectors in the countries of Eupean Union - Slovak, Poland, Austria, Czech Republic, France, Portugal and Greece. The selected countries are countries represented in the European Union, which have variously developed financial markets and banking sectors. Based on a cluster analysis, the creation of clusters would result, in which individual banking sectors will exhibit similar values in the selected criteria. Based on current research literature on the efficiency of banks, it is evident that in terms of evaluating the efficiency of banks, that there is a wide range of views and measuring the efficiency is therefore very difficult. There are numerous methods of measuring efficiency and the fundamental question is what indicators we can use to measure that efficiency. 72 Efficiency is often understood in the same sense as performance and profitability (such as Atemnkeng and Nzongang, 2006 or Molyneux and Thornton, 1992). Where banks are run efficiently, the operational costs are reduced, leading to an increase in profits realised by the banks. The authors Richard, Devinney, et al. (2009) found an analysis of more than 213 articles in leading international journals which use particular indicators based on accounting data to measure efficiency; these indicators mainly include cash flow, financial results, revenues and their growth and asset profitability indicators . In measuring the efficiency of banks, profitability was used, for example, by Altunbas (1998), Bonin and Hasan (2005), Abbasoglu, Aysan and Gunes (2007) and Berger et al. (1993). These authors evaluate the profitability of banks using return on assets (ROA) or return on equity (ROE). Bonin and Hasan (2005) also monitored the amount of total deposits, total assets, loans and liquid assets. The size of a bank is judged by its total assets (Dabla-Norris and Floerkemeier (2007), Fuentes and Vergara (2003)). Indicators of total assets, loans, and total loans/total deposits are used to assess the efficiency of banks, in addition to ROE and ROA, as well as Berger et al. (1993). Groenveled and de Vries (2009) use the capital ratio when measuring the efficiency of banks. Very often the efficiency of banks is evaluated by means of their ownership structure (Fuentes and Vergara (2003), Bonin and Hasan (2005), Mester (1993)). Some authors take into account the cost of labour when measuring the efficiency of banks (Stavarek (2013), Tulekns (2006), Berger et al. (1993)) and the cost of capital (Berger et al. (1993)). Another factor influencing the efficiency of banks is the interest margin (Stavarek (2013) or Dabla-Norris and Floerkemeier (2007)). These last authors also use the indicator of quickly nonperforming loans/total loans, liquid assets/total assets and quick liquid assets/short-term liabilities. 2 Methodology and Data Evaluating bank efficiency is a relatively complicated analytical problem. There is no generally accepted concept of efficiency nor is there a uniform system of indicators for measuring bank efficiency. It is even possible to use the method of financial analysis to measure bank efficiency. The goal of financial analysis is to evaluate the financial ratios for efficiency and competitiveness that were achieved in prior periods of time. In this paper, the following ratios are used for measuring bank efficiency: ROA, ROE, total assets, nonperforming loans/total loans, quick liquid assets/total assets, quick liquid assets/short-term liabilities, loans/deposits, and capital adequacy. The collective ratios for the entire banking sector will be compared for the selected countries for the years 2010-2014. The necessary data were obtained from the Bankscope database and were chosen with regard to the specifics of the selected banking sectors, international accounting standards and information requirements for the banks. A comparison was made of the average values of the selected indicators in individual banking sectors. Further scientific study could use a longer time series of selected indicators of selected banking sectors for a more detailed analysis. It would be possible to monitor factors which affect the efficiency of banking sectors (such as the period before the financial crisis, the impact of the financial crisis on selected criteria and subsequently track the clusters created, etc.). The peer analysis allows make a comparison of the financial variables according to the tables and graphs. For this peer analysis will used the traditional methods of multiple statistical analysis, especially cluster analysis and principal components analysis. The method of cluster analysis was used to compare the efficiency of the Czech banking sector with the banking sectors of the other selected European countries. Cluster analysis divides the selected countries into clusters according to similarity. Using the method of principal component analysis, it was determined that there are two main components that jointly explain nearly three-quarters of the variability. Cluster analysis The primary access for determining the similarity of quantitative variables is the factor analysis. It is based on principal component analysis, which is used to reduce the size of 73 the job (instead of many variables for further calculations determined by a small number of principal components, which can be expressed as linear combinations of the original variables). The Principal components analysis is computed by the Singular Value Decomposition of X. (Friedman et al. (2013)) The general formula (2) is: D ... diagonal matrix consisting of the set of all eigenvalues of C along its principal diagonal, and 0 for all other elements U ... an n-by-n matrix, the columns of which are orthogonal unit vectors of length n called the left singular vectors of X; W ... a p-by-p whose columns is orthogonal unit vectors of length p and called the right singular vectors of X. In the Principal Components Analysis (PCA), the data are summarized as a linear combination of an orthonormal set of the vectors. The first principal component accounts for as much of the variability in the data as possible, and each successive component represents as much of the remaining variability as possible (Zou (2006)). Components accounting for maximal variance are retained while other components accounting for a trivial amount of variance are not retained. These techniques are typically used to analyse groups of correlated variables representing one or more common domains. The result of PCA enters into the factor analysis. It is aim to assess the structure and relationships of selected indicators to see if allowed by their division into groups, in which the indicators chosen from the same groups together more than correlated variables from different groups. Cluster analysis is a collective term covering a wide variety of techniques for delineating natural groups or clusters in data sets. The article will be used hierarchical agglomerative clustering. Hierarchical agglomerative clustering start at the bottom and at each level recursively merges a selected pair of clusters into single clusters. This produces a grouping at the next higher level with one less cluster. Algorithm of hierarchical agglomerative clustering begins with every observation representing a singleton cluster. At each of the N-l steps the closest two (least dissimilar) clusters are merged into a single cluster, producing one less cluster at the next higher level. (Friedman et al., 2013) In the first phase clustering calculated the relative distances of objects and writes them into a matrix. This leads to a square symmetric matrix D ~ ^ which has zeros on the main diagonal. It used for calculating the metric distance matrix is normally used and it called a Euclidean method. It is based on the geometric model (Klfmek, 2005). The objects characterized by p characters are assigned to the points p-dimensional Euclidean R S space Ep, then two dots ( ' ) it is defined by the Euclidean distance given by general formula (3): On the basis of the distance matrix can be launched the second phase calculations, also clustering. Clustering method was used furthest neighbour (called too complete linkage). Complete linkage agglomerative clustering takes the intergroup dissimilarity to be that of the furthest (most dissimilar) pair according to formula (4): X = UDVF (2) where (3) d(R,S) = max OeR {j(0.,0.)}for R±S (4) 74 where R S ' ... represent two such groups d(R,S) represent dissimilarity between R and S in computed from the set of pairwise observation dissimilarities ^(^'O,)^ where one member of the pair ^ is in R, and the other ^j is in S. Methods of clustering is selected based on the degree of credibility, and it cophenetic correlation coefficient "CC". The higher the value of the correlation coefficient cophenetic (a value close to 1), the greater the credibility and the choice of a suitable model cluster. (Friedman etal. (2013), Romesburg (2004)) The result is graphical figure called a dendrogram with provided a highly interpretable complete description of the hierarchical agglomerative clustering. 3 Results and Discussion The basic condition for performing cluster analysis is rejected claim that the data are affected by multicollinearity. Multicollinearity could very significantly affect the final quality of the clustering and classification of the individual elements in the resulting clusters. It is necessary to establish the correlation matrix. Then eliminate those criteria in assessing the relationship reaching the correlation coefficient higher than 0.7. If left criterion which the correlation coefficient is above 0.7. It is necessary to provide a justification for its further occurrence of cluster analysis. For more information see Friedman et al. (2013). Based on the results of the correlation matrix, the ratio of nonperforming loans/total loans was removed from the analysis. This indicator showed very high levels of correlation with ROE, as well as the proportion of quick liquid assets/short-term liabilities, which is highly correlated with ROA. To obtain information on the impact of these indicators, the principal components method was applied followed by a factor analysis. Both methods are used for visualising data and obtaining input information. Visualization of date using factor analysis The principal component method determined that there are two main components which together explain nearly three quarters of variability (Table 1). The first principal component depletes approximately 47.96% of the total variability in the data, the second approximately 25.81%. The results of the factor analysis bring Table 1 and Figure 1. Table 1 shows which criteria are important for further exploration in terms of classification into certain objects, respectively clusters (bold face type). Table 1 The Result of the Factor Analysis - Two Main Components First principal Second principal component 1 component 2 Total assets 0.39 0.80 Liquid assets/total assets 0.77 0.26 Loans/Deposits 0.60 0.04 Capital adequacy 0.62 0.64 ROA 0.91 -0.12 ROE -0.19 0.97 Source: Own calculation Graphic representation of the data visualisation from the factor analysis assumes the possible creation of approximately four relevant clusters (Figure 1). 75 Figure 1 Factor Analysis - Number of Clusters Source: Own calculation Result of cluster analysis The cluster analysis method was used in comparing selected banking sectors. This analysis divides the selected countries into clusters according to their similarities. To perform a cluster analysis, we have assumed an agglomerative hierarchical clustering. For more information see Romesburg (2004). It was followed by selecting the clustering procedures, namely, a clustering method (the furthest neighbour method, or complete linkage clustering using statistical software), and the distance calculation method (Euclidean distance). The clustering method was selected based on the degree of credibility, namely, a correlation coefficient. The degree of credibility, or closeness degree, has been verified by the correlation coefficient. The higher the value (i.e., approaching 1), the greater the credibility and the choice of a suitable cluster model. The correlation coefficient was chosen on the basis of achieving a value approaching 1 with the furthest neighbour method. A prerequisite to performing the cluster analysis is that the data is not affected by multicollinearity. Determining the relevant number of clusters was started from the clustering schedule, which determined the degree of distance of approximately 60%. Below this level, the relevant number of clusters was determined (Figure 2). The division of the countries into four clusters with the values of the individual indicators can be seen in Table 2. Determining the relevant number of clusters was started from the clustering schedule, which determined the degree of distance of approximately 60%. Below this level, the relevant number of clusters was determined (Figure 2). The division of the countries into four clusters with the values of the individual indicators can be seen in Table 2. Figure 2 The Dendrogram - Wards Method 100 w S *> - ! « 20 Or«aci Portuaoi PalanG ALitris Frjirc» Ciacr Sia*, Source: Own calculation 76 Table 2 Average Values of Chosen Indicators (in %) Change in total assets Liquid ass./Tot. ass. Loans/ Deposits Capit. adeq. ROA ROE First Slovak 1.4 36.2 110 17.93 1.3 9.1 cluster Czech R. 8.9 33.8 132 17.08 1.27 16.2 Second France 9.6 39.1 81 15.03 0.5 8.4 cluster Austria -5.8 24.5 87 15.83 0.1 5.5 Poland 4 21.4 90 14.91 1.1 14 Third Portugal -7.7 16.9 117 13.20 -0.7 -11 cluster Fourth Greece -10.8 29.9 89 13.50 1.4 -169 cluster Average total -0.06 26.44 88.14 15.35 0.71 -18.1 Source: Own calculation by Bankscope 4 Conclusions Based on the cluster analysis, four clusters were created. From the point of view of the efficiency of the banking sectors using selected indicators with the first principal component, which explains almost 48% of the variability of the investigated group, the greatest correlations were the ratio of liquid assets to total assets and ROA. In the ratio of liquid assets to total assets, the best values were achieved by France, the Czech Republic and Slovakia. The Czech Republic and Slovakia faultlessly exceeded the average ROA limit, but not France. The average ROA value was also exceeded by Poland, but which does not record comparable results to those countries in the ratio of liquid assets/total assets. The first cluster is formed by the Czech Republic and Slovakia. Because France and Poland lag behind in one or the other indicators, they are clustered into another cluster together with Austria. The third cluster consists of only one country -Portugal. Portugal achieved the worst results in both indicators listed above. The Greek banking sector achieved better results than Portugal, but because it achieved very low levels in the indicator corresponding to the second part of the component, it forms a separate cluster. Especially in terms of ROE, it achieved high negative values, which prevents it from being compared to other countries, and thus Greece and forms the fourth separate cluster. Depending on the combination of selected indicators, the cluster composed of the Czech Republic and Slovakia can be qualified as a cluster with the highest possible efficiency in the banking sector. The first cluster achieves significantly better values of the indicators monitored than other banking sectors. These two banking sectors were not impacted by the global financial crisis (compared to Greece and Portugal, and to some extent, France). The average values of the monitored indicators of the first cluster are significantly above the average for all the markers in the selected banking sectors. Acknowledgments The paper has been created with the financial support of The Czech Science Foundation (project GACR No. 14-02108S - The nexus between sovereign and bank crises). References Altunbas, Y., Chakravarty, S. P. (1998). Efficiency measures and the banking structure in Europe. Economics Letters, vol. 60(2), pp. 205-208. Atemkeng, T., Nzongang, J. (2006). Market structure and profitability performance in the banking industry of CFA Countries: the case of commercial banks in Cameroon. Journal of Sustainable Development in Africa, vol. 8(2), pp. 1-14. Abbasoglu, O. F., Aysan, A. F., Gunes, A. (2007). Concentration, competition, efficiency and profitability of the Turkish banking sector in the post-crises period. Banks and Bank Systems, vol 2(3), pp. 106. 77 Bankscope. (2014). World banking information source. Bureau vank Dijk. Berger, A. N., Hancock, D., & Humphrey, D. B. (1993). Bank efficiency derived from the profit function. Journal of Banking & Finance, vol. 17(2), pp. 317-347. Bonin, J. P., Hasan, I., Wachtel, P. (2005). Bank performance, efficiency and ownership in transition countries. Journal of banking & finance, vol. 29(1), pp. 31-53. Černohorský, J. (2014). The Integration of Credit Markets. In Proceedings of 7th International Scientific Conference on Managing and Modelling of Financial Risks, pp. 127-135. Dabla-Norris, E., Floerkemeier, H. (2007). Bank efficiency and market structure: what determines banking spreads in Armenia?. IMF Working Papers, pp. 1-28. Friedman, 1, Hastie, T., Tibshirani, R. (2001). The elements of statistical learning, 1st ed. Berlin: Springer series in statistics. Fuentes, R., Vergara, M. (2003). Explaining bank efficiency: bank size or ownership structure?. In Proceedings of the VIII Meeting of the Research Network of Central Banks of the Americas, pp. 12-14. Groeneveld, 1 M., de Vries, B. (2009). European co-operative banks: first lessons of the subprime crisis. The International Journal of Cooperative Management, vol. 4, pp. 8-21. Hedvičáková, M., Svobodová, L. (2015). Analysis of Banking Fees and Clients'Needs. In Proceedings of the 12th International Scientific Conference on European Financial Systems 2015. Brno: Masarykova Univerzita, pp. 181-188. Klímek, P. (2005). Data mining a jeho využití. E+M Ekonomie a Management, vol. 8 (3), pp. 128 - 135. Mester, L. 1 (1993). Efficiency in the savings and loan industry. Journal of Banking & Finance, vol. 17(2), pp. 267-286. Molyneux, P., Thornton, 1 (1992). Determinants of European bank profitability: A note. Journal of banking & Finance, vol. 16(6), pp. 1173-1178. Richard, P. J., Devinney, T. M., Yip, G. S., Johnson, G. (2009). Measuring organizational performance: Towards methodological best practice. Journal of management, vol. 35(3), pp. 718-804. Romesburg, Ch. (2004). Cluster Analysis for Researchers. North Carolina: Lulu Press. Serdarevic, G., Teplý, P. (2011). The Efficiency of EU Merger Control During the Period 1990-2008*. Finance a Uver, vol. 61(3), pp. 252-276. Stavárek, D., Řepková, I. (2013). Efficiency in the Czech banking industry: A non-parametric approach. Acta Universitatis Agriculturae et Silviculturae Mendelianae Brunensis, vol. 60(2), pp. 357-366. Svobodová, L. (2013). Trends in the number of bankruptcies in the Czech Republic. In Proceedings of International Conference, Hradec economics days 2013, Hradec Králové: Univerzita Hradec Králové, pp. 393-399. Tokarčíková, E., Ponisciakova, O., Litvaj, I. (2014). Key Performance Indicators and their Exploitation in Decision-Making Process. In Proceedings of 18th International Conference. Transport means 2014, Kaunas: Kaunas University of Technology, pp. 372-375. Tulkens, H. (2006). On FDH efficiency analysis: some methodological issues and applications to retail banking, courts and urban transit. In Public goods, environmental externalities and fiscal competition, pp. 311-342. Zou, H., Hastie, T., Tibshirani, R. (2006). Sparse principal component analysis. In Journal of computational and Graphical Statistics, vol. 15(2), pp. 265-280. 78 Investment in Renewable Energy Technologies from the Perspective of Polish Venture Capital Funds Karolina Daszyriska-Zygadfo1, Tomasz Storiski2, Magdalena Ligus3 1 Wroclaw University of Economics Faculty of Management, Computer Sciences and Finance, Institute of Financial Management Komandorska 118/120, 53-345 Wroclaw, Poland E-mail: karolina.zygadlo@ue.wroc.pl 2 Wroclaw University of Economics Faculty of Management, Computer Sciences and Finance, Institute of Financial Management Komandorska 118/120, 53-345 Wroclaw, Poland E-mail: tomasz.slonski@ue.wroc.pl 3 Wroclaw University of Economics Faculty of Management, Computer Sciences and Finance, Institute of Financial Management Komandorska 118/120, 53-345 Wroclaw, Poland E-mail: magdalena.ligus@ue.wroc.pl Abstract: In the recent years ambitious targets to increase the share of renewable energy and reduce greenhouse gas emissions have been adopted by governments around the world. Variety of policies have been introduced. The question is what is the business potential and market response to that. This article adds a new perspective to the debate of business potential of investments in renewable energy sector. The aim of the paper is to focus on the venture capitalist interested in the new clean technology development and its diffusion. We analyze venture capitalists' perception of investing in projects from that sector, possibilities of financing and barriers that they meet in the process. We conducted survey and field study with managers of private equity funds from Poland. Our research results show that in the overall business history of the surveyed entities 67% of funds invested in projects related to renewable energy sector. In order to assure specific competencies for investing in renewable energy sector they were using external experts in most cases or past experience. If they specialize in the RE sector investments they were likely to employ their own experts in the investment department While main reason for not investing in that sector was lack of competence, but also unstable situation in the sector. Keywords: renewable energy, venture capital, environmental innovation, environmental policies JEL codes: Q55, Q58, G24 1 Introduction The growth potential for renewable energy investments attracts attention of private equity investors. Over next 25 years renewables will account for two thirds of total energy production. The New Energy Outlook for 2015 (Bloomberg 2016) finds that some $12.2 trillion will be invested in global power generation between 2015 and 2040. However, the majority of investments will take place in emerging market, the 22% will take place in OECD countries. In European perspective is particularly important since many countries had positioned themselves as a global leader in renewables investment. The peak of European investments of $132BN in 2011 has not been exceeded and recent sharp reversal of investment deviates the upward trend. The Bloomberg New Energy Fund (Bloomberg 2016) attributed this to fear over "the survival of the Euro, mistakes by policymakers, and some lingering effects from the global financial crisis". The existing literature underlines the importance of government support mechanism for renewable energy policies (Btirer, Wtinstenhagen 2009; Hoffman, Huisman 2012). Determined by the governments and the market regulators the investment climate is 79 shaped by different policies like: technology performance standards, tax credits or certificate trading etc. Policies are supported by the financial mechanisms which enhance investor to choose eco-efficient investments by increasing their returns on capital invested. Venture Capital and Private Equity (VC/PE) play important role in the sector. Focused on earning high returns the high risk capital is present in the beginning phase of technology development as well as in the last phase of mergers and acquisitions. Recent statistics draws an optimistic picture, since in 2015 venture capital and private equity investors invested $5.6bn into renewable energy sector, up 17% on the 2014 total but still far below the $12.2bn peak of 2008. However, it is worth to mention that the biggest VC/PE deal of last year was $500m in a well-established Chinese electric vehicle company NextEV. Our particular interest is the beginning stage of technology development, where the VC funds spend ca. $4.0bn (Bloomberg, 2016), 12% of global investments at this stage. Comparing to VC the other groups of early stage investors, Government R&D and Corporate Ventures, share different characteristics (Jackson, 2011). Since the technology adoption is responsible for at least a quarter of the cross-country variation of per capita income differences (Comin and Hobijn, 2006) we think country specific factors of technology development strongly influence the VC fund performance. This is the reason we focus on the VC fund which originate from single country, namely Poland, categorized by IMF as Emerging Europe (IMF 2016). Our objective is to focus on the venture capitalist interested in the new clean technology development and its diffusion. We would like to scrutiny the perception of development barriers and drivers of VC operations in renewable energy sector. Our research task is to identify the VC funds which declare the ability and readiness for investment in renewable energy sector. The survey by the questionnaire allows to collect information from Polish venture capital funds. Literature overview Early literature focuses on development constraints showing the main barriers (including financials) to possible growth opportunities of the renewable energy sector and measures to overcome them (Painuly, 2001). Although some of the barriers still exists, the rapid market expansion and changes in institutional landscape has altered the investment environment. Grubb (2004) presents the way the policies influence the renewable energy investment in the context of technology development cycle. The policies can be divided into the technology-push (early stage) and market-pull policies (late stage). Although appointed to different stages of technology development the policies are interconnected, since the technology-push policies are responsible for the "supply" of new technologies, whereas market-pull policies create a "demand" for new technologies. The pivotal moment in technology development cycle which, in fact, separates these two groups of polices is the successful introduction of the new technology to the market (Btirer and Wtistenhagen, 2009). The VC funds manage to bridge the gap between two groups of policies, which enables to provide high returns while facing: (1) relatively high attrition rate for projects within portfolio together with (2) large exposure to the market risk factors. The role of VC funds is not only to provide the capital for the most difficult stage of technology innovation called the cash flow "valley of death" (Murphy and Edwards, 2003) but also expertise and network to support entities. In order to do so, VC are able to find rapidly growing niche markets for technology with rapid growth opportunities, scalable ventures, and high returns (Hargadon and Kenney, 2014). In many cases, VC funds require relatively little capital for initial investments in the renewable energy sector, exploring the synergies with governmental support mechanisms. Ghosh, Nanda (2010) identified set of clean technologies with high-risk, low-capital characteristic perfectly suitable for Venture Capitalists. Majority of them originate from 80 renewable energy sector: energy efficiency software, electric drive trains, fuel cells/power storage, wind and solar components of unproved technologies. Renewable energy technologies are particularly challenging for VC since it takes longer for market acceptance than in traditionally safer VC investments like Internet technologies. With renewable energy sector, some technologies will take years to achieve market acceptance, and capital is needed for the duration (Tierney, 2011). Some surveys measure venture capitalists preferences to different policies of government support mechanisms. Regulatory risk is perceived as important role of market risk. The results of Btirer and Wtistenhagen (2009) show that VC fund managers prefer market-pull policies, while feed-in tariff policy is perceived as the most popular. However recent studies show marginally decreasing returns from investments supported by these policies (Criscuolo and Manon, 2015). A study of Hofman and Rusiman (2012) performed after the financial crisis shows decrease in popularity of many policies, especially for a trade based policies (i.e. certificate and C02 emission trading). That research was conducted on the sample of companies among which 35% represented early stage investments (4% of companies cover all investment stages), however the detailed statistics for smaller subsamples were not provided. Chassot et al. (2014) confirm that venture capitalist do not typically invest in the late stages of development and exhibit rather policy risk aversion. As a consequence they avoid investments if regulatory risk is perceived as high. On the other hand, they are willing to take some of the regulatory risk, especially if they worldviews accept government intervention over the "free market" attitude. An interesting resent research results confirm the positive correlation between perception of environmental policy stability and patenting applications in environmental technologies (Johnstone et al., 2010; Criscuolo and Manon, 2015). Polish market of Private Equity funds (PE) despite its dynamic growth, still finds itself at the early stage of development. In 2012, the ratio of PE investments to GDP amounted to 0.13%, a level much lower than the one recorded in more developed markets, such as Sweden (0.66%) and the UK (0.56%) According to the organization of the European Private Equity & Venture Capital Association (EVCA), Poland is a leader in the region of Central and Eastern Europe in terms of a number of companies financed (in 2014 there were 76 entities). Despite the development of the market, what is manifested by an increase in the number of funds and funded entities, the value of funds invested in recent years, recorded a decrease, which can be seen in figure 1. In accordance with the presented data, the share of venture capital funds in the private equity market in 2014 amounted to 7.7%, showing an increase of 1.3 percentage points in relation to the previous year 1. Figure 1 The Value of Investments in the PE Market in Poland and the Number of Financed Entities 800 100 2010 2011 2012 2013 2014 Value of investments (mln euro) Number of investments (right axis) Source: Central and Eastern Europe Statistics 2014, EVCA 81 As for the sector specialization of funds, the results of the survey presented in the KPMG report show that in 2004-2013 the most popular sectors for PE investors operating in Poland were information technology, media and communication (19%), industrial production (18%), medical industry (13%) and retail trade (12%). These sectors amounted to 62% of all investments in the period considered. However, in the case of energy industry, this share was about 4%. 2 Methodology and Data Survey was conducted on the sample of Venture Capital funds operating in Poland, data was collected on the basis of the questionnaire. Potential sample was constructed by excluding funds appearing in several groups, inactive funds and those not carrying out the recruitment for new investments from the overall population of Polish Venture Capital funds. Next step was to verify their investment policy. Attention was paid mainly to the preferred stage of implementation readiness for the ongoing investments and their compliance with the assumptions of Venture Capital investment (seed, start-up and early expansion). Additionally funds with clearly defined industry specification, which did not deal with renewable energy investments were excluded. Contact sample included 104 funds. In this group, there were several players who have been identified in the past as investors in the market of renewable energy (including Belchatow Kleszczow Industrial and Technological Park, Spinnaker Innovation - Incubator BonusCard, IdeaLab Centre for Innovation and Entrepreneurship), and also one fund that specializes in this industry - Skystone Capital. Surveyed players gained the capital from public funds, EU programs or from private investors, having assets under management in the value ranging from 2,5 up to 12 million EUR, operating on the market for over 2 years and having accomplished 10 to 30 capital entries. The survey was completely anonymous and consisted of 12 questions, as single-, multiple-choice and open question options. The answers were successfully obtained from 15 funds, resulting in the response rate of 23.4%. 3 Results and Discussion Survey results showed that 60% of the funds include in its investment policy sectoral or industrial specialization, with dominant sectors (most frequently mentioned by respondents) being: IT, energy, renewable energy (RE) and medicine. Due to the selection carried out on the contact sample, up to 87% of the surveyed funds indicated that their investment policy includes investments in the sector of renewable energy (RE), while 73.3% claims direct investments in the technologies of renewable energy, including 6.7% of energy producers and the same percentage of manufacturers or service providers for the RE sector (figure 2). Figure 2 "Does the Investment Policy of the Fund Include Investments in the Sector of Renewable Energy? If so, Identify in What Form."- Respondents' Answers 6,7% 6,7% ■ Technologies used in RE ■ Firms - energy producers ■ Firms - suppliers for RE ■ No RE coverage Source: Own elaboration. 82 Figure 3 "If the Fund Did not to Invest in Projects in the RE Sector, Select the Reason" - Respondents' Answers o% Lack of competence in the sector Lack of agreement as to the terms of the investment Unstable situation of the sector Lack of attractiveness of identified projects Lack of reaching final investment agreement Source: Own elaboration. In the overall business history of the surveyed entities 67% of funds invested in projects related to RE sector. In case of not investing in renewable energy sector, the reasons given to justify that were (figure 3): lack of competence in the sector (33%), lack of agreement as to the terms of the investment (17%), unstable situation of the sector (17%), lack of attractiveness of identified projects (17%), lack of reaching final investment agreement (17%). None of the analyzed funds did not carry more than five capital entries at the researched market. Study on sample of Polish VC funds to some extend confirms the earlier findings of Btirer and Wiistenhagen (2009) that regulatory risk is perceived as important market risk factor in the sector, due to the fact that unstable situation in the Polish RE sector results from regulatory environment uncertainty and government policies changes and discontinuities. Other reasons for not undertaking RE investments could also be connected with the findings of Chassot et al. (2014). Funds surveyed were asked to identify the specific competencies required to invest in the renewable energy sector (it was possible to select multiple answers). The results indicate that the majority (67%) of the funds have access to external experts in the industry, more than half of the funds has already completed at least one investment in a project from RE sector, while 33% of the fund's investment department employs people who have experience in investing in renewable energy sector. 13% of the funds did not prove the competence to invest in renewable energy. Results are presented in figure 4. Figure 4 "Does the Fund Have Specific Competence to Invest in Renewable Energy Sources (You Can Choose More than One)?" - Respondents' Answers 70% 60% 50% 40% 30% 20% 10% 0% Access to external experts Completed at least one Investment department Lack of competence in the industry investment in a project employs people who have from RE sector experience in investing in renewable energy sector Source: Own elaboration. It can be noticed that VC funds in Poland are willing to undertake projects from RE sector especially if they have proper competences to do so. They are using external experts in 5% 10% 15% 20% 25% 30% 35% 83 most cases or past experience. If they specialize in the RE sector investments they are likely to employ their own experts in the investment department. Among the introductory questions there was one about the financial sources that VC fund uses to invest in its projects. It was multiple-choice of answer question, so funds could choose more than one answer. Out of 12 questioned funds 11 (92%) used public funds and EU sources. Half of them used private investor funds and just 1 used capital sources at the stock exchange. Therefore it indicates high correlation between policy supported sector investments (RE) with particular funding strategies which stays in line with results of previous studies of Grubb (2004) and Chassot et al. (2014). 4 Conclusions Our research extends the up to date studies about VC funds' investments in RE sector by the issues of reasons for not undertaking these types of investments as well as required competencies used in order to do so. It also confirms previous studies findings in the field of high dependency of VC investment in RE sector with their prefunding by public or EU sources. It also supports the argument about high regulatory risk connected with RE investments that causes instability of the sector and unwillingness to invest in it by Polish VC funds. In our study large part (60%) of Venture Capital funds in its investment policy assumes sectoral specialization, while the selection of sectors indicates a diversification of investment portfolios. Among the dominant industries are: medicine, IT, energy and renewable energy. Most venture capital funds (87%) places in its investment policy projects in the renewable energy sector. This characteristic, to a certain extent, results from selection of the contact sample, which included the specification of investment policy of entities. The dominant area for investment in this market are the technologies used in renewable energy sector. Smaller interest is given to investment in energy producers (one response), manufacturers and service providers in the sector of renewable energy (one response), and separate installations (no responses). It may result from a bigger development potential perceived by funds in technological solutions in comparison with investments in companies or installations. That confirms the general findings about VC funds investment preferences (Hargadon and Kenney, 2014) and Ghosh and Nanda (2010) findings about type of clean-tech investments attractive to VC funds. 33% of the surveyed funds did not report in their history any investment in renewable energy sector. The main reason was the lack of competence of the fund in this sector. This could be due to the lack of access to external industry experts, lack of experience in the inputs of capital in the market or the lack of skilled employees in the investment departments of funds. Currently, lack of competence in the renewable energy sector is shown by less than half of the funds, that is only 13%, which may signify a considerable increase in interest in this sector. None of the surveyed funds reported more than 5 investments in RE sector. The reason for such a condition could be lack of sufficient number of attractive investment opportunities in this sector, as well as relatively high risk associated with it. Even though the sample of Polish VC funds engaged in RE sector investment might seem not representative conclusions drawn on the basis of the survey stay in line with the previous finds adding some novelty related to the competencies of VC funds in the RE sector as well as showing the reasons for not undertaking investments in it. Future research should be dedicated to risk assessment and perception of VC funds as well as analysis of rates of returns expected and obtained from investments in RE sector. That could shed light on the specifics of preferences of Polish VC funds and their risk acceptance level. 84 Acknowledgements This paper was prepared as part of the research projects "Value based management of investments in renewable energy sources" [UMO-2011/01/D/HS4/05925] and "Evaluation of the environmental effects in a cost-benefit analysis of the investments in low-emission energy sources" [UMO-2011/01/B/HS4/02322], executed by the Wroclaw University of Economics and financed by the National Science Centre, Poland. References Baum, J.A.C., Silverman, B.S. (2004). Picking winners or building them? Alliance, intellectual, and human capital as selection criteria in venture financing and performance of biotechnology start-ups. Journal of Business Venturing 19, pp. 411-436. BCirer, M. J., Wtistenhagen, R. (2009). Which renewable energy policy is a venture capitalist's best friend? Empirical evidence from a survey of international clean tech investors, Energy Policy, vol. 37, pp. 4997-5006. Comin, D. A., Hobijn, B. (2006), An Exploration of Technology Diffusion, NBER Working Paper Series, Working Paper 12314. Chassot, S., Hampl, N., Wtistenhagen, R. (2014). When energy policy meets free market capitalists: The moderating influence of worldviews on risk perception and renewable energy decisions, Energy Research & Social Science, vol. 3, pp. 143-151. Criscuolo, C, Menon, C. (2015). Environmental policies and risk finance in the green sector: Cross-country evidence. Energy Policy, vol. 83, pp. 38-56. Ghosh, S., Nanda, R. (2010). Venture capital investment in the clean energy sector. Harvard Business School Entrepreneurial Management Working Paper, (11-020). Gorman, M., Sahlman, W. A. (1989). What do venture capitalists do? Journal of Business Venturing, vol. 4, issue 4, pp. 231-248. Grubb, M. (2004). Technology Innovation and Climate Change Policy: An Overview of Issues and Options. Keio economic studies, vol. 41, no 2, pp. 103-132. Hargadon, A. B., Kenney, M. (2012). Misguided Policy? Following Venture Capital into Clean Technology, California Management Review, vol. 54, no. 2, winter, pp. 118 -139. Jackson, F. (2011). Corporates' ventures, Renewable Energy Focus, September/October, pp. 30-33. Johnstone, N., Hascic, I., Popp, D. (2010). Renewable energy policies and technological innovation: evidence based on patent counts. Environmental and resource economics, vol. 45, no 1, pp.133-155. Murphy, L.M., Edwards, P.L., (2003). Bridging the valley of death: transitioning from public to private sector financing. Painuly, J.P. (2001). Barriers to renewable energy penetration: a framework for analysis, Renewable Energy, vol. 24, pp. 73-89. Tierney, S. (2011). Venture Capital And Cleantech Symbiosis, The Industrial Geographer, vol. 8, Issue 2, pp. 63-85. Unruh, G. C. (2000). Understanding carbon lock-in. Energy Policy, vol. 28, pp. 817-30. Wtistenhagen, R., Teppo, T. (2006). Do venture capitalists really invest in good industries? Risk-return perceptions and path dependence in the emerging European energy VC. 85 Sustainable Value Creation - performance of European Manufacturing Companies Karolina Daszyriska-Zygadfo1 1 Wroclaw University of Economics Faculty of Management, Computer Sciences and Finance, Institute of Financial Management Komandorska 118/120, 53-345 Wroclaw, Poland E-mail: karolina.zygadlo@ue.wroc.pl Abstract: The research problem of this paper is how to measure the impact of corporate sustainability performance on company's value. Its aim is to show this impact in monetary terms, using opportunity costs based approach. Sustainable Value method presented and tested in this paper integrates environmental, social and economic factors. It measures the value created or destroyed by the company with the usage of particular set of resources, the value is referenced to the benchmark's value creation that could potentially be realized with the same set of resources. Research was conducted on manufacturing companies from EU-15 countries. It analyzes the prospects and advancement in sustainable value creation as well as mitigation of negative environmental impact in reference to value created by companies examined in the research of Figge, Hahn and coauthors in ADVANCE project (2006). Results show that majority of companies which were creating sustainable value in the period of 2001-2003 continued to obtain positive values in the period of 2004-2012, the same applied to value destructing companies. Keywords: sustainability, value creation, opportunity cost JEL codes: Q01, Q55, G32 1 Introduction Nowadays, global community is challenged by climate change issues. Businesses are also challenged to show how sustainable they are. From the corporate finance perspective, companies should be focused on wealth maximization while not being distracted by additionally imposed goals (Auppperle et al., 1985; Shleifer, 2004). Among theorists of the subject matter there is also a group that claims and proves that introducing social, environmental and governance goals into business conduct might be economically beneficial by improvement of: employees' efficiency due to higher job satisfaction; relations with clients and suppliers leading to better reputation, loyalty and favorable contracts' conditions (Guenster et al., 2010; Godfrey and Hatch, 2007). All that might have positive impact on financial performance of the companies leading to increase created value. Vast majority of researches aim at showing immediate or long term impact of performance related to sustainability on companies' financial results proving whether there exist a significant relationship between these two. Immediate effect is analyzed by event studies showing reaction of investors on the information disclosure about company being included (excluded) in (from) sustainability index (Consolandi et al., 2009; Cheung, 2011; Cheung and Roca, 2013; Daszyiiska-Zygadlo et al. 2014). The long term effect (usually calculated as a yearly effect over a certain period of time) is analyzed on the basis of financial performance (accounting or market measures) regressed over sustainability (or CSR) performance (scoring/rating result) (among the newest published research results being: Wang and Berens, 2015; Saeidi et al., 2015; Sloriski et al., 2016). Depending on the period of the analysis, proxies selected for financial performance and sustainability performance as well as a sample of companies and methodology of research assessed impact is either positive, negative or inconclusive and insignificant (Margolis et al., 2007). Therefore, it is hard to have a casting vote over the final conclusion on the relationship of sustainable performance of companies and their financial standing. 86 Potentially, it leaves room for studies driven by different objective and analyses of the impact of sustainability performance on value creation using different approach. The underlying problem might be the information load of sustainability performance disclosed and reported that would allow for inclusion in sustainability indices, scorings, ratings or databases. The drawback of researches conducted up-to-date on the big sample is that companies on average are moderately efficient in sustainability performance, they disclose all the necessary information, but in many cases there is no significant improvement or mitigation of negative environmental, social and governance impacts. This is why, assessment of sustainable value created could have much bigger potential in terms of explaining the impact of sustainability performance over value creation and financial results. This research takes initial steps to solve that puzzle and to identify the gap in the literature that could be explored in this and future studies. Additionally, a tool that shows the economic value of sustainability in monetary terms such as Sustainable Value (SV) can be beneficial for companies from managerial perspective. They could practically assess which social or environmental actions and investment projects undertaken create additional economic value and which destroy this value. In recent years, a big step towards creating analytical methods of sustainability assessment has been made. Including the ones that capture the sustainability effect or environmental or social effect in qualitative and quantitative way. Among recognized tools and metrics there are: FVTool, Measuring Impact Framework Methodology and Sustainable Value. Sustainability investments (understood as programs, initiatives or infrastructure investments done by companies to manage environmental and social risks and support the development needs of local communities) might be analyzed by FVTool developed in 2009 by International Finance Corporation's Infrastructure & Natural Resources Advisory team in partnership with Rio Tinto and Deloitte in order to justify their implementation from economic perspective. Measuring Impact Framework Methodology (by World Business Council for Sustainable Development and International Finance Corporation from 2008) could be used as complex tool for sustainability management within an organization, it includes quantitative and qualitative metrics which could be well aligned with more general and strategic goals of an organization. Sustainable Value initially introduced by Figge and Hahn (2004) and later implemented in ADVANCE project (2006) is an approach showing overall performance of the companies that report on 7 environmental resources, expressed in monetary terms in relation to a benchmark being GDP of EU15 in Gross Value Added. This method could be even perceived as a framework because it allows modifications of number and types of resources as well as a different choice of the benchmark (e.g. additional social resources and sector EBIT as a benchmark as proposed in Figge et al., 2014). Out of the three above mentioned methods SV seems to be the most aligned to the stated research problem of this paper. Due to the fact that its aim is to show the impact of corporate sustainability performance on value of the company in monetary terms. Moreover, sustainable value added reflects the value that could be generated if resources were relocated from inefficient to efficient users, assuming overall constant level of resources and all forms of capital being perfectly substitutable (weak form of sustainability) (Perman et al., 2003). 2 Methodology and Data Sustainable value (SV) methodology extends the logics of financial market to eco-efficiency theory. It deals with the pricing problem in monetary sustainability assessment, namely how to valuate resources that are not explicitly priced. It introduces opportunity cost thinking to sustainability assessment: if the return an economic entity achieves with the use of resources exceeds the opportunity cost of these resources, then this economic entity contributes to sustainable resource use at the benchmark level. The opportunity cost indicates how much return the benchmark alternative would create with the same set of resources. The return of the economic entity and the return of the 87 benchmark are then compared (Hahn et al., 2010). The SV accounts for how much value has been created as a result of the economic entity using the resources instead of the benchmark. It indicates how efficiently resources are being allocated between different economic entities (Figge and Hahn, 2004). In contradiction to what is usually used -analysis of the burden by internationalizing external environmental damages through complex pricing procedures. Furthermore, the investor's perspective about efficient allocation of resources is taken into account (Ang and van Passel, 2010) and not the business entity productivity perspective as indicated in the discussion paper of Kuosmanen and Kuosmanen (2009). A yearly value for an individual company can be derived on the basis of the following equation: Model Specification SVi - sustainable value added for company i, R - total numer of the resources, yi - gross value added for company i, y* - value added for the benchmark, xir - amount of the resource for company i, x*r - amount of the resource for the benchmark. General concept of sustainable value leaves space for certain flexibility in shaping factors and benchmark content. It allows to adjust model parameters according to the availability of the data. Sample for this research was chosen in order to analyze the prospects and advancement in sustainable value creation as well as mitigation of negative environmental impact in reference to value created by companies initially examined in the research of ADVANCE project (2006). Out of the 65 manufacturing companies from EU-15 countries it was possible to select 20 with a set of complete financial and environmental data. Results of ADVANCE project for the years 2001-2003 were compared with results obtained in this study for years 2004-2012. Data for the EU-15 countries were chosen for the benchmark following results of ADVANCE project. The dataset was obtained from Eurostat and European Environment Agency (EEA). Water usage data point was missing for 2013 year, it limited the period of analysis to 2012. It was recognized as less harmful to the research results than excluding water from the analysis. Water is an important environmental resource, though in the overall SV result it has a minor, even insignificant share. That is due to the fact that, following the original model, water is measured in cubic meters what makes the gross value added per cubic meter of water extremely low in comparison with other values per resource. From this observation general conclusion can be drawn about the impact of each resource for the overall value that is weighted by the size of emission or usage/generation of the resource. Therefore, additional weighting, as proposed in the recent work of Stakova (2015) might seem to be unnecessary effort. Out of initially chosen seven resources five were included in the analysis, namely: carbon dioxide (C02)-emissions, nitrogen oxide (NOx)-emissions, sulphur oxide (SOx)-emissions, waste generation and water use. Environmental data for companies were collected from Thompson Reuters Datastream ASSET4 database and financial data for companies were collected from Amadeus database, both accessed from Wroclaw University of Economics. Added value, understood as gross value added was calculated by adding depreciation to added value obtained from Amadeus database. Value added shows the contribution of particular company to both private and public income as well as its distribution among all stakeholders. It represents the economic value to be compared with GDP of EU-15 standing for the economic value of the benchmark. (1) 88 3 Results and Discussion Performance of 20 manufacturing companies was analyzed in the recent period of 2004-2012 in comparison with research results of ADVANCE (2006) for the period of 2001-2003. Two subsamples were created for a better clarity of results presentation. First one consists of companies creating positive sustainable value added and second one - of those destroying SV. It was an easy task to do due to the fact that companies were very consistent in their performance. Seven companies were permanently creating SV with some picks in 2006 and downturns in 2009 throughout the whole period. Remaining 13 were mainly destroying SV with some insignificant short-term improvements. Figure 1 shows that companies which were creating positive sustainable value in the ADVANCE research (2006) continue to do so in the following years. Majority of them improved their results. Only Henkel and Pirelli were creating relatively small values and demonstrating either minor progress (Henkel) or decrease in sustainable value over the period of analysis. Outstanding result was shown by Volkswagen which right after crisis year of 2009 outperformed all other companies (figure 1). Observing constant SV margin of approx. 30% (figure 3) it could be noticed that such results of Volkswagen resulted from very high increase in net sales (20-25% in years 2010-2012). Majority of companies that were destroying sustainable value reached lower results at the end of the period of analysis in reference to the beginning and to results from ADVANCE. Enel SPA data is not shown at figure 2 due to overly low value of almost -600 bilions of EUR in 2012. It was also performing poorly in 2003, being at the 45 place of the ranking. Nevertheless, companies in that subsample were successively creating less sustainable value than the benchmark, their SV margins (figure 5) and operating profit margins (EBIT margins) depicted at figure 6 stayed at almost constant level. Only ACEA, Italian company from utilities sector, had outstanding decrease in SV margin value (figure 5) and increase in EBIT margin in 2010. Financial results of that company show that in this year its sales decreased and it caused worsening of SV margin. Figure 1 Sustainable Value (Positive) of Companies in 2001-2012 (Billions of EUR) ■ DAIMLER AG ■VOLKSWAGEN AKTIENGESELLSCHAFT ■PEUGEOT SA ■ ASTRAZENECA PLC ■ HENKEL AG & CO. KGAA 0 -MP 4 f. T.T. 'IM,'1,1,1, • — HEIDELBERGER DRUCKMASCHINEN AG JCV>'' oV ■PIRELLI &C. S.P.A. Source: Own elaboration. 89 Figure 2 Sustainable Value (Negative) of Companies in 2001-2012 (Bn EUR) 50 o -50 -100 -150 -200 -250 -J5> JÖV rSP )( )( )( ) ■ACEA S.P.A. ■IMPERIAL CHEMICAL INDUSTRIES LIMITED ■BG GROUP PLC UPM-KYMMENEOYJ ■KEMIRA OYJ ■CENTRICA PLC ■BASF SE ■STORA ENSO OYJ ■EDISON S.P.A. ■FORTUM OYJ ■ENI S.P.A. ■BP P.L.C. Source: Own elaboration. Figure 3 Sustainable Value Margin for Companies in Years 2005-2012 (Positive SV) 70% 0% 2005 2006 2007 2008 2009 2010 2011 2012 ■ASTRAZENECA PLC ■HENKEL AG & CO. KGAA ■PIRELLI &C. S.P.A. ■VOLKSWAGEN AKTIENGESELLSCHAFT ■DAIMLER AG ■HEIDELBERGER DRUCKMASCHINEN AG ■PEUGEOT SA Source: Own elaboration. Figure 4 Operating Profit (EBIT) Margin for Companies in Years 2005-2012 (Positive SV) 50% 40% 30% 20% 10% 0% -10% -20% ASTRAZENECA PLC HENKEL AG & CO. KGAA PIRELLI &C. S.P.A. VOLKSWAGEN AKTIENGESELLSCHAFT DAIMLER AG ■HEIDELBERGER DRUCKMASCHINEN AG ■PEUGEOT SA Source: Own elaboration. Astrazeneca PLC is leader in both SV margin creation and operating profit margin creation (figure 3 and 4), it is British pharmaceutical company, it also had positive results in ADVANCE project study with 18th place (out of 65) in the ranking for 2003. Analysis of particular companies leads to the conclusion that sector specifics (high profitability, low GHGs emission, low polluting, etc.) could be the explanation of the outstanding performance. 90 Figure 5 Sustainable Value Margin for Companies in Years 2005-2012 (Negative SV) Source: Own elaboration. Figure 6 EBIT Margin for Companies in Years 2005-2012 (Negative SV) Source: Own elaboration. After analysis of absolute values of SV, it could be expected that Fortum OYJ (Finnish company operating in utilities industry, already poorly performing in ADVANCE, 56th place out of 65 in the ADVANCE ranking) will obtain very low negative results for SV margin. It ranged between -56% in 2006 and -2409% in 2009. It extended the scale of the figure 5 that is why it was excluded from it, together with two other very poorly performing companies - Edison SPA (Italian company operating in utilities industry, already with a very poor score in ADVANCE research, at 55th place) and Enel - SPA (Italy, utilities industry, around 60th place in ADVANCE). Analysis of sustainable value and relevant margins of this small sample showed couple of interdependencies. Choice of the benchmark determines the results among companies that operate in different sectors. Heavy industry, emitting a lot of GHGs and using a lot of resources will have bigger problems with reaching or outperforming the benchmark constructed on the basis of EU-15 results. It finds confirmation in results of ADVANCE project (2006) and study of Hahn et al. (2007). Later research of Hahn et al. (2014) changes the benchmark for sector EBIT values instead of EU-15 GDP. Positive results are more easily obtained by the companies realizing high profit margins and high increases in sales and profits. 4 Conclusions The purpose of this study was to discuss possibility of employing a different approach to analysis of the impact of companies' sustainability performance on their financial results and value creation. Sustainable value allows to show the individual value created by an entity in monetary terms using the opportunity costs approach. It shows it in reference to the benchmark. The concept is clear and methodology - coherent, but from the application perspective there are four important issues that require further studies and 91 consideration: 1. The choice of economic activity of companies (instead of gross value added one could use operating profit or cash flows), 2. The choice of resources and their units, 3. The choice of the return figure, 4. The choice of benchmarks (economic result at sector or national or EU level). Empirical study showed the development and prospects of companies initially tested in ADVANCE project (2006). In conclusion one could notice observable sector specific differences between the companies as well as positive tendencies among sustainable value creators and negative tendencies among SV destructors. This puts into the question the disclosure policies (sustainability policies) of those companies together with involvement in mitigation of their negative impact on the environment. The studied sample was too small to draw general conclusions, but it was an initial study that leaves space for more careful choice of the sample, benchmark and set of resources. Further studies will also explore the problem of alignment of sustainable value results with sustainability performance disclosure, the choice of companies being disclosing leaders might allow further interesting conclusions. References ADVANCE-project, (2006). Sustainable value of European industry: a value-based analysis of environmental performance of European manufacturing companies. Final report of the ADVANCE-project, full version. Ang, F., van Passel, S. (2010). The Sustainable Value approach: A clarifying and constructive comment. Ecological Economics, vol. 69, pp. 2303-2306. Aupperle, K. E., Carroll, A. B., Hatfield, 1 D. (1985). An empirical examination of the relationship between Corporate Social Responsibility and profitability. The Academy of Management Journal, vol. 28, no. 2. Figge, F., Hahn, T., (2004). Sustainable value added: measuring corporate contributions to sustainability beyond eco-efficiency. Ecological Economics, vol. 48 (2), pp. 173-187. Hahn, T., Figge, F., Barkemeyer, R., (2007). Sustainable value creation among companies in the manufacturing sector. International Journal of Environmental Technology and Management, vol. 7 (5/6), pp. 496-512. Figge, F., Hahn, T., (2009). Not measuring sustainable value at all: a response to Kuosmanen and Kuosmanen. Ecological Economics, vol. 69 (2), pp. 244-249. Hahn, T., Figge, F., Barkemeyer, R., Liesen, A. (2014). Operationalizing sustainability targets: An introduction to sustainable value approach. Global Compact International Yearbook. Godfrey, P. C, Hatch, N. W. (2007). Researching Corporate Social Responsibility: An agenda for the 21st Century. Journal of Business Ethics, vol. 70, no. 1, pp. 87-98. Guenster, N., Bauer, R., Derwall, J., Koedijk, K. (2010). The economic value of corporate eco-efficiency, European Financial Management, vol. 17, no. 4, pp. 679-704. Kuosmanen, T., Kuosmanen, N. (2009). How not to measure sustainable value (and how one might). Ecological Economics, vol. 69, pp. 235-243. Margolis, 1 D., Elfenbein, H. A., Walsh, 1 P. (2007). Does it pay to be good? An analysis and redirection of research on the relationship between corporate social and financial performance. Working Paper, Harvard University, Available: http://stakeholder.bu.edu/Docs/Walsh,%20Jim%20Does%20It%20Pay%20to%20Be%20 Good. pdf. Perman, R., Ma, Y., McGilvray, J., Common, M., (2003). Natural Resource and Environmental Economics, 3rd ed. Longman, New York. Saeidi S. P., Sofian S., Saeidi P., Saeidi S.P., Saaeidi S.A., (2015). How does corporate social responsibility contribute to firm financial performance? The mediating role of 92 competitive advantage, reputation, and customer satisfaction, Journal of Business Research, vol. 68, pp. 341-350. Shleifer, A. (2004). Does competition destroy ethical behaviour? NBER Working Paper no 10269. Sloriski, T., Daszyriska-Zygadlo, K., Zawadzki, B. (2016). The market value of CSR performance across sectors. Inzinerine Ekonomika-Engineering Economics, vol. 27(2), pp. 230-238. Strakova, 1, (2015). Sustainable value added as we do not know it. Verslas: Teorija ir Praktika / Business: Theory and Practice, vol. 16 (2), pp. 168-173. Wang, Y. and Berens, G. (2015). The Impact of Four Types of Corporate Social Performance on Reputation and Finance Performance. Journal of Business Ethics, vol. 131, pp. 337-359. 93 Systemic Risk Indicators in the Eurozone: An Empirical Evaluation Oleg Deev1, Martin Hodula2 1 Masaryk University Faculty of Economics and Administration, Department of Finance Lipová 41a, 602 00 Brno, Czech Republic E-mail: oleg@mail.muni.cz 2 VŠB-Technical University Ostrava Economic Faculty, Department of European Integration Sokolská třída 33, 701 21 Ostrava, Czech Republic E-mail: martinhodula@gmail.com Abstract: In this brief paper, we use combination of Markov-switching models and dynamic conditional correlation models to ex-post evaluate the performance of three widely used systemic risk measures (SRISK, CISS and term-spread) based on their ability to predict financial turmoil. We first compare systemic risk measures based on their dynamic correlations. Second, we identify three regimes for each indicator and evaluate them based on their ability to capture crisis information. We found that in practice, the correlations between studied systemic risk measures are indeed high and indicators are successful in capturing regimes of high financial stress. We have however, identified a few periods when the indicators are not overlapping, especially in pre- and post-crisis period. Keywords: systemic risk measures, Markov-switching models, DCC-GARCH models JEL codes: C34, C58, E44, G01 1 Introduction After the financial crisis economists started to recognize that adverse shocks to the financial sector can have a significant impact on the real economy. This ability of financial system stress (or financial instability) to trigger sharp macroeconomic downturns has fostered extensive research on systemic risk, either on its definition, measurement, or regulation. The identification of Systematically Important Financial Institutions (SIFIs) -the financial institutions that contribute the most to the overall risk of the financial system - was initially the main interest among practitioners and researches. As SIFIs pose a major threat to the system, regulators and policy makers from around the world have called for tighter supervision, extra capital requirements, and liquidity buffers for SIFIs under the microprudentional regulatory framework. Currently, as there is a growing consensus among central banks about the financial stability objective and its definition, the policy interest has shifted from microprudential to macroprudential policies. The job of these policies is to ensure that the financial system does not become extremely vulnerable or that some shocks would not ultimately cause financial instability in the form of a crisis. The two main tasks of macroprudential policy - to prevent systemic risk and, if prevention fails, to mitigate the impacts when it materializes - are given by the existence of two phases of development of systemic risk: (i) accumulation phase, which can be explained as the build-up of systemic risk in the economy, and (ii) materialization phase, when economic agents revise their expectations and radical change in their behavior will occur. Banks will revise their evaluation of the credit, market and liquidity risk accumulated in their balance sheets, increase credit margins or credits spreads, and tighten lending conditions (more on this topic can be found in Fraitand Komárkova 2011). In their pursuit for financial stability, the monetary authorities need to identify and correctly assess the evolution of systemic risk over the financial cycle. Many systemic risk indicators have already been proposed in the aftermath of the 2007-2009 financial crisis, but the question remains of to what extent are they able to trace the development of the systemic stress? Hence, the goal of this brief paper is to propose a comprehensive 94 comparison of some of the most used systemic risk indicators for analysis in the Eurozone area. To our knowledge, this research is the first attempt to derive and compare major systemic risk measures within the common Eurozone framework. We hope that our analysis allows us to uncover the theoretical and empirical link between systemic risk measures. Such research is much needed as there is no consensus on the topic of which of the indicator's type is the most appropriate to uncover the accumulation of systemic risk within the financial system. It should be noted that the goal of this research is not to find the best systemic risk indicator as the "one-size-fits-all" approach. Overview and Systemic Risk Measures Characteristics There are a few approaches to measure systemic risk and to identify and assess the evolution of systemic risk over the financial cycle. Because we are interested in the former, we will consider only those indicators that can be used not only to measure systemic risk, but also to assess the systemic risk build-up (accumulation phase) in the financial system. We are also interested only in those indicators that cover the time dimension (this criterion is given by the nature of our analysis as we compare indicators in time). We also do not cover such indicators that are not available for Eurozone or do not have sufficiently long time series or frequency. Figure 1 Systemic Risk Indicators 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 - CISS---SRISK - SPREAD Source: VLAB, ECB. Note: Time series were normalized to scale 0 to 1. First, we consider SRISK indicator as a representative of institution-specific market-based risk measure. The indicator is designed to capture an individual bank's contribution to economy-wide systemic risk. SRISK (proposed by Brownlees and Engle, 2015) measures the capital shortfall of the financial institution, if there is another crisis in the financial system. Simply put, it estimates the amount of capital needed by a financial firm in the event of a crisis. The SRISK equation is calculated as follows: SRISK= k.DEB T- (1 - k) EQ £7/71(1 - LRME§ (i) where k is the capital requirement and LRMES is the long-run marginal expected shortfall. By multiplying the components in (1), we obtain the total differential of SRISK: ASRISK= k.dDEBT- (1 - k).(1 - LRME§dEQUIT¥- (1 - k) EQUITHLRME. (2) The change in SRISK can be hence decomposed into three parts: 95 - k{DEBT) = k.dDEBri - the contribution of firm's debt to SRISK: as the company takes on more debt, it increases its leverage and the contribution to the systemic risk will be positive; - A(EQUITJ=-(l-k).(l-LRMES)dEQUIT can be described as firm's equity position: as firm's market capitalization declines, the SRISK contribution rises; - A(RISK)=(l-k)EQUIT¥lLRME> shows a potential increase in firm's risk attributes, such as increased correlation or volatility. The SRISK measure is mostly used to identify SIFIs at micro-level (Banulescu and Dumitrescu 2015; Benoit et al. 2015), but there already exist a growing number of studies using this indicator as a proxy for aggregate systemic risk at macro-level in the financial system (Engle et al. 2015; Grinderslev and Kristiansen 2016; Langfield and Pagano 2015). Second, we analyze the term spread (the slope of the yield curve) calculated as the difference between 10-year swap rate and 3-month EURIBOR. This can be viewed as a measure that takes into account liquidity and credit conditions in financial markets. Gerlach (2009) considers interest rates a useful indicator of systemic risk as they are continuously available with no delay and contain information about markets participants' views of a range of different risks. Wheelock and Wohar (2009) provide a survey of empirical evidence on how the term spread predicts output growth and recessions up to one year in advance. If we consider the systemic risk as a main driver of recessions, the term spread acts as a good measure of systemic risk. Third, we consider the official ECB financial stability indicator - Composite Indicator of Systemic Stress (CISS). The index is the composite indicator which aims to represent the level of systemic stress in the Eurozone's financial system based on 15 mainly market-based financial stress measures from the financial intermediaries sector, money markets, equity markets, bond markets and foreign exchange markets. All components carry equal weightings, thereby allowing the stress index to place relatively more weight on situations in which stress prevails simultaneously in several market segments. The stress index falls within 0 and 1, where higher stress levels are closer to one. There are of course many indicators of systemic risk that are not covered here and will be incorporated in future analysis (such as indicators based on Value at Risk, non-performing loans, CDS spreads etc.). 2 Methodology and Data Since the main goal of any financial risk indicator (or in this case systemic risk) is to identify such level of risk that is able to disrupt the market and endanger economy as a whole, it is possible to identify two basic approaches on how to do so. First approach is to set some benchmark level of financial stress and classify its level as significant in case it exceeds one unit of variance (Illing and Liu 2006). Such approach must, however, explicitly assume that the indicator follows normal distribution, which assumption in many cases cannot be fulfilled. The second approach separates periods of high or extreme financial stress from periods with only moderate or low levels of stress based on the assumption that the time series properties of particular indicators are state-dependent. This approach may seem appropriate, as it is assumed for financial stress to cluster around local attractor levels across different regimes, thereby, displaying some intra-regime persistence. Hence, the measure acts stochastically, unpredictably. Therefore, it is useful to first analyze the distribution of particular indicators. It is visible from Figure 2 that the distribution of indicators cannot be described as Gaussian. The distribution is heavily skewed towards its right (CISS) or left (SPREAD) tail or even multimodal (SRISK). This means that the empirical density function should be represented as a mixture of distributions, each characterizing a separate regime. 96 Figure 2 Histogram Display for Analyzed Indicators OSS SRISK SPREAD Note: Histograms calculated for analyzed indicators are based on monthly averages of monthly data from July 2000 to June 2015. Smoothed histogram is based on Epanechnikov kernel. In order to model this regime-dependence, we estimate several variations of Markov-Switching models with up to three states (st): Rt = a{st) + P{st)Rt_x +a{st)ut, for s, = {0,1,2}, (3) where Rt = [CISS,SRISK,SPREAD], all coefficients (CC,P, Pl\0 Pon An All Po\2 Pl\2 P2\2 1- Po\o Pw ' Po\o ~ A|0 1- An Pill ' Po\\ ~ Pl\[ 1- Po\2 Pl\2 ' Po\2 ~ Pl\2 (4) where the third row conditional probabilities have been replaced by adding-up constrains for probabilities. For a model with three regimes only six out of nine transitions probabilities can be estimated independently). As seen from Table 1, our preferred model specification is MS(3)-AR(1), that is a Markov-Switching model with three regimes. Some summary statistics suggest superiority of the model with regimes in terms of log-likelihood values and information criteria. More importantly, all model specifications survived the residuals misspecification tests of non-normality and autocorrelation. Table 1 Different Markov-Switching Model Specifications Indicator Model Log-likelihood AIC Normality (D-W) Autocorrelation CISS MS(3)-AR(1) 333.889 -3.601 1.853 0.227 MS(2)-AR(1) 318.279 -3.503 1.984 0.714 SRISK MS(3)-AR(1) 290.025 -3.954 1.949 0.421 MS(2)-AR(1) 262.356 -3.719 1.888 0.327 SPREAD MS(3)-AR(1) 305.651 -3.132 1.981 0.351 MS(2)-AR(1) 294.561 -3.056 1.915 0.153 Source: Own estimation Notes: MS(s) stands for a Markov-Switching models with 2 or 3 regimes, AR(p) denotes an autoregressive model of order p. Estimations are based on monthly data from July 2000 to June 2015. We have generated 50 sets of random starts with 20 iteration refinements. 97 3 Results and Discussion Before we proceed to Markov-Switching models results, it is useful to look at time-varying correlations between our modelled indicators. We estimate dynamic conditional correlation GARCH (1,1) models whose results are plotted in Figure 3. First, it is evident that our analyzed indicators are highly correlated during the crisis period from 2008 to 2013. Second, estimated correlations are somewhat mixed during the pre- and after-crisis periods. In terms of identifying risk materialization phase, all of the indices seem to be successful, but they are posing different results in the risk-accumulation phase from 2005 to 2007. Figure 3 DCC-GARCH(1,1) Model Estimations -0.25 --0.50 --0.75 - -1 i !\* V if V y -i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i— 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 CISS_SPREAD CISS_SRISK SRISK_SPREAD Source: Own estimations Notes: Time-varying correlations were computed as dynamic conditional correlation model estimated by log likelihood for integrated process since the data in levels have a unit root. Mathematical propositions of the model are described in Engle (2002). Used calculation code is available upon request. Results of Markov-switching model estimations are shown in Figure 4. Regime 1 represents the case when financial stress is relatively low and we can identify it as "normal" times. Regime 3 identifies structural breaks, shocks and levels of extremely high financial stress, such is only visible during rare events (financial crisis, 9/11 attack, oil shocks). Regime 2 is somewhat problematic for interpretation. Judging from our estimations, it can be interpreted as intermediate level of systemic stress as it occurs in periods before and after Regime 3 events. Fostel and Geneakoplos (2008) speak of "anxious economy" and this truly might be the case. The central question is however different - does Regime 2 capture the accumulation phase of systemic risk? The answer varies among different indicators and it is not as straightforward as needed. In terms of regime probabilities, the estimations of each indicator nicely show their differences: First, the CISS indicator as a representative of aggregate systemic risk indexes successfully identifies rare events under Regime 3, such as terrorist attacks 9/11 in the US, after which even European markets had exhibit increased volatility and sold-outs, or Lehman Brothers bankruptcy and Greek crisis. Regime 2 is displayed before and during the dot-com bubble burst and subsequent bear market and also shortly before and after the financial and debt crises. Regime 1 displays the great moderation period from 2002 to 2006 and the after-crisis period of economic recovery after 2013. Overall, CISS index appears successful in capturing materialization phase of systemic risk. As for the build-up phase of systemic risk, we would expect to see Regime 2 active in 2006, when the 98 financial risk was truly accumulating and many potential non-performing loans contracts were made. The CISS index seems to fail to comprehensively display the build-up phase of systemic risk (which is in line with some recent research, such as Giglio et al. 2015). Second, SRISK (financial sector equity return volatility measure) is more focused on particular failures from micro viewpoint under Regime 3. It coincides strongly with CISS indicator in this Regime. They differ, however, in the interpretation of Regime 2. SRISK places Regime 2 especially under the increased systemic volatility after dot-com bubble burst in 2002 and more importantly at the beginning of 2006, which is now identified as a direct pre-crisis period and it is characterized by a strong build-up of systemic risk and credit risk in the mortgage market in the US. It seems that financial sector equity return volatility variables are the most informative individual predictors of downside macroeconomic risk. These findings are consistent with the view of Schwert (1989). His empirical analysis highlights the co-movement among aggregate market volatility, financial crises, and macroeconomic activity. Third, we analyzed the term spread measure. It captures liquidity and default risk, but also the fact that during periods of turmoil investors lend against treasury bills (the best form of collateral), measuring the "flight to quality" effect (Rodrfguez-Moreno and Pena 2011). These stylized facts are nicely reflected in Regime estimates. Regime 3 marks the start and end of a credit cycle in the period from 2001 to 2005. First, it identifies the end of 2001, when ECB started to slowly downgrade its interest rates to boost the economy after the dot-com bubble. Second, it highlights the end of 2005, when the expansionary monetary policy ended and the ECB raised its official interest rates. Regimes 1 and 2 should then capture the periods of low and moderately high liquidity and default risk. It does so nicely, as we know that the pre-crisis period from 2006 to 2007 was characterized by seemingly low liquidity and default risk. The increased risk premium has only started to rise after the Lehman's bankruptcy. Figure 4 Smoothed Regimes Probabilities from Markov-Switching Model Estimations OSSJ CIS5_2 ; ^^^^^^^^^^^^^^^^^^^^^^^^^^mtpM»TT>.r..TT.; ilu i..Tr;i■.■.1111■ f ■1111..r'.'T"^^^^^^Tr^^^^^^^^^^ uu ^l.t,.^li,t.,l[il^,li[li.t,l.p,.tli.li.,t.,l|.,,T,llr,^ DaHCO]OiKKII0in«1tUl3U1! 00 Di 02 0} 0* 05 C6 07 C« » 10 11 i? ij W 15 00 01 « 03 04 05 06 07 H » 10 11 12 13 14 IS Source: Own estimations 99 4 Conclusions In this brief paper, we quantitatively examined a collection of systemic risk measures that are proposed and used in the literature. We argue that systemic risk measures should be able to capture not only the materialization phase of systemic risk, but also the build-up (accumulation) phase. We analyzed three widely used systemic risk indicators: CISS, SRISK and term spread. We assume the empirical density functions of analyzed indicators can be represented as a mixture of distributions, each characterizing a separate regime. In other words, we assume the systemic risk indicators to be regime-dependent. In order to model this regime-dependence, we estimate several variants of Markov-Switching models with up to three states. So far, we can present here a few stylized facts. First, systemic risk indicators analyzed here have an especially strong association with the materialization phase of systemic risk in the financial system, but mostly fail to capture the build-up phase. Second, indicators based on financial sector equity volatility (such as SRISK) were particularly informative about future real activity and seem to identify successfully the accumulation phases that might eventually lead to financial instability. In our future research we aim to incorporate more indicators generally used in the literature (such as indicators based on Value at Risk, non-performing loans, CDS spreads etc.). Study results might be improved by prolonging time series to include more business cycles and including a dynamic regression Markov-switching model proposed by Doornik (2011) instead of described fully fledged model. Acknowledgments Authors acknowledge the support of the project GAČR 16-13784S "Financial sector policy and institutions: Current challenges in balancing financial development and Stability" and Masaryk University internal grant MUNI/A/1025/2015. We are also thankful to V-Lab Institute for sharing their data. References Banulescu, G. D., Dumitrescu, E. I. (2015). Which are the SIFIs? A Component Expected Shortfall approach to systemic risk. Journal of Banking & Finance, vol. 50, pp. 575-588. Benoit, S., Colliard, J., Hurlin, Ch., Perignon, Ch. (2015). Where the Risks Lie: A Survey on Systemic Risk. HEC Paris Research Paper, No. 1088. Brownless, T. C, Engle, R. F. (2015). SRISK: A Conditional Capital Shortfall Measure of Systemic Risk. Working Paper. Retrieved from: http://dx.doi.org/10.2139/ssrn.1611229. Doornik, 1 A. (2011). Markov-Switching Models, University of Oxford and OxMetrics Technologies Ltd., Mimeo. Engle, R. (2002). Dynamic Conditional Correlation. Journal of Business & Economic Statistics, vol. 20(3), pp. 339-350. Engle, R. F., Jondeau, E., Rockinger, M. (2015). Systemic Risk in Europe. Review of Finance, vol. 19(1), pp. 145-190. Fostel, A., Geneakoplos, J. (2008). Leverage Cycles and the Anxious Economy. American Economic Review, Vol. 98(4), pp. 1211-1244. Frait, 1, Komárkova, Z. (2011) Financial stability, systemic risk and macroprudential policy. In: Financial stability report 2010/2011. Praha: Czech National Bank, pp. 96-111. Gerlach, S. (2009). Defining and Measuring Systemic Risk. Directorate General for Internal Policies Note, No. 2009/040. Giglio, S., Kelly, B. T., Pruitt, S. (2015). Systemic risk and the macroeconomy: an empirical evaluation. NBER Working Paper Series, No. 20963. 100 Grinderslev, O. J., Kristiansen, K. L. (2016). Systemic risk in Danish Banks: implementing SRISK in Danish context. Danmarks Nationalbank Working Paper, No. 105. Illing, M., Liu, Y. (2006). Measuring Financial Stress in Developed Country: An Application to Canada. Journal of Financial Stability, vol. 2(3), pp. 243-265. Langfield, S., Pagano, M. (2015). Bank bias in Europe: effects on systemic risk and growth. ECB Working Paper Series, No. 1797. Rodrfguez-Moreno, M., Pena, J. I. (2011). Systemic risk measures: the simpler the better? BIS Papers, No 60. Schwert, G. W. (1989). Business cycles, financial crises, and stock volatility. In: Carnegie-Rochester Conference Series on Public Policy, vol. 31, pp. 83-125. Wheelock, D.C., Wohar, M. (2009). Can the Term Spread Predict Output Growth and Recessions? A Survey of the Literature. Federal Reserve Bank of St. Louis Review, No. 91. 101 The Impact of Contingent Convertible Bond Issuance on Bank Credit Risk Oleg Deev1, Vlad Morosan2 1 Masaryk University Faculty of Economics and Administration Department of Finance Lipová 41a, 602 00 Brno, Czech Republic E-mail: oleg@mail.muni.cz 2 Masaryk University Faculty of Economics and Administration Department of Finance Lipová 41a, 602 00 Brno, Czech Republic E-mail: 440132@mail.muni.cz Abstract: Contingent convertible bonds are designed to provide additional capital to banks in times of distress and discourage the risk-taking incentives of the stockholders and, hence, decrease bank credit risk. In this paper, we study bank CDS spreads as a proxy of credit risk during the periods around the announcement of contingent convertible bond issuance. We analyze whether investors see these bonds as signs of possible bank distress or the stabilizing mechanism decreasing the probability of bank default. We use event study methodology where abnormal CDS spreads are identified based on constant mean return model and basic market model. Our data sample consists of 41 banks with 109 current unique issues taken from Bloomberg. Our results indicate that CDS spreads show a systemic reaction to the announcement of contingent convertible instruments and are economically significant for the bond holders that value the decreased default probability and reduced risk incentives of the issuing institution. Keywords: contingent convertible bonds, event study, CDS prices, bank credit risk JEL codes: G21, G14 1 Introduction The financial crisis of 2007-09 revealed that financial institutions had insufficient adequate loss absorbing capital and very little insight on the credit and liquidity risks related to subprime mortgages and other risky investments. As a result, regulators and governments had to intervene and bail out the global systemically important banks by nationalizing, using liquidity insurance schemes and injecting public sector funds i.e. taxpayers' money in the statutory capital of the banks, in order to address the risk of an even bigger systemic crisis. This led to a moral hazard problem where banks have no incentive to prioritize and empower risk management, since taxpayers' money is used as insurance (Calomiris & Herring 2011). As the response to the financial crisis, the Basel Committee on Banking Supervision issued a set of proposals (Basel III) with the main priority to strengthen the quality, consistency and transparency of the regulatory capital base (BIS 2009). Under these new requirements, only truly loss absorbing capital will be recognized as regulatory capital consisting of common equity, retained earnings and contingent convertibles - a new asset class in the family of hybrid financial instruments (BIS 2010). According to Hilsher and Raviv (2014), the addition of contingent convertibles to regulatory capital represents "one of the most prominent suggested solutions for the shortfall of capital in bad times". Contingent convertible bonds (CoCos) are hybrid capital securities that absorb losses on a going concern basis when the capital of the issuing bank falls below a predetermined level (Avdjiev et al. 2015). Based on the contractual agreement, the loss absorption mechanism can either take the form of an automatic mandatory conversion into equity or take place through a full or partial principal write-down of the CoCos' face value (Spiegeleer et al. 2014). This mechanism of swapping debt into equity or "bail-in" is 102 considered potentially valuable since it is triggered on a going concern basis, unlike other hybrid instruments, such as Tier 1 bonds, which were outlawed in Basel III due to their failure during the recent near collapse of the financial system (Hilsher & Raviv 2014). Central bankers believe that the issuance of CoCo bonds reduces the default probability of the issuing bank by providing an additional buffer that can absorb losses meanwhile discouraging risk taking incentives of the stockholders in times of crisis. In this context, CoCos represent an attractive mean to raise the necessary capital in order to meet the requirements due to lower cost of capital as coupon payments are tax deductible and dividends are not. On the other hand, from the investors' standpoint, CoCo bonds offer a higher yield compared to other bonds. Therefore, it is no surprise that according to Bloomberg the market of contingent convertibles is growing with nearly USD 77 billion issued in 2015 alone, and that in the following years this market will have an even stronger growth. The issuance of contingent convertible bonds has a stabilizing effect on the issuing institution by providing more loss absorbing capital and discouraging risk taking incentives of the stockholders in times of crisis (Flannery 2009). Consequently, if investors assume that the bank can decrease its probability of default, then the issuance of CoCos should have a positive announcement effect on the market based indicators, such as CDS spreads. On the other hand, opposing effects are also being considered. A prevailing positive effect of the CoCo issuance resulting in a decreasing CDS spreads around the announcement date is possible, since bond investors are facing the downside risk and do not gain from the firm's additional risk taking. At the same time, considering that CDS spreads are negatively correlated with the value of the firm, the decrease in share prices after the announcement could imply also a higher default probability, and so, it is possible that CDS spreads to increase. The purpose of this paper is to investigate the effects associated with the issuance of contingent convertible capital instruments on banks' financial stability. In order to test this impact, we apply an event study analysis on a sample of 41 banks with 109 unique issues of contingent convertible bonds. As a proxy of financial soundness of bank, we use a market-based indicator - credit default swap spreads. The use of market-based indicator was determined by their general availability at high frequency and better risk-signaling qualities compared to accounting-based indicators (Cihak 2007). Our research adds to the literature on contingent convertible instruments by analyzing market's perception of the announcement of CoCos issued in countries with different levels of development, while in the previous studies the focus has been put on issuers from a single country (Schmidt & Azarmi 2014) or from a group of countries with similar characteristics (Avdjiev et al. 2015, Rudlinger 2015, Ammann et al. 2015). We provide a broader view of the impact of the event on bank stability and perceived default risk and compare the results using two different models of abnormal CDS spreads calculation (market model and constant mean return model). Related Literature Flannery (2002) introduces the theoretical concept of a new type of bond for large financial institutions called "Reverse Convertible Debenture". Banks' balance sheet should include an instrument that would initially take the form of a subordinated debt but that would automatically convert into common equity if the market capital ratio of the bank would fall under a predetermined level. This mandatory conversion mechanism would facilitate the deleveraging of the bank in times of distress with little effects on risk taking incentives. Only after the financial crisis of 2007-09 this instrument was considered to reduce the systemic risk of large financial institutions. BIS (2011) provides the regulatory framework for this type of capital consisting only of fully absorbing instruments in times of distress. The literature on CoCos mainly concentrates on the optimal design and pricing of these financial instruments, while there is little empirical evidence on the effects of contingent convertible capital on bank risk-taking and investor behavior mainly due to the fact that CoCos are still a new asset class with the first placements being made in 2009. 103 One of first attempts to estimate the effect of issuance of contingent convertible bonds on bank value and perceived default risk was done by the Schmidt and Azarmi (2014) for the first issue of CoCos made by Lloyds Banking Group in 2009. This first evidence suggests that CoCos can have a negative effect on a bank's creditworthiness and firm value. Avdjiev et al. (2015) find that the issuance of CoCos has a negative impact on the issuer's CDS spreads. The sample used in this study was based on the CoCos using mandatory conversion to equity issued by banks from all advanced economies with the exception of the euro area periphery. Table 1 presents a summary of existing studies that examine the announcement effect of contingent convertible instruments on market-based indicators along with the applied methodology and scope of the event study. Table 1 Comparison of Studies on CoCo Announcement Effect Study Scope Period Methodology Findings Schmidt and Azarmi (2014) Lloyds Banking Group (UK) 2009 MM Significant positive effect Ammann, Blickle and 34 large banks 2009 - MM Significant negative Ehmann (2015) (mostly European) 2014 effect Rüdlinger (2015) 12 large banks (Eurozone) 2009 -2014 MM, CMRM No significant effect Avdjiev et al. (2015) Advanced economies (except PIGS) 2009 -2015 MM Significant negative effect Source: Own elaboration Note: MM - market model, CMRM - constant mean return model 2 Methodology and Data In order to examine the effect associated with the issuance of contingent convertible instruments on banks' financial stability, we conduct an event study. Given rationality in the marketplace, the effect of an event is instantly captured by the market and reflected in CDS prices. In this context, we follow the event study methodology of Campbell et al. (1997) and Mackinlay (1997) to determine the announcement effect of the CoCo bonds issuance on credit default swap (CDS) spreads. An event study analysis has the following steps: Defining the event. Event day is represented by the announcement day that corresponds with the official disclosure of new information on the issuance of CoCo bond. The day after the event day is also imperative for analysis since it captures the impact on market prices, if the information was released after the market was closed. The period preceding the event day should also be analysed, because market participants might acquire some information before the public announcement which can affect the prices of securities of the affected firm. According to Avdjiev et al. (2015), banks do not publicly announce the issuance of CoCos. Instead, the intention is revealed privately to the groups of potential buyers, usually over the course of 2 weeks before the date of issuance. Hence, the event date cannot be easily associated to a specific date that would represent the intention of the bank to issue contingent convertible bonds. Considering that the information is being spread prior to the date of issuance, the information about the upcoming issuance could be incorporated in CDS spreads before the actual issuance. For this reason, as suggested by Ball and Torous (1988), when the exact event day is unclear the maximum likelihood estimation length should be applied. The common practice is to consider 41 days, 20 days before the event (t-20 to t-1), the event day (t), and 20 days after the event (t+1 to t+20). Firm selection criteria. Bloomberg provides information on 229 unique issues of contingent convertible bonds for 132 banks over the period from January 1, 2009 to December 31, 2015. However, due to the lack of actively traded CDS (we consider only the banks with corresponding USD denominated daily 5-year CDS as they are the most liquid credit derivative instruments present on the credit market), the initial sample is reduced to 41 banks with 109 unique issues. 104 Measurement of normal and abnormal CDS spreads. The normal CDS spread is defined as the return that would have occurred, if the event did not take place. Abnormal spreads or returns eiit are estimated by subtracting the normal return from the actual ex post return of the security in the event period. Considering a security i at the event day t the abnormal return is defines as: e£,t = Rt,t-E[R<>t|Xt], (1) where Rix is the realized return, E(Rix) is the return of the security and Xt is the conditioning information for the performance model. Because realized returns Ra are observable from the market, realized return has to be modelled using the conditioning informationXt from one of the two benchmark models. Through the constant mean return model (CMRM) Xt is a constant equal to the mean return. In this case the normal return of the security i is equal to a mean return of the firm in the estimation window. The market model (MM) assumes a stable linear relationship between the returns of the assessed security and the market returns. The market model normal returns are determined using the geographical iTraxx CDS bank indices (European, American or Asian). CMRM assumes a realized return Ra of a security i in the period t equal to the mean return m and an error term <^t as follows: Rt,t = \k + to, E[ <;,t] = 0, Var[ <;,t] = S2. (2) where Rix is the -tth element of Rt for period t and to is the error term for security with an expectation of zero and a constant variance 62^. The parameters of the constant mean return model for the security i are m and h\. The market model assumes that the asset returns have a joint normal distribution and has the following form: R.« = + p,Rmt + £«, E[ = 0, Var[ £,t] = 52£, (3) where Rix denotes the return of the securityi, Rmt is the market return in the period t, and sit is error term for security with a zero mean and a constant variance 8\.. Therefore, the parameters in the MM area^, (3^ and 8\.. The advantage of using the MM over CMRM is that it reduces the variation of abnormal return, as it removes the portion of return attributed to the variation in the market's return. Consequently, using the MM can give us a better perspective on the effects of the event. In order to analyze longer-term inferences of the abnormal returns for the event of interest daily abnormal returns of each security i in the event window are aggregated to cumulative abnormal returns (CAR). The cumulative abnormal return CARi(T) of security i in the event window from xt to t2 , where Tt < rt < t2 < T2, is defined as: CARiC^T,) = y'fr, (4) where y is a (L2 x l) vector having ones in the positions from xt to t2 and zeros elsewhere and represents the vector of abnormal returns with the dimensions of L2 x land which is based either on the constant mean return model where \l = %\, or the market model where £j = ?t. Since we cannot conclude about the results based on the tests of one singular event, the abnormal returns (AR) and the cumulative abnormal returns (CAR) are aggregated. The aggregation of AR and CAR is performed along two dimensions - through time, by aggregating the data in the event window by date from (t-20) to (t+ 20); and across securities, by averaging the abnormal returns across all securities included in the sample. For this aggregation we use the simplifying assumption of no correlation between the abnormal returns of different securities. Therefore, the aggregation of individual abnormal returns can be performed using the results from the constant mean return 105 model (CMRM) where ^ = or the market model (MM) where£j= e\ for each event periodt = Tx + 1, ...,T2 as follows: (5) Cumulative average abnormal returns CAR(TlT2) for the period from tx to t2across all the securities in the event window is computed as follows: CAR ) - ^E?=iCARw (Tl,T2), (6) Estimation procedure. Model parameters are assessed based on the data from the estimation window. Estimation window is the period of time that is not included in the event window, preferably prior to the event that is used for estimating parameters of the normal performance, but that are not influenced by the information from the event. The parameters for the benchmark model used for the computation of the expected CDS spreds are usually computed based on the estimation window of 80 trading days prior to the event window (t-100 to t-21). Data testing. In order to conclude on the significance of the results, parametric t-tests are used. A t-test is a parametric hypothesis test that is based on the assumption that returns are independent and identically distributed, i.e., knowing the abnormal return ARi(T) of the securityi for the dayx, the standardized abnormal returns on the day t of the individual security i: Jar(t) = SAR(T) = ^-£1 (7) where S;(t) denotes the standard deviation of the abnormal return on the day t, being equal to the square root of the (T,x)th element of the covariance matrix v^. Under the null hypothesis, the distribution of JARi(t) is Student-t with L2 - 2 degrees of freedom. To test the null hypothesis of no effect of the announcement on the individual returns of securityi, standardized cumulative abnormal return can be used as follows: jcar(t) = SCAR _ car(t1,t^ 5(tl,T2) (8) - ( 1 \l/2 where 6(Tl>T2)is measured based on 6(Tl>T2) = [r^'Zi=182(T1,T2)j and is calculated with the consideration that under the null hypothesis SCARi(TliT2) is Student-t with L2 - 2 degrees of freedom. Figure 1 CDS Spreads around C0C0 Bond Issuance 1: CMRM—Average abnormal returns (AAR) 2: MM— Average abnormal returns (AAR) 0,01 0,015 -0,01 Day relative to event day -0,01 Day relative to event day 3: CMRM— Cumulative average abnormal returns (CAAR) 4: MM— Cumulative average abnormal returns (CAAR) 106 0,01 0,015 -0,04 Day relative to event day Day relative to event day Source: Own elaboration based on Bloomberg data 3 Results and Discussion Our event study results show that the overall impact of a CoCos issue on the average abnormal CDS spreads of the issuing banks is significantly negative. However, the significance of results obtained by using the constant mean return model (CMRM) and the market model (MM) are different. This can be easily explained by the fact that using the market model as the normal return model leads to a reduced abnormal return variance by removing the share of the return associated to the variation in the market return. This results follow the conclusion of Campbell et al. (1997) that the market model is more precise in estimating the event effects and depends on the R2 of the market-model regression. A higher R2 leads to a greater variance reduction of the abnormal CDS spreds and, in consequence, to the more precise inference. Table 2 Abnormal CDS Spreads around the Announcement Date CMRM MM 1 Day AARs t-stat Sig. CAARs t-stat Sig. AARs t-stat Sig. CAARs t-stat Sig. t-20 0.574% 0.743 0.574% 0.743 0 984% 1.273 0.984% 1.273 t-19 -0.280% -0.363 0.294% 0.380 0 007% 0.009 0.991% 1.282 t-18 -0.358% -0.463 -0.064% -0.083 -0 100% -0.129 0.891% 1.153 t-17 -0.713% -0.922 -0.776% -1.005 -0 334% -0.432 0.557% 0.721 t-16 -0.464% -0.601 -1.241% -1.606 -0 238% -0.308 0.319% 0.413 t-15 -0.319% -0.412 -1.525% -1.973 ** 0 010% 0.013 0.286% 0.370 t-14 -0.263% -0.341 -1.788% -2.313 ** 0 085% 0.111 0.371% 0.480 t-13 -0.270% -0.349 -2.058% -2.663 *** -0 523% -0.676 -0.149% -0.193 t-12 0.567% 0.733 -1.518% -1.965 ** 0 495% 0.640 0.390% 0.505 t-11 -0.180% -0.233 -1.699% -2.198 ** -0 023% -0.029 0.368% 0.476 t-10 0.107% 0.139 -1.577% -2.041 ** 0 043% 0.055 0.372% 0.481 t-9 -0.498% -0.645 -2.075% -2.685 *** -0 198% -0.256 0.173% 0.224 t-8 -0.886% -1.147 -2.972% -3.846 *** -0 616% -0.798 -0.403% -0.521 t-7 0.598% 0.774 -2.374% -3.072 *** 0 535% 0.693 0.132% 0.171 t-6 -0.163% -0.211 -2.537% -3.282 *** -0 125% -0.162 0.007% 0.010 t-5 -0.170% -0.220 -2.707% -3.502 *** -0 171% -0.222 -0.164% -0.212 t-4 -0.256% -0.331 -2.962% -3.833 *** -0 269% -0.348 -0.433% -0.561 t-3 0.113% 0.146 -2.850% -3.687 *** -0 083% -0.107 -0.516% -0.668 t-2 0.074% 0.096 -2.776% -3.592 *** -0 074% -0.096 -0.590% -0.763 t-1 0.035% 0.046 -2.741% -3.546 *** 0 002% 0.002 -0.588% -0.761 to -0.357% -0.461 -3.097% -4.008 *** -0.213% -0.275 -0.801% -1.036 t+1 -0.065% -0.085 -3.163% -4.093 *** -0 133% -0.173 -0.935% -1.210 t+2 -0.082% -0.106 -3.245% -4.199 *** -0 085% -0.110 -1.020% -1.320 t+3 -0.384% -0.498 -3.629% -4.696 *** -0 221% -0.286 -1.238% -1.602 t+4 0.275% 0.356 -3.354% -4.340 *** 0 291% 0.377 -0.947% -1.225 t+5 0.086% 0.111 -3.268% -4.229 *** -0 292% -0.377 -1.235% -1.598 t+6 0.120% 0.155 -2.952% -3.820 *** 0 023% 0.029 -1.026% -1.327 t+7 -0.694% -0.898 -3.646% -4.718 *** -0 380% -0.492 -1.408% -1.822 * t+8 0.867% 1.122 -2.779% -3.596 *** 0 627% 0.811 -0.781% -1.011 t+9 -0.364% -0.471 -3.342% -4.325 *** -0 009% -0.011 -0.890% -1.152 t+10 0.072% 0.093 -3.156% -4.084 *** 0 129% 0.167 -0.719% -0.931 t+11 0.061% 0.079 -3.096% -4.006 *** 0 072% 0.093 -0.738% -0.955 t+12 0.129% 0.167 -2.967% -3.839 *** -0 239% -0.309 -0.977% -1.264 t+13 0.705% 0.912 -2.378% -3.077 *** 0 781% 1.011 -0.236% -0.305 t+14 -0.010% -0.013 -2.388% -3.091 *** 0 234% 0.303 -0.001% -0.002 t+15 0.070% 0.090 -1.885% -2.439 ** -0 107% -0.138 0.046% 0.059 t+16 0.817% 1.057 -0.713% -0.922 0 662% 0.857 0.932% 1.206 t+17 0.291% 0.377 -0.554% -0.717 0 130% 0.168 0.924% 1.196 t+18 -0.045% -0.058 -1.036% -1.340 -0 646% -0.837 -0.015% -0.019 t+19 0.502% 0.650 -0.793% -1.026 0 418% 0.540 0.353% 0.457 t+20 -0.077% -0.099 -1.075% -1.391 0 137% 0.178 0.480% 0.622 Source: Own calculation based on Bloomberg data 107 Note: The table reports the average abnormal CDS spreads (AARs) and the cumulative average abnormal CDS spreads (CAARs) based on the constant mean return model (CMRM) and the market model (MM) for different periods around the event day (tO). Test statistics are represented in the column with the header "Sig." indicating the significance of returns where *, **, and *** indicate statistical significance at the 10%, the 5%, and the 1% levels, respectively. As we can see in the results for the cumulative average abnormal returns in Figure 1, most of the decrease in the CDS spreads tend to start in the period from (t-18), according to the CMRM, and (t-6) according to the MM. This indicates that the information about an upcoming CoCo issuance is being spread prior to the announcement date. Table 2 shows that while the average abnormal returns are not significant over the entire event period, they are on average negative, indicating that the credit market values the CoCo emission. However, cumulative average abnormal returns are statistically significant for the event period from (t-15) to (t+15) according to the constant mean return model (CMRM), and only on (t+7) based on the market model (MM). Table 2 shows cumulative average abnormal CDS spread changes for the period from (t-9) to (t + 14) are statistically significant at 1% level, and 6 results outside this timeframe are statistically significant at 5% level, using the constant mean return model (CMRM). At the same time, there is just one statistically significant result at the 10% level, according to the market model (MM). The discrepancy in the significance of the results can be explained by comparing the abnormal return variances of the two models, since the lower market model variance is carried forward to the aggregate results leading to more price results. Despite the discrepancies in the significance of results, we can conclude based on both the market model (MM) and to the constant mean return model (CMRM) that CDS spreads show a systemic reaction to the announcement of contingent convertible instruments and are economically significant for the bond holders that value the decreased default probability and reduced risk incentives of the issuing institution. Although cumulative average abnormal returns are not statistically significant, we observe that there are much more positive than negative, suggesting that the issue of CoCos is considered as a positive signal for the shareholders. The result is in accordance with the hypothesis of mixed signals where the positive effect of the decreased default probability is canceled by the risk of dilution and the increase of the bond portfolio for which the bank has to pay coupons. At the same time, we find significant negative CDS spread changes around the announcement date. The negative effect on the CDS spreads can be explained by the additional protection buffer offered to senior creditors by the contingent convertible capital. 4 Conclusions Our results confirm that market shows a systemic reaction to the announcement of contingent convertible instruments and that bondholders value the decreased default probability and reduced risk incentives of the issuing institution. The obtained results follow most closely the previous event study findings of Avdjiev et al. (2015) regarding the announcement effect of CoCos on CDS spreads. However, our approach is different from the one used by Avdjiev et al. (2015) considering that our sample consists of banks from countries with different levels of development while they focused on advanced economies with the exception of the euro area periphery (Greece, Ireland, Italy, Portugal and Spain). The similarity of results proves that in many contexts the differences between the two approaches is not significant. This can be explained by the fact that the vast majority, both by number (84%) and by issuing volume (72%), of the CoCo issuing institutions are from Europe. In this context, we can conclude that the additional banks added to the sample do not have a significant impact on the estimation. At the same time, we obtain different results compared to the studies that used a shorter estimation period (Schmidt and Azarmi, 2014; Ammann et al., 2015; Rudlinger, 2015). This is due to the relatively young nature of the CoCos market, with the first issuance made in 2009. 108 The total amount issued in 2015 is about USD 76 billion, which is 24% of the total amount issued from 2009 to 2015. Therefore, our research contributes to the previous findings on contingent convertible instruments by providing up-to-date information on the developments of contingent convertible capital and their effect on the perceived default probability of banking institutions. The results of our study illustrate that the issuance of contingent convertible instruments is viewed as positive information by the credit markets. The decreased probability of default of the issuing institution is reflected in the lower CDS spreads. Additionally, obtained results allow us to conclude that the market sentiment regarding the issuance of CoCos follows the intention of the Basel Committee on Banking Supervision to strengthen the quality and consistency of the regulatory capital base. However, important question emerged and that is how these instruments will react in financial distress, considering that none of CoCos has been tested in crises. Also, in order to provide a more detailed analysis on the effects of issuance of contingent convertible capital instruments on banks' financial stability, further research should evaluate the CoCos announcement effect on other, more elaborate, market indicators, such as distance to default (DD). Acknowledgments The support of the Masaryk University internal grant MUNI/A/1025/2015 is gratefully acknowledged. References Ammann, M., Blickle, K., & Ehmann, C. (2015). Announcement effects of contingent convertible securities: Evidence from the global banking industry. University of St. Gallen, School of Finance Research Paper, No. 2015/25, pp. 1-39. Avdjiev, S., Kartasheva, A., & Bogdanova, B. (2013). CoCos: a Primer. Bank for International Settlements Quaterly Review, September 2013, pp. 43-56. Ball, C. A., Torous, W. N. (1988). Investigating security-price performance in the presence of event-date uncertainty. Journal of Financial Economics, vol. 22(1), pp. 123-153. BIS (2009). Strengthening the resilience of the banking sector. Consultative Document, Retrieved from: http://www.bis.org/publ/bcbsl64.pdf. BIS (2011). Basel Committee issues final elements of the reforms to raise the quality of regulatory capital. Consultative Document. Retrieved from: http://www.bis.org/press/ pll0113.pdf. BIS (2011). Basel III: A global regulatory framework for more resilient banks and banking systems. Consultative Document. Retrieved from: http://www.bis.org/publ/ bcbsl89.pdf. Calomiris, C. W., & Herring, R. J. (2011). Why and How to Design a Contingent Convertible Debt Requirement. Financial Institutions Center at The Wharton School Working Paper. Campbell, 1 Y., Lo, Andrew W., MacKinlay, A. C. (1997). The Econometrics of Financial Markets, Princeton University Press. Cihak, M. (2007). Systemic Loss: A Measure of Financial Stability. Czech Journal of Economics and Finance, vol. 57, pp. 5-26. Flannery, M. J. (2002). No Pain, No Gain? Effecting Market Discipline Via 'Reverse Convertible Debentures'. University of Florida Working Paper, pp. 1-32. Flannery, M. J. (2009). Stabilizing Large Financial Institutions with Contingent Capital Certificates. University of Florida Working Paper, pp. 1-34. 109 Hilsher, 1, & Raviv, A. (2014). Bank Stability and Market Discipline: The Effect of Contingent Capital on Risk Taking and Default Probability. Journal of Corporate Finance, vol. 53, pp. 1-45. Mackinlay, A. C. (1997). Event Studies in Economics and Finance, Journal of Economic Literature, vol. 35, pp. 13-39. Rüdlinger, M. (2015). Contingent Convertible Bonds: An Empirical Analysis of Drivers and Announcement Effect. Dissertation of the University of St. Gallen. Schmidt, C, & Azarmi, T. (2014). The Impact Of CoCo Bonds On Bank Value And Perceived Default Risk: Insights And Evidence From Their Pioneering Use In Europe. Journal of Applied Business Research, vol. 31(6), pp. 2297-2306. Spiegeleer, J. D., Schoutens, W., & Hulle, C. V. (2014). The Handbook of Hybrid Securities: Convertible Bonds, CoCo Bonds and Bail-In. Wiley. 110 Hedging Case Study in the Exchange Rate Commitment Regime Environment Jaroslava Dittrichová1, Libuše Svobodová2, Ivan Soukal3 1 University of Hradec Králové Faculty of Informatics and Management, Department of Economics Rokitanského 62, 500 03 Hradec Králové, Czech Republic E-mail: jaroslava.dittrichova@uhk.cz 2 University of Hradec Králové Faculty of Informatics and Management, Department of Economics Rokitanského 62, 500 03 Hradec Králové, Czech Republic E-mail: libuse.svobodova@uhk.cz 3 University of Hradec Králové Faculty of Informatics and Management, Department of Economics Rokitanského 62, 500 03 Hradec Králové, Czech Republic E-mail: ivan.soukal@uhk.cz Abstract: The paper deals with topic of hedging as an instrument of foreign exchange exposure treatment. This case study is focused on traditional export-oriented company PETROF, s. r. o. Analysis of scenarios is performed by the loss-profit comparison. The study dataset is consisted of internal financial statements gathered in period from 2010 to 2015 and provided the company representatives. Analyses are based on spot rates effective at studied periods and singed hedging contracts - forward derivatives. Analyses include data processing and proposals for consideration of macroeconomic factors that determine the company's business environment. Qualitative research is concluded with a discussion on proposed approaches and instruments. Despite the fact that in November 2013 Czech national bank entered foreign exchange interventions regime with exchange rate commitment of 27 CZK/EUR and so the companies knows the expected rate of the Czech currency, the topic is still current. The importance of the topic will be greater with the upcoming end of intervention regime in 2017 when we can expect increased volatility of the Czech crown again. Keywords: Financial derivatives, forward, hedge, foreign exchange risk, monetary policy JEL codes: F31, F37 1 Introduction On 7 November 2013 the supreme governing body of the Czech National Bank (thereinafter as CNB) started to use foreign exchange market interventions. This decision regarding this instrument was made already on the autumn 2012 after reaching "technical zero" interest rate. Czech koruna exchange rate was the only one more or less conventional instrument left because negative interest rates were not taken into consideration. This decision outlined the exchange rate asymmetric commitment of CZK appreciation close to CZK 27/EUR. Further appreciation would and is being prevented by purchases of foreign currencies against the Czech koruna on the foreign exchange market by the CNB. This is the managed zone of managed floating regime. A floating approach is applied at levels weaker than CZK 27 to the EUR. It allows the CZK exchange rate to float according to supply and demand on the foreign exchange market. Statement of the Bank Board from 4 January 2016 informs about a plan to use foreign exchange interventions at least till 2017. Weaker home currency increases imported goods and services prices creating inflation pressure although low oil prices prevented from clear manifestation in a way of reaching the inflation goal. On the other hand the effect of domestic output consumer preference, due to foreign goods and services price increase, was and could not be affected by oil market situation. Increase of domestic output is supported by the lowest interest rates in history for more than a year. Another stimulus again comes from the exchange rate 111 respectively from increased price competitiveness improving financial results of domestic export-oriented companies. The positive effects of higher investment activities again improve figures related to the domestic economy increasing even the lagging macro-indices such as wages and employment rate. CNB is determined to intervene as long as necessary for reaching the inflation goal. Foreign exchange market interventions are going to end when significant inflation pressure would manifest and or when the Bank Board will be under control of new members with a negative attitude regarding the interventions. It is highly probable that the first situation won't take place. If so then restrictive monetary policy would have taken the place increasing among other the interest rates. However the eased monetary policy is likely to continue more than one year. The paper is focused on the impacts of foreign exchange market interventions in business practice. Cooperation with the PETROF, s. r. o. company (thereinafter as PETROF), with a domicile in Hradec Králové, was arranged. By different means of communication we acquired from this company internal accounting data and business documents regarding the contracts from 2010-2015 period. Older data are not available due to a new information system implementation. At first a literature overview is performed, followed by methodology and the goal. The main part of the paper concerns the PETROF company and its case study of possible foreign exchange exposure treatment by forwards. Since PETROF did not use forward hedging the forward rates had to be taken from other company. Consequently we present an example of how the foreign exchange interventions influenced foreign exchange exposure management since the November of 2013. This case is compared to forward contracts from the period before the interventions. Forward offer is taken form the offer made for EXCON, a. s. (thereinafter as EXCON) since PETROF did not use the forward hedging. Jílek (2005) describes forward as a contract regarding the exchange of previously agreed amount of money in one currency to another at specific date and with specific exchange rate. The exchange expresses the ration of exchange between two currencies. It states how much of contract currency units equal the value of a base currency. Regarding direct quotation 25.630 CZK equal to base currency which is in our case EUR. The ration is inversed in case of indirect one and so 0.039 EUR equals 1 CZK. (Dittrichová et al., 2016) A case study, as one of the main qualitative study instruments (Hendl, 2012), would allow suggesting ways to achieve better results. The results are sum up at the end of the paper. 2 Methodology and Data Theoretical approach for the case of an internationally competitive but domestically monopolistic firm under exchange rate uncertainty (the model that fits the PETROF's market position regarding the high quality classic keyboard music instruments market) is represented e.g. by Wong (2007). The general model's outcome is that the effect of financial hedging on output is always positive. In other words it is profitable. However some empirical evidence clearly shows the problem that forward hedging could result in loss because the forward-real future spot rate correlation varies. The empirical study concerning exchange rate prognosis successfulness by forwards was performed e.g. by Ptatscheková and Draessler (2013). Another factor to mention is that nowadays monetary policy rather less conventional. Unlike the free floating regime the exchange rate commitment is now applied. Among other things it improves market transparency. That can be considered as by "publicly observable signal". These signals, as Broil and Eckwert (2009) claim, are the factors decreasing prices and contracts traded on the currency forward market. So again it can be considered as factor supporting the forward hedging. The main objective of this paper is to analyse development and use of instruments of CNB monetary policy and impacts on selected export-oriented company. The analysis will be accompanied by model examples using the acquired data. This paper can be beneficial, especially in processing of acquired data with consequent evaluation and 112 conclusions, not only for the selected company but also for other export-oriented companies and expert community. The main purpose is the analysis of impacts of eased monetary policy and financial derivatives hedging usage as foreign exchange exposure treatment for export-oriented company PETROF (the closest company form in British corporate law for s. r. o. is private company). PETROF did not use forward hedging. Then the forward derivatives offer for EXCON was taken into the calculation. This offer was up to date for given period and adequate regarding the contacts' value. The case study is based on accounting and business materials of PETROF company. The object of the analysis is long-term contract with Russian customer with an application of the forward rates taken from previously mentioned offer. The business case will be applied on exchange rate commitment regime of CNB. Qualitative research is sum up by evaluation of volatility of the Czech crown exchange rate, impacts of foreign exchange market interventions and conclusions on elaborated issues. The data presented are supplemented by tables and graphs. The article is based on primary and secondary sources. The primary sources are represented by the information from the companies. The secondary sources are represented in greater extent. They comprise information from expert literature, information collected from specialized press and other official sources, previous participations in specialized seminars and conferences related to the given subject. The processed calculations come from supervised theses (Simek, 2013; Sumpich, 2016). 3 Results and Discussion The PETROF company regularly delivers its products to the Russian company LLC SALON «ACCORD». The goods of nearly 23 million Czech koruna value were delivered in 2011-2012 period. The forward contracts of EXCON are used because the PETROF company did not used financial derivatives for hedging, see table 1 and 2. Table 1 EXCON's Forward Contract with Maturity in 2011 Contract Date of Foreign Home Forward ID signature Due date currency Volume currency rate 49750018 5.1.2011 19.7.2011 EUR 200.000 CZK 24.7 50256647 24.5.2011 29.7.2011 EUR 376.000 CZK 24.44 50380119 24.6.2011 29.7.2011 USD 47.000 CZK 17.03 49750020 5.1.2011 17.8.2011 EUR 650.000 CZK 24.70 50256661 24.5.2011 31.8.2011 EUR 176.000 CZK 24.44 49750022 5.1.2011 19.9.2011 EUR 200.000 CZK 24.70 49750032 5.1.2011 19.10.2011 EUR 200.000 CZK 24.70 49750036 5.1.2011 18.11.2011 EUR 200.000 CZK 24.70 50227993 17.5.2011 30.11.2011 EUR 170.000 CZK 24.24 49750040 5.1.2011 19.12.2011 EUR 650.000 CZK 24.70 50237799 19.5.2011 2.1.2012 EUR 170.000 CZK 24.30 50237828 19.5.2011 31.1.2012 EUR 170.000 CZK 24.30 50380051 24.6.2011 31.1.2012 USD 150.000 CZK 17.03 50105452 30.6.2011 18.7.2011 EUR 176.000 CZK 24.29 Source: (Simek, 2013) The volume is nominated in EUR and USD. Signed forward rate is e.g. 24.70 CZK per 1 EUR using the direct quotation CZK/EUR in case of contract 49750020 with maturity 17.8.2011. 113 Table 2 EXCON's Forward Contract with Maturity in 2012 Contract Date of Foreign Home Forward ID signature Due date currency Volume currency rate 50706167 13.9.2011 31.1.2012 EUR 150.000 CZK 24.35 50746085 21.9.2011 31.1.2012 EUR 600.000 CZK 24.66 50741533 20.9.2011 29.2.2012 EUR 150.000 CZK 24.60 50872333 21.10.2011 29.2.2012 EUR 2.300.000 CZK 24.85 50406172 13.9.2011 30.3.2012 EUR 150.000 CZK 24.35 50741535 20.9.2011 30.3.2012 EUR 150.000 CZK 24.60 50938842 9.11.2011 30.3.2012 EUR 800.000 CZK 25.19 50748088 21.9.2011 27.4.2012 EUR 600.000 CZK 24.66 50706175 13.9.2011 30.4.2012 EUR 150.000 CZK 24.35 50741537 20.9.2011 30.4.2012 EUR 150.000 CZK 24.60 50706177 13.9.2011 31.5.2012 EUR 150.000 CZK 24.35 50741538 20.9.2011 31.5.2012 EUR 150.000 CZK 24.60 50706184 13.9.2011 29.6.2012 EUR 150.000 CZK 24.35 50741539 20.9.2011 29.6.2012 EUR 150.000 CZK 24.60 50706191 13.9.2011 31.7.2012 EUR 150.000 CZK 24.35 50746055 21.9.2011 31.7.2012 EUR 150.000 CZK 24.66 50746090 21.9.2011 31.7.2012 EUR 600.000 CZK 24.66 50605195 13.9.2011 31.8.2012 EUR 150.000 CZK 24.35 50746058 21.9.2011 31.8.2012 EUR 150.000 CZK 24.66 50706197 13.9.2011 27.9.2012 EUR 150.000 CZK 24.35 50746063 21.9.2011 27.9.2012 EUR 150.000 CZK 24.66 50706200 13.9.2011 31.10.2012 EUR 150.000 CZK 24.35 Source: (Simek, 2013) The forwards, used for case study in the following part, are greyed in two previous tables. These forwards of EXCON company are applied on the business case of PETROF company with the Russian customer LLC SALON «ACCORD». The following table 3 presents profits and losses from the applied forwards, see the right-hand greyed column. The first column contains the date of maturity. Left columns consist of received payments from customer in CZK and consequently in EUR. This allows us to calculate the exchange rate of transaction. Forward taken from offer for EXCON company is the last but one column. Table 3 Forward Rates Application for the PETROF-LCC SALON «ACCORD» Case Date of signature CZK EUR Contract rate Forward rate Profit / loss 17.8.2011 301.516 12.480 24.16 24.70 6.739 17.8.2011 1.259.388 52.127 24.16 24.70 28.148 17.8.2011 2.425.591 100.397 24.16 24.70 54.214 6.10.2011 2.318.772 93.217 24.88 24.70 -16.312 6.10.2011 1.616.700 64.993 24.88 24.70 -11.373 18.11.2011 1.161.598 46.399 25.04 24.70 -15.543 18.11.2011 1.892.646 75.600 25.04 24.70 -25.326 18.11.2011 1.415.478 56.540 25.04 24.70 -18.940 12.391.694 501.753 1.604 Source: Own elaboration 114 Transactions in year 2011 took place in three days in various quantities. Forward rate was 24.70 CZK/EUR for all these three transactions. Forward with maturity nearest to the given business case was used for all the cases. Exchange rate of the transaction was however changing according to the exchange rate development in year 2011 (see figure 1). Czech Koruna appreciated against euro by one unit during the autumn. In the case of PETROF company in 2011 forward would pay off due to transactions on 17.8.2011 when profits from rate differences amounted to 89.102 CZK. This applies even despite of later transactions from October and November with loss-making forward rate. The result is profit of 1.605 CZK from hedging with the contract forward rate in year 2011. Figure 1 CZK/EUR Exchange Rate Development in 2011-2012 26 25.5 fa] 24.5 1 24 23.5 -1-1-1-1-1-1 January "11 May rl 1 September '11 January'12 May'12 September'12 January Source: Czech National Bank (2016b) Similar development of forward usage was in year 2012 in which the forward contract would have paid off during transactions on March 3rd. During business in August and November were analysed forwards creating the loss. Especially in the case of invoices due on August 15th by receiving payment in the amount of 167.601 EUR. In this case was forward rate (24.35 CZK/EUR) approximately by one koruna per euro lower than spot one (25.36 CZK/EUR). Fluctuations in the development of CZK/EUR exchange rate can be seen on the figure 1 and so the exchange rate profit opportunity from forward use was present. However it is true that forward contracts are not an instrument of speculations. However the EXCON company did not signed forwards at the best moment and so their use has negative consequences also on the business case with Russia for the PETROF company. Given contracts appear not to be profitable on the contrary - the loss during the 2011-2012 period is as high as 233.427 CZK regarding the trade volume of 22.882.987 CZK. Instead of previously mentioned forwards we can use CNB commitment which determines the lower bound of the exchange rate to euro since 7 November 2013. Economic agents can be sure that the exchange rate won't drop under the value of 27 CZK/EUR. The "post-intervention" or we can say "CNB commitment regime" period 2014-2015 transaction overview is in the table 4 bellow. The total value traded in this period is 18.477.519 CZK. 115 Table 4 The Commitment Regime and the PETROF-LCC SALON «ACCORD» Case Date of Contract Commitment signature CZK EUR rate rate Possible loss 26.3.2014 1360662 49750 27.35 27.00 17412 26.3.2014 1634545 59764 27.35 27.00 20917 5.8.2014 2466565 89239 27.64 27.00 57112 5.8.2014 875911 31690 27.64 27.00 20281 2.10.2014 3046418 110799 27.49 27.00 54845 2.10.2014 1307744 47563 27.49 27.00 23543 10691848 388805 27.00 194113 27.1.2015 396968 14331 27.7 27.00 10031 27.1.2015 1321068 47692 27.7 27.00 33384 13.5.2015 1111183 40606 27.36 27.00 14821 1.9.2015 1682211 62258 27.02 27.00 1245 1.9.2015 1587019 58735 27.02 27.00 1174 3.11.2015 1454700 53679 27.1 24.70 5367 3.11.2015 232518 8580 27.1 24.70 858 7785670 285881 66883 Source: Own elaboration One of the outcomes of recent regime is possible hedging substitution when the export company does not need to use derivatives to counter the possibility of the loss due to home currency appreciation. This situation also refutes the argument or rather a myth that due to the hedging the exporters won't have any gain from interventions. The companies never hedge 100 % of the trade and now it would be logical to have even lower volumes of hedged trade. To be more specific a survey of CNB and the Confederation of Industry of the Czech Republic showed that 37 % of the export was hedged in the second quartile of 2013. The number is lower by 2 per cent points regarding the expected export contracts hedging share in the next 12 months. Moreover we have to mention that small companies have more limited access to hedging instruments. Studied business case of PETROF shows that under the "CNB commitment regime" with sureness of 27+ CZK/EUR rate the intervention is more profitable than standard hedging by derivatives. Of course we mean short-time planning within the frame of the intervention regime. The exchange rate fluctuations result in maximal possible loss of 260.997 CZK in 2014-2015 period. Still it is worth mentioning that hedging brings another benefit in form of planning. With forward contract the cash-flow, liquidity and other financial management is much more predictable and accurate. So there is a question whether the company would still pay for forwards to make their planning more precise. The question is as well what if other currency would be taken into the consideration. Another step could be cross-hedge strategy or multi-currency diversification. As Alvarez-Dfez et. al (2015) show this strategy can solve hedge needs as well as generate return accordingly the value at risk ratio. 4 Conclusion The example of forward usage was given in our model case showing the CNB exchange rate commitment is now a part of financial management of the export company. It can be seen as a way of hedging instead of a standard forward contract. The forward rate offer of EXCON company was applied on Czech PETROF and Russian LLC SALON 116 «ACCORD» company regarding the period 2011-2012. Then since 7 November 2013 we applied CNB exchange rate commitment of 27+ CZK/EUR. The company would have profited less in case of using the forward contracts than without them at all. It is greatly influenced by the exchange rate development showing the appreciation trend. The only profitable forwards were at the turn of the year. For instance profitable forwards would have been those from 5 January with forward rate of 24.7 CZK/EUR with maturity of the August, September and November. During the August the profit would have been 89.102 CZK with so far profitable exchange rate. But at the end of the August strong depreciation of CZK took place and so the invoice payments resulted in loss of 87.497 CZK regarding the rest of the year 2011. In spite of this fluctuation the forward rate was profitable for 2011 with the exchange rate difference positive outcome of 1.605 CZK. It could have been much more without the depreciation of CZK. see fig. 1. In case of the year 2012 the EXCON forward rates from September and November 2011 were applied. November forward rate contract of 25.19 CZK/EUR appeared to be profitable because of its maturity in March concerning the PETROF-LLC SALON «ACCORD» case. The exchange rate profit was 33.505 CZK. On the other hand the forward contract from September 2011 with maturity in October 2012 with the 24.35 CZK/EUR rate created the loss. The EXCON obviously expected the appreciation of the CZK but the development was the opposite one. Later appreciation did not balance it and so the final result is a loss of 268.537 CZK in total for 2012. The forward rate was very unprofitable. The second case was using the CNB commitment from 7 November 2013 as a form of hedging. This commitment of 27+ CZK/EUR rate will be held at least till the 2017.If this would be considered as the foreign exchange resistance value the export company must profit from it and has no hedging issue since 2013. The second case shares the same idea with the first one however the forward rate was substituted for the commitment exchange rate. Virtually there is no space for loss and despite hedging price decrease suggested by Broil and Eckwert (2009) the company is better off. In 2014 the maximum possible loss resulting from difference between the commitment rate and the contract rate was 194.114 CZK. However CZK started to appreciate at the turn of the June forcing the CNB to come up with new interventions. Since then the exchange rate is above 27 CZK/EUR. From transactions from the first half of 2015 the maximum possible loss of 66.883 CZK might have been generated due to the difference of contract rate and commitment rate. Acknowledgments This paper is written with financial support of IGP from Faculty of Informatics and Management of the University of Hradec Králové to the Department of Economics. References Álvarez-Díez, S. et al. (2015). Hedging foreign exchange rate risk: Multi-currency diversification. European Journal of Management and Business Economics, vol. 25(1), pp. 2-7. Broil, U., Eckwert, B. (2009). Modelling information and hedging: The exporting firm. Economic Modelling, vol. 26(5), pp. 974-977. Czech National Bank (2016a). Foreign exchange market information. Retrieved from: http://www.cnb.cz/en/financial_markets/foreign_exchange_market/index.html. Czech National Bank (2016b). Selected exchange rates - charts. Retrieved from: http://www.cnb.cz/financni_trhy/devizovy_trh/kurzy_devizoveho_trhu/grafy_form.jsp. Czech National Bank (2016c). Statement of the Bank Board for the press conference following the monetary policy meeting. Retrieved from: 117 https://www.cnb.cz/en/ moneta ry_policy/bank_board_minutes/2016/160505_prohlaseni. html. Dittrichová, J., Svobodová, L., Soukal, L, Jindra, V. (2014). Základy financí. 2. vyd. Hradec Králové: Gaudeamus. Hendl, J. (2012). Kvalitativní výzkum: základní teorie, metody a aplikace. 3. vyd. Praha: Portál. JÍLEK, J. (2010). Finanční a komoditní deriváty v praxi finančního trhu. Praha: Grada. Ptatscheková, J., Draessler, J. (2013). Empirické Ověření Prognózy Prostřednictvím Forwardového Kurzu. E + M Ekonomie a Management, vol. 15(2), pp. 129-137. Šimek, P. (2013). Finanční deriváty pro zajišťování kurzových rizik. Diploma thesis. University of Hradec Králové. Šumpich, M. (2016). Monetární politika centrální banky. Diploma thesis. University of Hradec Králové. Wong, K. P. (2007). Operational and financial hedging for exporting firms. International Review of Economics and Finance, vol. 16(4), pp. 459-470. 118 Influences on Consumer Rationality Bohuslava Doláková1, Jan Krajíček2 1 Masaryk University Faculty of Economics and Administration, Department of Finance Lipová 41 a, 602 00 Brno, The Czech Republic E-mail: 175975@mail.muni.cz 2 Masaryk University Faculty of Economics and Administration, Department of Finance Lipová 41 a, 602 00 Brno, The Czech Republic E-mail: jan.krajicek@mail.muni.cz Abstract: The aim of this paper is to evaluate key components in consumer behaviour. Consumer behaviour is influenced by several factors; one with the greatest influence is genetic predisposition of each individual and also experience and knowledge acquired during the life of the society. The consumer is often irrational even in those cases in which we could expect rational behaviour (regarding to the prevailing economic opinion and rational choice theory). People are subjects to irrational influences. These influences come from their surroundings, feelings, emotions and other sources. Keywords: consumption, behaviour, consumer behaviour, rationality, irrationality JEL codes: G02, H31 1 Introduction As for the evaluation of results of the actions, as well as a numerical expression of this assessment, the in literature usually uses designations benefit or value to the result. The assessment, which is used for comparison, is usually referred to as benefit decision. The introduction of a specific label assessment, which is actually used, is particularly timely in light of the fact that, according to the results of the decision-making research, there are more types of evaluation result when all types are in some way related to decisions, but not the same origin. People in the moment used as a decision-benefit just one of them. Rational behaviour defined Pareto two conditions, named axioms. We can say that these axioms are built completely full microeconomic theory consumers. This theory states that the individual is only mechanism, acting on genetically embedded instincts. According to this model, a person should receive stimuli from the environment and by It would determine its next action. But the situation is not as simple as it might seem. Process hides many simplifications and assumptions with a significant influence. It is important to note that sensory perception is not perfect. Many medical experiments demonstrated that the ability to recognize human senses the reality is very limited. Moreover, the load capability repeatable mistakes that we cannot avoid results of psychological experiments point to the fact that behaviour patterns are frequent cases learned - they are influenced by perceptions surrounding accompany us throughout our lives. Psychological knowledge in this case is undoubted assistance, precisely because psychology seeks to define complex systems of human perception. The results can then expand the capabilities of microeconomic analysis. And that is the goal wide behavioural economics - to extend the neoclassical mathematical-logical model of significant quantities discovered in psychology. It is a huge step that will develop new areas of research, and especially also contribute to a more realistic perception economic science. Author Max Planck this problem indicates Phantom term problem. In the case that we are trying to define the problem too narrowly, it is likely that becomes so limiting factor on the basis of the theory will be build. On the basis of this opinion, we conclude that Pareto defined rationality of economic man to become such a limiting factor. Noyes this fact and their findings on this issue published in 1950. Phantom term problem with his entire theory of the vanguard crisis economic theory as such. All conclusions that creates, are in 119 fact applicable to almost nonexistent kind of behaviour. It is therefore almost impossible to such a theory has become a tool for economic policy. It is therefore the a place to entertain the question whether it is necessary to individuals and define its behaviour so strictly and close it so within the framework of the utility function, which is defined only one parameter. This parameter is consumption. Psychological knowledge in this case is undoubted assistance, precisely because psychology seeks to define complex systems of human perception. The results can then expand the capabilities of microeconomic analysis. And that is the goal wide behavioural economics - to extend the neoclassical mathematical-logical model of significant quantities discovered in psychology. It is a huge step that will develop new areas of research, and especially also contribute to a more realistic perception economic science. Consumption One of the most important driving forces of the market economy is consumption. Consumption of various goods and services is typical for a man because it allows people to live at a certain standard of living. For years, the consumption was seemed as a logical rational outcome of basic needs of the individual. Rationality Many experts explain consumer behaviour of individuals on the basis of rational decisionmaking economic theories. These theories see a man as a rational being that is in his economic decisions motivated primarily by the facts that he is able to reasonably consider the situation and its consequences in order to obtain the greatest benefit for the least expenditure of labour. Decisions of a rational man (homo economicus) are made by the judgments based on a sufficient amount of relevant information and efforts to get the greatest value at the lowest cost. Many recent researches have shown that consumer behaviour is much more driven by emotions, intuition, and others - these are for neoclassical economic theory utterly irrelevant influences (Kahneman; McKenzie, 2003). Current research findings of behavioural economists show that the consumer is often irrational even in those cases in which we could expect rational behaviour (regarding to the prevailing economic opinion and rational choice theory). Traditional economists argue that human decisions are rational, that people are able to estimate the value of all goods and pragmatically calculate, what will bring them greater benefits. People are therefore trying to maximize their profits and optimize their costs. Postmodernism puts behavioural economics in direct contrast to the classical economic theories. Behavioural economists believe that humans are subjects to irrational influences. These influences come from their surroundings, feelings, emotions and other sources. People decide rather on the basis of short-sighted decisions than by long-term plans. People decide according to other stimuli, than for a rational satisfaction of needs. Consumer behaviour Consumer behaviour is a type of human behaviour. A behaviour that is associated with the use of a particular object (product or service). It includes the reasons that lead consumers to use certain goods and ways in which it carried out, including the effects that accompany this process. Consumer behaviour is part of the human behaviour and cannot be understood independently. It is influenced by several factors, one of which has the greatest influence genetic predisposition of each individual as well as experience and knowledge acquired during the life of the society. For this reason, it is not possible for consumer behaviour mark only conduct associated with the immediate purchase or using the product, but also some other influences that determine it. 2 Methodology and Data The main purpose of this work is to create a model of rational consumer behaviour under the influence of irrational factors. Another purpose of this work was to provide general recommendations to improve the success of transmission of advertising messages in 120 those cases where it is supposed influence of irrational factors of human behaviour on rational consumer behaviour. Models of consumer's perception and interpretation of consumer behaviour Figure 1 Models of Consumer's Perception and Interpretation of Consumer Behaviour Marketing models Source: Own elaboration Rational models Rational models of consumer behaviour consider a consumer as rational being who is in purchasing decisions influenced only by relevant information and an effort to maximize the satisfaction of his needs. Emotional, psychological and social elements in this process have only marginal role. The models assume a consumer with certain characteristics: fully informed of all the parameters, all options can create algorithm deciding which also consciously observes. Customer then monitors the link between income, prices, budgetary constraints, marginal benefits, cross-elasticity of indifference curves. Psychological models Psychological models are based on the influence of deeper motivation structures, showing how to reflect consumer behaviour influences coming from the unconscious. Consumer behaviour is explained by the relationship stimulus - reactions consequence of psychological processes. The consumer response is a response to a stimulus. These models of consumer behaviour are related to laws of conditioning defined by I. P. Pavlov. Importantly, this approach does not take into account the mental/thinking processes. They consider just external factors, such as various external stimuli and incentives. (Skinner, 1984). Sociological models Sociological models explain consumer behaviour as based on influences of the social environment of consumers. This environment is made up of social circumstances and social groups (Koudelka, 2006). Models are based on the assumption that individual has a strong need to adhere to social norms. This need and effort influences actual purchasing process. 121 Marketing models Consumer behaviour is influenced by many different factors that original models do not reflect. Original models do not reflect the impact of such things as consumer's habits, shopping conditions, or the consumer mood. Also a response to a stimulus is involved in the model, as Figure 2 illustrates. Figure 2 How the customer decides Alternatives evaluation Need identification Information search Aftershopping behavior Decision to shop Source: Own elaboration Irrationality in consumer behaviour With the growing importance of behavioural economics it is often possible to observe application of its results into practical advertising practice. Advertising experts are aware of irrationality, a factor which significantly affects consumers in the process of its decision making. However, it should be noted that not all inexplicable in the process of consumer's behaviour is also irrational (Koudelka, 2006). It is important to distinguish between general rational human behaviour and rationality, as it is perceived by the rational choice theory. In the first case, the wide definition says that rational behaviour can be defined as personal views and behaviours based on logic and objective analysis of all available information (Koukolfk, 2010). From the perspective of rational choice theory, it is considered that the rational individual evaluates every action time, cost, effort, money, investment, spending, spending, loss of (...) on one side, while good, favours, favour, benefit, advantages, benefits on the other side (Koukolfk, 2010). Factors influencing rationality Decision-making is greatly influenced by the way a person perceives and risk assessment, respectively, emerging risks. When a person is familiar to evaluated risk, it is possible to estimate the likelihood for something to happen. However, the uncertainty of probability is not known. The decision affects cultural sphere in which people grew up. The risks those are very unlikely to happen have people a tendency to overestimate, while the risks that happen highly are likely underestimated. Decision on the same issue may turn out quite differently, decides if the individual in a very positive or very negative emotional state. Emotions affect decisions at the time that precedes it, to the next time. People receive and use information in the form in which it is acquired without about them in this regard they were thinking. The decision is affected by the learned rules. There is a greater influence of people from constructed within a hierarchy, and on the contrary, it reduces the influence of the people with lower ranking. Deciding changes the degree of uncertainty with which people make decisions. Simultaneously the same influence on decision making has the value that people give to objects and things. Deciding is influenced by the stereotypes. Especially those that relate to beliefs about the characteristics and behaviour of members of human groups (automatisms, delusions and prejudices). With the knowledge of factors affecting rationality can say that thinking, talking or conduct that is less useful or less logical than it would have been their rational alternatives are termed irrational (Koukolfk, 2010). 122 3 Results and Discussion The work provides an insight on the impact of various factors of irrationality on decisionmaking process of consumer especially on change of the hierarchy of effects that the advertising messages create in consumer's mind. The model itself provides sake of explanation of mechanism for activation of irrational human behaviour factors in consumer decision-making. If these new findings will be taken into account by the process of developing advertising strategies, they can become very valuable. Thus, these findings can greatly contribute to increase the success rate of transmission of advertising messages and also can help to communicators to achieve the desired change in consumer's behaviour. Irrationality Irrational thought is a process, which leads to the conclusion or decision in the light of the evidence and considering time that was available, not the best that could have been achieved (Sutherland, 1994). At the same time there are some irrational judgments and decisions arising from the continuous ideological distortion, which can be avoided (Sutherland, 1994). There are special cases where is involved irrational behaviour determined by situations that person is located in. In many cases it is a situation where people are in difficult acute and chronic stress, especially if compromised their lives. Equally irrational behaviour is typical for people with disabilities person, mentally ill or intoxicated with narcotics (Koukolfk, 2010). Healthy people, who tend not exposed to the huge level of stress, make also irrational behaviour. This irrational behaviour is also one of the major factors that affect the economic behaviour of individuals. Irrational behaviour This type of irrational behaviour due to many different factors that can be divided into four basic groups - a mistake resulting from a first impression, emotional and social causes, thought errors and problems of intuition (Koukolfk, 2010). Mistake From first impression (availability error) The first and most crucial factor that negatively affects the rational individual behaviour is the first impression (Sutherland, 1994). It is probably the most important cause of irrational behaviour. Koukolfk (Koukolfk, Drtilova, 2002) notes the mistake is similar to the influence of propaganda. So is present a message that is specific, but simple, and emotional are addressed to imagination. Misleading first impression negatively affects the risk assessment. One type of availability error is called Halo-effect. It lies in the suppression of negative characteristics based on a particular positive feature individual. For example, experiments have shown that beautiful people are often considered smarter than you really are (Koukolfk, 2010), opposite the halo-effect called Devil effect when people are ugly compared to the actual ascribed worse properties or low intelligence. Emotional and social causes Emotional and social causes have a major negative impact on the rational behaviour. Sutherland (1994) identified seven of them: • obedience, • conformity, • belonging to group, • irrationality of organizations, • misplaced consistency, • incorrect use of rewards and punishments, • emotions, • thought errors. It is a different strain of thought processes and the judgements, which frequently arise under the pressure of the imperfections of human thought. Sutherland (Sutherland, 1994) presents ten basic intellectual errors: 123 • people ignore the views that testify against their opinion, • people distort evidence which is against their opinion, • people distort reality, • people form non-existenting connections, • people misinterpret the facts, • people do not know elementary statistical rules, • people have unreliable memory, • people committing incorrect conclusions, • generation people return to the average, • people are wrong on the issue of dependence and independence. Intuition Intuition and reliance on it is a very common cause of people's irrational behaviour (Koukolik, 2010). The belief in the superiority of intuition beyond rational analysis is part of the magical thinking that is evolutionary "natural", i.e. more or less innate, while critical thinking and statistical analysis must be learnt. 4 Conclusions The paper focuses on consumer decision making risk conditions. Part of this work was to point to behavioural theories. Consumer behaviour is influenced by several factors, one of which has the greatest influence genetic predisposition of each individual as well as experience and knowledge acquired during the life of the society. Consumer behaviour is also driven by emotions, intuition. Consumer takes into advance various alternatives for his decision. The amount of information is crucial to determine an optimal solution. Acknowledgements Support of Masaryk University within the project MUNI/A/0916/2015 "Behavioral and knowledge aspects of trading and pricing financial assets" is gratefully acknowledged. References Kotler, P. (1998). Marketing Management. Praha: Grada Publishing. Kotler, P. (2004). Marketing. Praha: Grada. Koudelka, J. (2006). Spotřební chování a segmentace trhu. Vyd. 1. Praha: Vysoká škola ekonomie a management. Koukolík, František a Jana Drtilová (2002). Živots deprivanty. Praha: Galén. Koukolik, F. (2010). Mocenská posedlost. Vyd. 1. V Praze: Karolinum. Mckenzie, Craig R.M. (2003). Rational models as theories not standards of behavior. Trends in Cognitive Sciences. Vol. 7, č. 9, pp. 403-406. Retrieved from: http://linkinghub.elsevier.com/retrieve/pii/S1364661303001967. Simon, H. A. (1991). Bounded Rationality and Organizational Learning. Organization Science., vol. 2, č. 1, pp. 125-134. Retrieved from: http://orgsci.journal.informs.Org/cgi/doi/10.1287/orsc.2.l.125 Skinner, B. F. (1984). The operational analysis of psychological terms. Behavioral and brain sciences, vol. 7, no. 4, pp. 547-581. Sutherland, S. (1994). Irrationality: why we don't think straight!. New Brunswick, N.J.: Rutgers University Press. Simon, H. A. (1955). A behavioral model of rational choice, The Querterly. Journal of Economics. 124 Analysis of the Relationship between Taxes and Social Benefits and Transfers in the EU Nadiya Dubrovina1, Jana Peliova2, Erika Neubauerova3 1 National Economy Faculty, University of Economics in Bratislava Dolnozemska 1, Bratislava, 852 35, Bratislava, E-mail: nadija@mail.ru 2 National Economy Faculty, University of Economics in Bratislava Dolnozemska 1, Bratislava, 852 35, Bratislava, E-mail: peliova@euba.sk 3 National Economy Faculty, University of Economics in Bratislava Dolnozemska 1, Bratislava, 852 35, Bratislava, E-mail: erika.neubauerova@euba.sk Abstract: Taxes have main contribution to the government revenue, tax revenue made up about 90% of total government revenue in the European Union. Government revenue, expenditure and deficit/surplus are main objectives of fiscal policy and the analysis of their dynamics plays very important role in the formation and coordination of the strategic and tactic tasks for socio-economic development in the countries. For the more detail analysis of dependence between taxes and social benefits and transfers we used time series of these indicators for EU countries. For the modeling the initial data as absolute values of the different taxes, social contributions and social benefit and transfers per inhabitant were used for 2002-2012. It should be noted that most of time series included linear trends that is why for the panel data analysis we used the first differences of these indicators. These models explain the complicated relationship between changes of taxes, social contributions and social benefits and transfers with individual countries effects (fixed effects) for each EU countries and can be used for prediction of features of social benefits and transfers as results of changes in fiscal policy in different EU countries. Keywords: taxes, social contribution, social benefits, social transfers, econometric modeling J EL Classification: H20, H53, C50 1 Introduction The problem of rationales for taxes and transfers is very popular among as researches as well as policy makers. Taxes and social transfers had very important role in the formation of the concept of "social state" and improving the living standards in the period of the Second World War in USA and, especially, in West Europe. Taxes as important part of public finance fulfill the three important functions: 1) allocation; 2) distribution and redistribution and 3) stabilization (Owsiak, 2005). Function of allocation is concerned with the governmental expenditures on socio-economic needs and their optimal proportions between public and private sectors in the economy. Functions of distribution and redistribution are related with the shift of part of incomes and wealth from rich part of population to the poor population. Here, for the realization of these functions the social transfers as instruments are used (Schultzova, 2009). Function of stabilization is very important as preventive measure for the negative consequences of the cyclic development of market economy. Social transfers cover the social help given by central, state or local institutional units. They include: old-age (retirement) and survivors' (widows' and widowers') pensions; unemployment benefits; family-related benefits; sickness and invalidity benefits; education-related benefits; housing allowances; social assistance; other benefits (Zubal'ova, 2008). 125 In most countries the total taxes are about 40% of national income and total monetary transfers are approximately 15% of national income. Usually monetary transfers are public pensions, unemployment and family benefits, means-tested transfers. Other government spending or in-kind transfers is made approximately 25% of national income and they are used for education, health care, police, defense, roads, etc. On long-run dynamics the ratio of taxes to national income is essentially changed, from less than 10% in the early of twenty century to 40% in nowadays. It is should be noted that different countries have own experience in the tax-transfers or tax-benefit systems and their efficiency in the reducing inequality between regions, social and ethnical groups, migrants and native population, etc. (Goni et al., 2008) A lot of publications concerned the study of income transfers and their role in reducing inequality was published in the USA. D.Betson and R.Haveman (1981) studied the role of income transfers in the observed reduction of income inequality among regions in the USA during two decades in the XX c. (from 1960 till 1980); geographical distribution of regions with greatest income inequality; the efficiency of income transfers programs in reducing regional market inequality and set of factors determine the impact of transfers in reducing inequality within states and regions. Based on the empirical analysis and econometric models, the mentioned authors found that the marginal impact of transfers on inequality reduction is larger in states with higher unemployment rates, larger average family sizes, a higher proportion of female-headed families and a higher proportion of aged persons (Betson and Haveman, 1981). Another authors, such as X.Wu, J.Perloff and A.Golan (2006) studied the effects of taxes and other governmental policies on income distribution and welfare in the USA during period of 1981-1997. They examined the distributional effects of major government tax and welfare policies in the USA and found that marginal tax rates have larger income redistribution and equilibrating welfare effects than social insurance or direct transfers programs. These authors analyzed the impact of all major government programs that directly or indirectly transfer income to the poorest members of society and variation of these transfers in real terms over time or across states during the period of 1981-1997. Due to the model they showed that the marginal income tax rates and the Earned Income Tax Credit play a more important role in equalizing income than do the other government programs. In addition, the mentioned authors found that several of the other government programs have undesirable distributional effects (Wu et al., 2006). Nevertheless, the large difference in the efficiency of tax-transfers or tax-benefit systems is observed between macro regions and countries in the world. M. Luebker (2004) shows results of the impact of taxes and transfers on inequality for different macro regions and countries. According to the analysis provided in his paper, the Latin American and East Asian countries have mildly redistribute transfer systems, but European countries have well-developed social security systems. Australia, Canada, Israel and the USA have noticeably higher inequality of disposable incomes than Europe (M. Luebker, 2004, 2011). Luebker argued that the income inequality growth over the past decades was driven by a greater dispersion of market incomes, but countries with the same market inequality achieved different outcomes, so political choice and institutional factors in the formation of effective redistribute results are very important in the national tax and transfers systems. Another important empirical research devoted the role of taxes and transfers in the solution of income inequality and growth was published in the report for the OECD countries (OECD, 2012). In this report the six important facts were observed: 1. Inequality of income before taxes and transfers is mainly driven by the dispersion of labor income. 2. Tax and transfers systems reduce overall income inequality in all countries. Approximately 75% of the reduction in inequality is due to the transfers and 25% to direct household taxation. 126 3. In some countries, cash transfers are small in size but highly targeted on those in need. In others, large transfers redistribute income mainly over the life-cycle rather than across individuals. 4. The personal income tax tends to be progressive, while consumption taxes and real estate taxes often adsorb a larger share of the current income of the less well-off. 5. Some reforms of tax and transfers systems entail a double dividend in terms of reducing inequality and raising GDP per capita. Reducing tax expenditure, which mostly benefit the well-off, contributes to equity objectives while also allowing for a growth-friendly cut in marginal tax rates. 6. Other reforms may entail trade-offs between these two policy objectives. Shifting the tax mix to less-distorting taxes from social contributions to consumption would improve incentives to work and save, but stimulates the raising inequality. 2 Methodology and Data In our article the purpose was to analyze the relationship between tendencies of the main taxes in EU and the development of social benefits and transfers. The data of main components of government revenue and expenditures across EU countries were used for the econometric analysis and modeling. For the modeling the initial data as absolute values of different taxes, social contributions and social benefit and transfers per inhabitant were used for 2002-2012. In first stage of our research we analyzed time series of the mentioned data and built the linear trend models, which were reflected the character of tendencies in the most EU countries. Then we calculated the correlation matrices for the components of government revenue and expenditures across EU countries and found that these indicators were strong correlated for most of EU countries during 2002-2012. It should be noted that most of time series included linear trends and have strong correlations. That is why in the second stage of our research, i.e. for the panel data analysis we used the first differences of these indicators. The first differences of the mentioned indicators are count as differences between the current and former values and equal to annual the changes of the indicators in the original time series. The tests for the analysis of the causality between these time series were used. We used Granger causality test and found that for most EU countries the dynamics of taxes and social contributions defined the social benefits and transfers tendencies (Richter and Paparas, 2013). The different models on the panel data were tested and the models with fixed effects were used as the best. The fixed effects reflected the individual effect for each country in the econometric model based on the panel data. The built econometric model with fixed effects was used for the prediction of the annual changes of social benefits and transfers as dependence from annual changes of taxes and social contributions in each EU country. The soft Statistica and Eviews were used for the calculation and tested of the models. 3 Results and Discussion In 2013 for EU-28 taxes and net social contributions made 88.2% of total government revenue, in 2014 this value was 88.6%. Across different countries sum of taxes and net social contributions are changed more, from lowest level in Bulgaria (77.5% in 2013 and 77.7% of total government revenue) to highest level in Belgium (91.6% in 2013 and 92.2% in 2014). Coefficient of variance for these values in different EU, calculated as ratio of sample average to its standard deviation, is small, less than 5%. It means that sum of taxes and social contributions are varied slightly in EU during last time period. For EU-28 sum of social transfers, other current transfers and subsidies is varied about 51% of total government expenditure. In 2013 this value was 50.9% and in 2014 it raised up 51.3%. Across different EU countries sums of social transfers, other current transfers and subsidies as percentage of total government expenditure are changed more, from 37-38 % in Malta till 60-61 % in Germany and Luxembourg. Coefficient of variance for these values in different EU is about 15%. It means that policy concerned 127 social benefits and transfers are visible varied in different countries of EU, especially it is significant for the different design of main components in government expenditure. For the more detail analysis of dependence between taxes and social benefits and transfers we used time series of these indicators for EU countries. For the modeling we used initial data as absolute values of different taxes, social contributions and social benefits and transfers per inhabitant for 2002-2012. The recent data will be used for the comparative analysis of the real and predicted data. Most of time series of mentioned indicators included linear trends. It means that the linear trend models should be analyzed in the first stage of this research. In linear trend model we reveal two parameters of the linear function: intercept (or a0) and slope (or aj). The estimation of the parameter a0 is the estimated initial level of the indicator, when time variable is 0. The estimation of the parameter ai is the constant slope or the estimated fixed annual change of the indicator over [t-l,t] period. In table 1 the characteristics of the linear trend model are shown for the description of the taxes and social contributions dynamics across EU countries. Table 1 The Characteristics of the Linear Trend Model for the Taxes and Social Contributions Dynamics Country Taxes on production and Current taxes on income, Social contributions, imports, receivable wealth, etc., receivable receivable, (euro per inhabitant)_(euro per inhabitant)_(euro per inhabitant) linear trend: estimations linear trend: estimations and linear trend: estimations and correlation correlation and correlation a0 R a0 ax R a0 ax R BE 3271.82 102.51 0.96 4370.53 99.97 0.87 3824.41 192.76 0.99 BG 279.29 53.66 0.94 116.87 16.19 0.8 168.62 43.94 0.98 CZ 732.79 98.81 0.99 765.43 33.46 0.73 854.05 117.19 0.98 DK 6094 127.46 0.81 10219.85 289.45 0.9 5447.34 176.72 0.94 DE 2565.07 96.56 0.97 2647.49 106.77 0.9 4801.37 39.97 0.83 EE 650.6 107.15 0.95 468.67 40.28 0.79 376.54 108.4 0.96 IE 5036.94 -74.06n 0.33 4513.93 -12.41" 0.1 2648.71 286.05 0.97 EL 1899.99 54.49 0.69 1214.75 52.68 0.91 2107.54 155.78 0.92 ES 2311.7 7.21n 0.1 2099.87 31.24" 0.3 1812.02 172.59 0.99 FR 3742.57 98.84 0.96 2802.14 64.15 0.71 4144.56 186.22 0.99 HR 1326.36 59.55 0.86 425.81 29.84 0.72 925.91 55.02 0.94 IT 3261.48 47.35 0.73 3069.48 83.58 0.82 3604.45 142.83 0.99 CY 2551.07 87.33 0.55 1550.83 99.36 0.67 1637.39 140.26 0.98 LT 452.63 452.63 0.88 319.02 50.69 0.72 254.1 81.44 0.95 LV 451.5 73.22 0.94 486.07 11.66" 0.2 253.33 109.99 0.95 LU 6905.57 328.88 0.93 7341.39 423.54 0.97 7284.69 527.36 0.99 HU 1054.27 67.18 0.95 790.96 2.92" 0.1 977.56 64.34 0.87 MT 1420.21 75.13 0.96 1075.28 100.93 0.97 1256.07 74.52 0.98 NL 3611.45 69.5 0.74 3152.7 106.04 0.82 3052.31 113.7 0.98 AU 3839.17 129.23 0.99 3560.42 110.61 0.88 4926.8 187.17 0.99 PL 607.56 70.72 0.9 310.66 41.81 0.83 782.73 59.63 0.94 PT 1956.27 26.31 0.57 1131.05 39.29 0.81 1584.35 123.71 0.99 RO 221.08 62.67 0.93 113.28 31.8 0.88 98.15 70.22 0.95 SI 2008.43 51.25 0.85 1056.25 44.3 0.68 1795.23 123.45 0.99 SK 535.48 80.77 0.96 303.85 42.81 0.91 411.78 133.66 0.99 FI 3621.25 122.26 0.94 5084.43 61.96 0.56 4138.05 210.92 0.97 SE 4396.2 283.05 0.94 6310.69 122.06 0.57 4752.23 93.93 0.78 UK 3772.62 -7.53" 0.1 4728.7 -15.1" 0.1 3554.29 74.85 0.79 Note: n - estimation is not significant at level p<0.1; Source: Own statistical calculation based on Eurostat data 128 It is clear seen from the table 1 that the most of estimations of the parameters are statistically significant and correlations of the linear models are closed to 1. Nevertheless for some countries the stable linear trends in these indicators are absent. For example, there are no linear tendency in the taxes on production and imports dynamics in Spain and United Kingdom, and very weak linear trend of such indicator is in Ireland. In table 2 the characteristics of the linear trend model are shown for the description of the social benefits and transfers dynamics across EU countries. Table 2 The Characteristics of the Linear Trend Model for the Social Benefits and Transfers Dynamics Country Social benefits other Social transfers in - Other current than transfers in-kind kind transfers (euro per (euro per inhabitant) (euro per inhabitant)_inhabitant) linear trend: linear trend: estimations linear trend: estimations and and correlation estimations and correlation correlation a0 ax R a0 ax R a0 ax R BE 3824.41 192.76 0.99 1579.45 110.61 0.99 532.9 33.86 0.97 BG 168.62 43.94 0.98 12.16 7.75 0.95 19.04 8.71 0.55 CZ 854.05 117.19 0.97 378.85 48.89 0.99 47.91 20.14 0.98 DK 5447.34 176.72 0.94 470.79 20.73 0.95 864.21 39.37 098 DE 4801.37 39.97 0.83 1814.13 69.52 0.97 382.7 27.75 0.98 EE 376.54 376.54 0.96 57.27 16.21 0.98 44.71 16.49 0.92 IE 2648.71 286.05 0.97 497.6 46.13 0.96 435.64 0.71n 0.04 EL 2107.54 155.78 0.92 -12.72n 69.72 0.85 239.46 4.47n 0.25 ES 1812.02 172.59 0.99 433.14 24.25 0.88 263.4 12.52 0.75 FR 4144.56 186.22 0.99 1318.94 50.6 0.99 613.85 40.49 0.99 HR 925.91 55.02 0.94 130.24 9.01 0.85 103.69 -0.38n 0.08 IT 3604.45 142.83 0.99 605.2 14.21 0.87 324.69 10.84 0.85 CY 1637.39 140.26 0.98 10.12n 1.37n 0.43 331.25 28.42 0.77 LT 254.1 81.44 0.95 14.67 9.1 0.95 33.53 34.93 0.84 LV 253.33 109.99 0.95 47.91 14.93 0.97 4.4 14.41 0.9 LU 7284.69 527.36 0.99 2597.59 136.56 0.97 1415.3 120.28 0.94 HU 977.56 64.34 0.87 234.44 3.47n 0.29 147.32 11.82 0.82 MT 1256.07 74.52 0.98 45.15 6.07 0.97 108.46 18.78 0.95 NL 3052.31 113.7 0.98 1943.89 221.82 0.98 532.61 7.85 0.51 AU 4926.8 187.17 0.99 1320.64 70.92 0.99 650.35 16.75 0.83 PL 782.73 59.63 0.94 81.39 13.06 0.97 49.28 18.93 0.89 PT 1584.35 123.71 0.99 306.33 47.83 0.93 265.34 15.57 0.9 RO 98.15 70.22 0.95 15.92 7.2 0.91 1.99 12.39 0.92 SI 1795.23 123.45 0.99 244.11 14.29 0.96 169.55 21.22 0.9 SK 411.78 133.66 0.99 79.31 58.59 0.97 73.6 13.83 0.91 FI 4138.05 210.92 0.97 458.34 48.29 0.99 589.52 45.48 0.99 SE 4752.23 93.93 0.78 741.02 65.41 0.92 990.38 11.46n 0.46 UK 3554.29 74.85 0.8 0 0 0 832.18 5.67n 0.31 Note: n - estimation is not significant at level p<0.1; Source: Own statistical calculation based on Eurostat data From the table 2 that the most of estimations of the parameters are statistically significant and correlations of the linear models are closed to 1. Nevertheless for some countries the stable linear trends in these indicators are absent. For example, the social transfers in-kind are absent in United Kingdom and there are very weak linear trends of this indicator are observed in Hungary and Cyprus. It should be noted that most of time series were included linear trends that is why for the panel data analysis we used the first differences of these indicators. 129 Such as endogenous variables the first differences for social benefits other than transfers in-kind, euro per inhabitant (Zl); social transfers in -kind, euro per inhabitant (Z2), and other current transfers, euro per inhabitant (Z3) were used. As exogenous variables we used first differences for taxes on production and imports, euro per inhabitant (Tl), current taxes on income and wealth, euro per inhabitant (T2) and social contributions, euro per inhabitant (T3).Results of models defined the relation between taxes, social contributions and social benefits & transfers with fixed effects for each EU countries are shown in table 3. Table 3 Results of the Panel Data Analysis for Taxes, Social Contributions and Social Benefits and Transfers (Euro per Inhabitant) Zl Z2 Z3 Tl -0.1166** 0.0567*** 0.0739** (0.0459) (0.0198) (0.0351) T2 -0.0487 -0.0496*** -0.0518* (0.0357) (0.0154) (0.0307) T3 0.5613*** 0.118*** 0.0935*** (0.0681) (0.0294) (0.0333) Fixed Effects CI (BE) 125.1406 93.9903 18.0231 C2 (BG) 34.5592 2.5221 1.7213 C3 (CZ) 52.1974 26.5313 5.2033 C4 (DK) 218.1221 25.75 47.8415 C5 (DE) 17.7659 54.6382 19.495 C6 (EE) 57.7935 0.5328 3.0584 C7 (IE) 233.2791 45.8192 5.9564 C8 (EL) 108.6714 47.0607 3.294 C9 (ES) 126.2116 10.8221 3.1387 CIO (FR) 119.5574 33.2246 23.8413 CH (HR) 23.9773 3.9897 -4.5259 C12 (IT) 103.9254 3.7224 2.3726 C13 (CY) 113.7694 -10.489 15.793 C14 (LT) 43.2404 -0.1718 23.0126 C15 (LV) 51.6875 0.5864 2.3795 C16(LU) 360.0692 99.6259 95.8033 C17(HU) 44.8971 -5.1592 7.2458 C18 (MT) 65.6564 2.3635 26.4736 C19 (NL) 20.7012 177.4117 -8.2815 C20 (AU) 113.588 49.1385 -4.4077 C21 (PL) 27.0581 4.3524 9.9589 C22 (PT) 103.7356 40.5725 7.9155 C23 (RO) 42.8072 1.9258 6.0488 C24 (SI) 65.2408 -0.0694 7.7305 C25 (SK) 71.7887 37.4122 -0.6164 C26 (FI) 156.8223 24.9913 22.8367 C27 (SE) 202.9336 63.1172 15.504 C28 (UK) 79.7929 -6.2085 4.4317 R2 0.4854 0.5024 0.2712 F 117.4439 125.7112 46.3333 DW 1.9890 1.9976 2.5956 S.E. 119.7047 51.727 52.7232 Note: * - estimation is significant at level p<0.1; **- estimation is significant at level p<0.5, *** - estimation is significant at level p<0.01. In the parentless the standard deviations of the estimated parameters are given. Source: Own statistical calculation based on Eurostat data 130 It is should be noted that relationship between annual changes of social benefits other than transfers in-kind (Zl), social transfers in -kind (Z2) and annual changes of main kinds of taxes (Tl, T2 and T3) are more strong (R2is 50%) that the relationship between annual changes of current transfers (Z3) and annual changes of main kinds of taxes (Tl, T2 and T3). The positive and negative estimations of the parameters for the first differences (or annual changes) means the cyclic character of dynamics of social benefits and transfers as dependence from annual changes of main kinds of taxes and social contributions. We used these models with fixed effects for the predictions of the annual changes of social transfers and benefits. We assumed that for the next forecast period in studied EU countries the annual changes of taxes and social contributions per inhabitant will be equal the related estimations of the parameters dy (see table 1), so the planned annual changes of the taxes and social contributions per inhabitant will be constant over + period. Thus, introducing the planned annual changes of the taxes and social contributions per inhabitant in the econometric models with fixed effects for each EU country, we get the predicted annual changes of the social benefits and transfers per inhabitant. The predicted annual changes of the social benefits and transfers per inhabitant are shown in Fig.l. Figure 1 The Predicted Values of the Annual Changes of Social Benefits and Transfers in _EU Countries (per Inhabitant)_ 700 600 500 400 300 200 -{ 100 JU L L Hi I_ In.rJljL L L L BE BG CZ DK DE EE IE EL ES FR HR IT CY LT LV LU HU MT NL AU PL PT RO SI SK Fl SE UK □ Z1 P 1Z2 P OZ3 P Source: Own elaboration These models explain the complicated relationship between changes of taxes, social contributions and social benefit and transfers with individual countries effects (fixed effects) for each EU countries and can be used for prediction of features of social benefits and transfers as results of changes in fiscal policy in different EU countries. 4 Conclusions Taxes and social contributions influence to social benefits and transfers and define the wealth and social standards in the countries. In the EU we observed some diversity in the social benefits and transfers, in one hand, and their efficiency in point of social equality, in other hand. The efficiency of tax and transfer systems in the different EU countries depend not only from the more appropriate variant of composition taxes and social contributions, their redistribution in the forms of social benefits and transfers, but also from the various institutional factors, the transparency of national fiscal systems, their 131 synchronization and capacities to fulfill related administrative functions by their central, state and local governments. Acknowledgments This paper is the result of the research project Experimental investigation on incentives of economic subjects for paying taxes co-financed by Research Grant Agency of the Ministry of Education of the Slovak Republic. References Betson, D., Haveman, R. (1981). The role of Income Transfers in Reducing Inequality between and within Regions. In: Economic Transfers in the United States. University of Chicago Press. Retrieved from: http://www.nber.org/books/moon84-l. Goňi, E., López, J.H., Servén, L. (2008), Fiscal Redistribution and Income Inequality in Latin America. Policy Research Working Paper 4487.Washington DC. World Bank. Luebker, M. (2004). Globalization on and perceptions of social inequality. International Labour Review, vol. 143, No. 1-2, pp. 91-128. Luebker, M. (2011). The impact of taxes and transfers on inequality. TRAVAIL Policy Brief No. 4, ILO, Switzerland, pp. 1-8. OECD (2012). Income inequality and growth: The role of taxes and transfers. OECD Economics Department Policy Notes, No. 9. January 2012. Owsiak, S. (2005). Finanse publiczne. Teória i praktyka. Warszawa. Richter, Ch., Paparas, D. (2013). Tax and Spend, Spend and Tax, Fiscal Synchronization or Institutional Separation? Examining of the Case of Greece. In: Working Paper. International Network for Economic Research. Schultzova, A. (2009). Daňovnictvo: daňová teória a politika. Bratislava. Wu, X., Perloff, J., Golan, A. (2006). Effects of taxes and other Government Policies on Income Distribution and Welfare. Retrieved from: http://are.berkeley.edu/~perloff/PDF/tax.pdf. Zubaľová, A. (2008). Daňové teórie a ich využitie v praxi. Bratislava. 132 The Role of Life Insurance in the Context of Cover the Needs of the People in the Czech Republic Eva Ducháčková1 1 University of Economic in Prague Faculty of Finance and Accounting, Department of Banking and Insurance nám. W. Churchilla 4, 130 67 Prague 3, Czech Republic E-mail: duchack@vse.cz Abstract: The life insurance is considered to be an instrument to cover the needs of people, on the one hand, a tool of covering the consequences of the risk (death and other risks - accident, invalidity, illness etc.), and on the other hand, a tool for savings to cover the needs of people in post-productive age. At present, many factors affect the development of life insurance and especially its efficiency. In the use of life insurance as a means of addressing the needs of people in old age is in the last period on the Czech insurance market a number of problems. The problems arise from the form of life insurance products, from regulatory approaches in life insurance, from approaches to selling life insurance contracts. The paper focuses on the analysis and evaluation of the role of life insurance in current conditions and on the question of consumer protection issues in connection with the life insurance. The aim of the paper: analyze the role of the life insurance by the cover the needs of the people nowadays in the Czech Republic in the context new changes on the market, in the new conditions on the insurance market. Keywords: life insurance, death, unit linked life insurance, technical provision of life insurance, risk portion of premium, investment portion of premium JEL codes: G22, G28 1 Introduction Life insurance is intended for life natural events, i.e. demise and the rest of one's life. Its content and concept have been changing throughout its development. The original form connected with covering risks of the family breadwinner's demise and the consequences for the family members has been extended to cover additional risks affecting people's lives, such as injuries, invalidity, illness etc. by the savings portion (rest of one's life). Recently there have been a few tendencies and factors affecting life insurance development on the Czech life insurance market. At the same time, they have an impact on the life insurance efficiency from the perspective of covering the needs of people (Daňhel, Ducháčková, 2013). There are a few open questions: • recent developments on financial market, especially on interest rates and their impact on life insurance, • impact of the gender directive on life insurance (judgment of the European Court of Justice (ECJ) in March 2011 which removes the ability of insurers to use gender as a factor in pricing and benefits from 21 December 2012), • intermediated sale of life insurance, • tax questions of life insurances, • an increasing role of unit-linked life insurances, • products of lump-sum paid life insurances, • the effects of the use of Solvency II method for life insurance operation. Recent developments on the Czech insurance market indicate a decline in the field of life insurance and that is a decline both in the amount of premium written and in the number of insurance policies. If the indicators vary, it is possible to anticipate changes in the role of life insurance of cover the peoples' needs. On the other hand, there is a need to distinguish several roles of life insurance. Life insurance is used to provide risk solution 133 primarily the risk of death and the other risk of injuries, invalidity, critical illness etc. In this connection, there is a need to highlight a continuity between life insurance and credit risk solution -that implies the use of life insurance for covering credit risk of consumers. At the same time, there is an emphasis on saving portion that means the use of life insurance as a saving instrument especially for covering the needs of people in retired age. The paper analyzes some selected factors mentioned above. 2 Methodology and Data In this paper the author inquires about the effects which are the consequence of the changes on the life insurance market to the coverage of people's needs through life insurance for the next period. In processing the contribution were used the methods of description, deduction, analytic comparisons and literature review. The author uses life insurance information available primarily from resource materials of statistics of the Czech National Bank and the Czech Insurance Association. 3 Results and Discussion Recent Developments on the Czech Life Insurance Market In the period after the year 2000, developments on the Czech life insurance market have been destabilized due to a number of factors (at the same time we can see unstable developments both in European and also in a global perspective). Since 2001, State aid has been applied in form of tax advantage within life insurance. At that time, lump-sum paid life insurance played an important role. Life insurance is influenced by the state's approaches to covering of people's needs within social insurance (the higher amount of risk cover within social insurance the less room for commercial insurance) and existence of state-supported pension products. Furthermore, developments of life insurance was affected by gender directive as well. Recent developments on the market are determined especially by the situation on the financial market (very low interest rates or negative interest rate) and largely by the situation in life insurance mediation. In analyzing life insurance one cannot look away from the impact of changes in demographic structure of the population, because there has been a significant population decline in the age group of 30-45 year-olds. Figure 1 Premium Written on the Czech Life Insurance Market in Mil CZK 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Source: Basic information on life insurance, available at: http://www.cnb.cz/cs/dohled_financnLtrh/souhrnneJnformace_fin_trhy/zakladni_ukazatele_fin_tr hu/pojistovny/poj_ukazatele_tab04.html All these factors, some of which will be further specified, affected developments in premium written (see Figure 1), there was a significant decline (in 2015 compared to 2005, there was a 13% decline of premium written). This tendency is going on in 2016. 134 At the same time, recent developments also demonstrate a decline in the number of insurance policies (see Figure 2). Figure 2 Number of Life Insurance Policies in the Czech Republic 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Source: Basic information on life insurance, available at: http://wwwxnbxz/cs/dohled_financnLtrh/souhrnneJnformace_fin_trhy/zakladni_ukazatele_fin_tr hu/pojistovny/poj_ukazatele_tab04.html Until recently, emphasis has been placed on saving portion within the construction of life insurance - that means solutions to fill people's needs in retirement age (State aid moves towards that). The decreasing demand for life insurance leads insurance companies to offer life insurance products with the higher emphasis on risk portion -primarily solutions for critical illness, permanent consequences of accident, long-term care etc. Appreciation of financial means invested in life insurance Capital life insurance, where the risk connected with the investment of technical provisions of life insurance is carried by the insurance company, applies the regulation of the investment of technical provisions that are created within this insurance. Insurance companies can invest mainly in more conservative investment instruments (e.g. government bonds). The insurance company ensures the appreciation of the financial means invested in life insurance by applying the technical rate of interest. The level of the technical rate of interest is determined by the insurance company. On the one hand, the technical rate of interest means guaranteed minimum appreciation of the financial means invested in life insurance (investment portion of premium) and on the other hand it affects the price of life insurance. When determining the technical rate of interest, the insurance company has to take into consideration the situation at financial markets (from the perspective of achievable investment yields), the competitiveness of the life insurance product as well as the regulation of this technical rate of interest (since 2000, the technical rate of interest has been regulated by the state). There is Freedom of investment in unit linked life insurance, where the insured chooses the method of investing technical provision of life insurance, however, at the same time it carries investment risk. The problem in terms of running the unit linked life insurance lies in the approach towards its sale. It is common that the unit linked life insurance is sold to clients who do not understand the way it works, i.e. they expect outlined yield which is not achieved subsequently. Unlike the capital life insurance, the feature of the unit linked life insurance is that the insurance company does not guarantee the payment of the arranged part at the end of the insurance period; the size of the insurance benefit is based on the value of share units. Insurance companies do not specify the settlement to their clients in advance (the usual practice is that the amount of the settlement is calculated by means of insurance-mathematical methods depending on the duration of the period of insurance) although the client could obtain information for the respective years of the insurance period 135 related to the paid insurance premium, capital value of reserves with the guaranteed appreciation and capital value of reserves with the expected value of appreciation, including the method used by the insurance company to determine the level of profit-sharing. Intermediated sale of life insurance Life insurance is commonly arranged through insurance intermediaries or financial advisors. These sellers are remunerated with commissions for the arranged life insurance. As for the life insurance, these commissions are currently set at high values (savings products of life insurance show around 150-200 % of annual premium). Higher commissions are connected with the sale of the unit linked life insurance. Insurance companies pay such high commissions in order to enhance insurance contracts set of life insurance. On the other hand, this fact often results in efforts of the sellers (of several tens of thousands operating on the Czech insurance market) to sell the life insurance products (especially unit linked life insurance) even to clients these products are not suitable for. A current problem also related to the sale of life insurance is the so-called „over-insurance", i.e. sellers of insurance products convince the insurant to cancel the existing life insurance contract giving the reason he does not find it beneficial any more, and to conclude a new contract of an innovated form. This method is certainly profitable for the seller who collects money for the arranged insurance, however, completely disadvantageous for the insured person. Life insurance, especially life insurance with savings portion, has a long-term character and the technical provision of life insurance based on which the insurance company pays the settlement is created only after two or three years after the insurance is arranged, i. e. the client if he cancels the contract prematurely basically pays the cost of arranging the insurance policy and additional cost and the indemnity is not paid to him, or possibly only in a small amount. Products of lump-sum paid life insurance The period after 2000 shows the trend to apply lump-sum paid life insurance products on the life insurance market. They usually represent products with prevailing investment portion and minimal risk portion arranged for four to five years. These characteristics make it clear that it is actually no longer an insurance product but a savings product. These products basically represent an alternative to some bank products, especially in connection with changes in the area of bank products (e.g. related to the financial crisis) and they are largely offered by bank-insurance institutions. Products of lump-sum paid life insurance substantially influence the indicators of life insurance market (see Figure 3) although by their nature they are not real insurance products (Duchackova, Schlossberger, 2015). Unit linked life insurance Another area to deal with is represented by the products of unit linked life insurance. In the past, the unit linked life insurance was considered to be suitable for insureds who already are sufficiently provided with the standard life insurance (Hagelschuer, 1991, p. 44; Duchackova, Danhel, 2006, p. 382). Generally speaking, this is suitable for clients with certain knowledge and experience of investments. The significance of the unit linked life insurance was growing fast in the 90's of 20th century (Nyfeler, Lehmann, 1995, p. 6). The significance and share of the unit linked life insurance on the Czech insurance market substantially increased after 2000 (see Table 1). Increase in the share of the unit linked life insurance in the global prescribed premium has been determined by a number of factors, but mostly by the profitability for the insureds (shift of the investment risk to the insurant), thus resulting approach toward the sale through intermediaries (higher commission compared to other insurance products). The unit linked life insurance puts emphasis on flexibility, state subvention and the possibility of higher appreciation. 136 Figure 3 Development of the Premium Written of Life Insurance in the Czech Republic in Billion CZK 50 -e _ 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 □ current paid policies ■ lump-sum paid policies Source: http://www.cap.cz/statisticke-udaje/vyvoj-pojistneho-trhu A big challenge in relation to clients in life insurance is the issue of transparency and comprehensibility of life insurance products. Life insurance products are often intricately formed and incomprehensible to the client. Standard capital life insurance is viewed by the client as a guaranteed product where the insurance company is obliged to pay the previously arranged capital sum. However, misunderstanding the product by the clients may result in taking wrong steps - e.g. premature termination of the insurance policy (especially shortly after arranging the insurance policy) and wrong ideas about the amount of the so-called settlement. Table 1 Indicators of Unit Linked Life Insurance in the Czech Republic 2010 2011 2012 2013 2014 2015 Number of policies 2 024 2 618 2 867 3 029 3 145 3 178 798 238 261 495 950 659 Number of new 709 748 753 756 729 873 614 011 584 955 533 634 policies Premium written 33 956 36 584 37 799 34 763 37 246 35 125 363 233 094 925 298 244 -lump-sum paid 18 495 19 756 18 173 13 813 15 401 13 418 718 620 544 301 693 227 -new policies 19 581 19 160 18,632,9 12 628 13 972 13 173 999 608 16 742 142 488 Acquisition costs of 6 818 7 415 7 010 5 969 5 868 5 391 insurance policies 930 447 366 490 079 763 Cost of provisions 5 577 5 736 5 495 4 526 4 612 4 168 for intermediaries 270 525 480 489 539 726 Source: Basic information on life insurance, available at: http://www.cnb.cz/cs/dohled_financnLtrh/souhrnneJnformace_fin_trhy/zakladni_ukazatele_fin_tr hu/pojistovny/poj_ukazatele_tab04.html The unit linked life insurance is connected with the application of a whole range of fees that are intricately formed and often non-transparent. Regular fees of the unit linked life insurance are as follows: • administrative charges for administrative expenses charged for the whole policy period, • collection fees covered from each premium payment, 137 • fees of a certain percentage from the difference between the purchase price and the selling price of share units (bid/offer spread), • fees to cover initial costs (50 - 70 % of the annual premium of the savings portion for the period of 2-3 years). Furthermore, the unit linked life insurance is connected with other fees, such as fees for insurance policy cancellation, changes in insurance policies, extra withdrawal of share units. As a result of the structure of fees, the value of accumulated share units as part of the savings portion of unit linked life insurance at the level of paid premium is achieved only around the tenth year of the policy period (especially due to the fee to cover the initial costs of concluding the insurance policy). It implies that this insurance is suitable only in the long-term perspective. Considering the long-term nature of unit linked life insurance, an incorrect approach is the termination of the policy after a short time (or concluding a new insurance policy). Also, given the non-transparent structure of the product and the fees, the product is non-transparent. It is possible to enhance transparency through indicators characterising this product. The client should be provided with more detailed information about the structure of the insurance price (SUN indicator - standardized cost indicator). The price of unit linked life insurance consists of following portions: • risk portion - intended to cover risks, such as death, injury, invalidity, illness etc., • cost portion - intended to cover costs of assurer, such as acquirable commission, collection fee, administrative expenses, claims expenses etc., • investment portion,- a part of premium intended to purchase share units. In terms of the product transparency, it is suitable to inform the client about the structure of the insurance premium in the initial phase of the insurance with regard to the basic purpose of the insurance policy and client's expectations. In order to enhance the transparency of unit linked life insurance, other indicators may be recommended that provide a better picture, especially about the product expense ratio: • TANK indicator - expense ratio indicator generally used for investment products with periodical investment - basically an equivalent to RPSN (annual percentage rate of costs). TANK is a ratio of all expected paid fees to expected future investment. TANK = all fees/ future value of investment • TER indicator (Total Expense Ratio) indicates how much of the investor's property value is represented by the fund administrator costs per year (Cipra, 2006). TER = total operating expenses of the Fund/Total assets of the Fund • PER indicator (Product Expense Ratio) evaluates the product expense ratio and how much percent of the paid premium will not be invested but used to cover the administrative expenses. • Ongoing charge indicator related to individual funds of unit linked life insurance and expressing the amount of total costs of unit linked life insurance in relation to average controlled assets within the fund (in short, if a shares fund appraises the means by 5.3 % and the fund expense ratio is 2.1 %, the net appreciation of a share unit is then 3.2 %); the indicator includes all costs connected with the control of the shares fund and generally speaking if the funds are actively controlled, there are higher 138 costs, and the passively controlled funds show lower costs (Duchackova, Schlossberger, 2015). • RiY indicator (Reduction I Yield) that does not take into account the insured part of the product and all fees are expressed with one percent; its flaw is that it can be influenced by the inclusion of some cost portions in the risk portion of premium, which distorts the final evaluation through this indicator (Duchackova, Schlossberger, 2015). SUN indicator, which explains the layout of premium during the insured period, is the most important indicator from client's perspective. TANK indicator represents total costs of investment component of Unit Linked life insurance. TER indicator is suited for evaluation of cost of individual investment funds. 4 Conclusions Developments on the Czech insurance market can be described as unstable after the year 2000. Nevertheless, a decline in both premium written and number of insurance policies is typical for the last period. This decline was largely affected by the situation on the financial market. The insurance institutions are not able to achieve assessment of recourses of technical reserves of life insurance (primarily saving portion), which they were in the past. The fact will be emphasized by the need of risk solutions in life insurance operations following the use of Solvency II method. At the same time, problems in intermediated sales and unresolved regulation of intermediated sale have a negative effect on life insurance development. There is the problem with reworking of the life insurance policies (there is the role of provision size). The life insurance policies should be long-term, but currently this is not true. The next factor affecting the life insurance market development was the tightening in using tax advantage of life insurance. The state supports the life insurance, but the support is not effective (although the change was made from 2015). There are high costs connected with the life insurance especially provisions (see Table 1). In connection of a reduction in the trust in life insurance at present, there is an effort for increasing the transparency of life insurance, primarily United Linked life insurance. The indicators used for that purpose introduce closer the content of a particular product of life insurance especially within united linked life insurance. SUN indicator, TANK indicator and TER indicator are the most important indicators. Nowadays we can say that the life insurance is not covering the needs of peoples in full. The life insurance has often very small risk portion (for death or other risks). A trend for the next period is to operate life insurance with the emphasis on its risk portion primarily critical illness and long-term care. These questions are getting more serious with regard of population ageing. Also the saving portion of life insurance is nowadays problematic. Majority of life insurance are unit linked polities: the polities have short validity and that the insurance cannot fulfill the role of safe instrument for old age. The unit linked polities are not suitable for all clients, it is not adequate for risk-averse clients. Some life insurance product are not inherently insurance (they are more savings products, but disadvantageous) and so they cannot play the primary role of life insurance. This is especially true for lump-sum paid life insurance policies. Acknowledgments The paper has been prepared under financial support of project IGA Fl/21/2016 at the University of Economics and under financial support of constitutional support to long-term conceptual development of research at the Faculty of Finance and Accounting of the University of Economics Prague in 2016 IP 100040/1020. 139 References Cipra, T. (2006). Finanční a pojistné vzorce. Ind ed. Prague: Grada. Daňhel, J., Ducháčková, E. (2013) Today's Social Dilemma: Does a New Paradigm in Economics Demand Higher Ethics or Wider Regulation. Journal of Emerging Trends in Economics and Management Science (JETEMS), vol. 4 (1), pp. 98-102. Ducháčková, E., Daňhel, 1 (2006) Nové prvky v architektuře pojistných trhů v současné globalizační éře. Politická ekonomie, vol. LIV(3), pp. 382-393. Ducháčková, E., Schlossberger, O. (2015) Life Insurance and the Role of Financial Arbitrator for the Resolution of Disputes within Life Insurance. In: Proceedings of the 7th International Scientific Conference Finance and Performance of Firms in Science, Education and Practice [online]. Zlín, 23.04.2015 - 24.04.2015. Zlín : Univerzita Tomáše Bati ve Zlíně, 2015, s. 217-231. Hagelschuer, P. B. (1991). Lebenversicherung, In: Gabler Verisicherungensenzyklopädie, 5. vol., Wiesbaden, Gabler. Nyfeler, S., Lehmann, A. (1995). Lebensversicherungsprodukte im europäischen Vergleich. Versichrungswirtschaft, No. 1, Karlsruhe. Samoregulační standardy. Retrieved from: http://www.cap.cz/odborna-verejnost/samoregulacni-standardy-cap/modelace-vyvoje-pojištěni. Základní informace o životním pojištění. Retrieved from: http://www.cnb.cz/cs/dohled_financni_trh/souhrnne_informace_fin_trhy/zakladni_ukazat ele_fin_trhu/pojistovny/poj_ukazatele_ta b04.html 140 Quality and Efficiency of Bank Branch Services Vlastimil Farkašovský1, Ľubomír Pinter2 Matej Bel University in Banská Bystrica Faculty of Economics, Department of Finance and Accounting Tajovského 10, 975 90 Banská Bystrica, the Slovak Republic E-mail: 1 vlastimil.farkasovsky@umb.sk 2 lubomir.pinter@umb.sk Abstract: The goal of the paper is to deepen understanding between the two key indicators of bank branch production, the quality of services and the technical efficiency. The indicator of service quality is a composite indicator harmonized at the European level that includes the assessment of three main areas: the quality of customer service, the quality of sale of financial products, and customer satisfaction. The technical efficiency of the bank branches is analysed with the aid of non-radial SBM model by Tone. A spatial aspect is found to be a factor that affects both, quality and technical efficiency of the bank's branches. The paper reveals, that individual managers have still enough room to improve their skills in managing their branches. Keywords: quality, technical efficiency, the SBM model, correspondence map, cartogram JEL codes: R12, G21, G14 1 Introduction The economic growth and development of any economy hinges strongly on the proper functioning of the banking system. It is considered to be the backbone of the most economies and plays a vital role in attaining economic growth. The performance of the banks is determined by internal factors specific to the banks and by external macro- and micro-economic factors unique to the environment in which the banks perform their activities. Specific features of the Slovak banking sector can be studied in Horvátová (2014, 2015) and an impact of the crisis on the financial system of Visegrád coutries is explored in Lawson and Zimková (2009). As all these factors influence the degree of efficiency of the banks, the nexus between the internal and external bank environment demands the special attention of the regulatory bodies, shareholders, and managers of banks as well. In the production process of a commercial banks the retail segment belongs to the core of the business. Since financial crisis it have undergone a radical change and the notion of performance evaluation become even more studied. In this paper, we present results from an ongoing study on the multi-criteria assessment of the production process of bank branches. Main objective was to propose and elaborate alternative options for evaluating retail production process of 185 branches of one of the largest commercial bank in Slovakia and thus to complement the existing system of production evaluation of commercial bank branches with additional independent information. The broader study examined issues of technical efficiency, profitability, quality and dynamics of service by the mean of the data envelopment analysis and by modified decision-making matrix and the results have been reported in Zimková (2015). In this paper, the spatial analysis of the technical efficiency and quality of services is revealed by the means of correspondence map and cartogram to demonstrate the relation between the two key indicators of the bank branch production. Most DEA models developed for banks and bank branches consider issues of technical or operating efficiency and/or profitability (see Berger and Humphrey (1997), Berger (2007), Fethi and Pasiouras (2011) and Paradi and Zhu (2013) for an international survey of recent studies). A bank branch, however, needs to ensure not just high volume of output, but also volume of high quality. A branch may report high volume of products and services offered, as well as profits, but lose this advantage in the long run because of eroding service quality. Each branch utilizes some consumable resources to provide some level of service quality. The DEA model compares branches on how well they transform these 141 resources (inputs) to achieve their level of service quality (output), given their client base. In their paper, Soteriou and Stavrinides (1997) developed a DEA model which incorporated service quality output to provide bank branch benchmarks of internal customer service quality perceptions. As they stated, the model cannot be used alone to assess branch performance since it only considers a single service quality output which may ignore other important bank branch performance measures. Nevertheless, the model provides direction forwards service quality pitfalls. The main aim of the paper is to deepen understanding between the two key indicators of bank branch production, the quality of services and the technical efficiency. The indicator of service quality is expressed by a composite indicator harmonized at the European level that includes the assessment of three main areas: the quality of customer service, the quality of sale of financial products, and customer satisfaction. The technical efficiency of the bank branches is analysed with the aid of non-radial SBM model by Tone. A spatial aspect is found to be a factor that affects both, quality and technical efficiency of the bank's branches. The paper reveals, that individual managers have still enough room to improve their skills in managing their branches in both volume and quality of services as well. To the best knowledge of the authors, this paper is the only empirical inquiry into the relationship between technical efficiency and quality of services through the optics of correspondence map and cartogram. The spatial analysis was also applied in the research of Bod'a, Farkašovský and Zimková (2016) that revealed outcomes of the relationship between technical efficiency and profitability. Save the introductory and concluding sections, the body of the paper is organized into two other sections. The next section gives an overview of the database and methodology. The third section reports the results and includes their interpretation. 2 Data and Methodology Well established commercial bank divides its branches according to several criteria, in this contribution we will divide them according to average registered number of employees and according to region, in which the branches operate. In terms of the number of employees the branches are divided into four typologically different groups: branches of type I usually have 20 employees, branches of type II usually have from 10 to 19 employees, branches of type III usually have up to 10 employees and finally branches of type IV usually have up to 3 employees Technical efficiency is within the paper analysed in the context of nine regions chosen by bank, where branches of commercial bank operate. It is the Bratislava region - East (abbr. BAE), Bratislava region - West (abbr. BAW), Banská Bystrica region (abbr. BB), Košice region (abbr. KE), Nitra region (abbr. NI), Prešov region (abbr. PO), Trnava region (abbr. TR), Trenčín region (abbr. TRE) and Žilina region (abbr. ZI). Two of the nine territorial divisions of the analysed branches of bank are located in the capital city. One of the key factors in applying DEA in banking is the selection of the set of inputs and outputs. The selection of input and output variables is determined by identifying the particular economic problem of the production process being studied. An insight into the specification of the input-output set for DEA-based bank efficiency measurement provide Ahn, Le (2013) and Bod'a, Zimková (2015a). The selection of input and output variables in SBM model corresponds to the production theoretical principle and it reflects an important requirement of bank representatives, so that DEA model includes those variables, which are directly influenced by directors of branches of a commercial bank. The input side of the production process is therefore number of employees of a commercial bank branch and the output side is the volume of accepted deposits, volume of granted loans and the volume of participation certificates sold to customers in thousadns of euro. The anonymized data set was valid on December 31, 2014. 142 In measuring technical efficiency of individual branches, the non-oriented SBM model under the assumption of variable returns to scale is used. The non-oriented SBM model conforms to the fact that branch managers are capable of controlling and managing both the input and output side of retail branch production. The choice of the SBM model reflects the desire to measure technical efficiency in the sense of Pareto and Koopmans (see e.g. Zimkova, 2014; Bod'a and Zimkova, 2015b), and this measurement is then accomplished in a more comprehensive way than common or basic DEA models. Finally, as direct proportional links between inputs and outputs can scarcely be anticipated in the case of bank branch production, it is variable returns to scale that are acceptable as a valid and reasonable assumption. The SBM model yields for each bank branch in the sample an efficiency score (i.e. an estimate of the true value of the SBM) from the interval [0,1]. The value of one therein signifies that the concerned bank branch operates technically efficiently, and it cannot improve further on the quantity of labour force used or the amounts of services produced without affecting its production negatively. The technical details on the SBM model are well described in the paper by Tone (2001) who gave it a solid theoretical foundation. The indicator of service quality provided by analysed commercial bank branches is a composite indicator, which includes the assessment of three main areas: quality of customer service, quality of sale of financial products and customer satisfaction. The indicator of service quality is from the interval [0,100], the value of 100 signifies that the concerned bank branch provides service of the highest standard. 3 Results and Discussion In order to examine the effect of regional affiliation on the two performance characteristics under research, the bank's branches were cross-tabulated according their regional affiliation and ordinalized technical efficiency as well as according to their regional affiliation and ordinalized profitability. Table 1 The Correspondence Table of Regional Division of Commercial Bank Branches and Their Technical Efficiency Technical Region Frequency efficiency BAE BAW BB KE NI PO TR TRE ZI Lowest* 1 1 9 15 3 9 3 4 5 50 Low** 2 1 8 6 6 7 5 4 7 46 Medium*** 4 4 2 3 7 6 9 3 6 44 High**** 10 12 3 0 3 1 5 8 3 45 Frequency 17 18 22 24 19 23 22 19 21 185 Source: The authors. Notes: *The lowest level of technical efficiency, score between 0.00 - 0.22. ** Low level of technical efficiency, score between 0.221- 0.3, *** Medium level of technical efficiency, score between 0.31 - 0.45,**** High level of technical efficiency, score 0.451 and more. The score of technical efficiency is a measure of managerial skills of directors of commercial bank branches to utilize inputs of commercial bank branch (in this case the ability of employees) to produce financial services (in this case the volume of accepted deposits, granted loans and volume of sold participation certificates). The directors of commercial bank branches managing the branches of the commercial bank in Bratislava region reached the highest level of technical efficiency, as the absolute majority of them had technical efficiency at a high level, and 4 branches from both regions reached technical efficiency on a medium level. Into the group of managers technically able to ensure high technical efficiency of branches can be included also half of the directors of branches of commercial bank from Trenčín region, where 3 branches had technical efficiency on a medium level. The directors of branches with higher level of technical efficiency represent natural models for managers of branches with low or the lowest 143 technical efficiency. Among branches with the lowest level of technical efficiency belong branches in the Košice region and Prešov region. Table 2 The Correspondence Table of Regional Division of Commercial Bank Branches and Indicator of Quality Service Quality Region Frequency Service BAE BAW BB KE NI PO TR TRE ZI Lowest* 7 7 1 5 7 5 6 4 4 46 Low** 6 4 5 4 5 6 3 4 8 45 Medium*** 1 5 7 9 3 4 6 7 5 47 High**** 3 2 9 6 4 8 7 4 4 47 Frequency 17 18 22 24 19 23 22 19 21 185 Source: The authors. Notes: *The lowest level of indicator of service quality, value between 0.00 - 87.25. ** Low level of indicator of service quality, value between 87.26 - 88.55.. *** Medium level of indicator of service quality, value between 88.56 - 90.15... **** High level of indicator of service quality, value 90.16 and more. From the correspondence table of regional division and indicator of service quality in commercial bank branches follows that when evaluating the service quality branches of commercial bank reach different results than in case of employee profitability. The results of correspondence table do not confirm the assumption that the level of employee profitability depends on the service quality. On the contrary, Bratislava region with the highest employee profitability is the most numerous in the category of the lowest evaluation of quality of their services along with branches of the Nitra region. The most frequent evaluation of quality on the high level from the side of customers had branches in regions, which reached the lowest employee profitability, namely branches in Prešov and Trnava region. These two contingency tables were analysed in the framework of correspondence analysis whose results are exhibited Graph 1. It provides information about regional division of bank branches compared to technical efficiency and indicator quality indicator. Figure 1 Correspondence Maps of Regional Division of Branches and Selected Criteria Regions & Technical Efficiency Regions & Quality Service Row and Column Points Symmetrical Normalization Row and Column Points Symmetrical Normalization 1.0-c £ 0.0" 5 | b metfiLHn o1™ po 0 bac tn bb lowest "bavW ° ke O Oer Okral Or" N 0,0-c Q « * é 3 -05- baz lowest °tr£^e m o J* O po o q Okra Oq -1,0- 1 1 1 1 1 1 1 -1,5 -1,0 -0,5 0,0 0,5 1,0 1,5 Dimension 1 -li 1 1 1 1 1 1 -i.5 -1.0 -as 0,0 0.5 1.0 Dimension 1 Source: The authors. It reveals that the highest technical efficiency in general have branches in Bratislava region. Up to 10 branches in Bratislava region - East and 12 branches in Bratislava 144 region - West reached technical efficiency with chi-square distance closest to the highest level of technical efficiency. Further, four branches from each Bratislava region reached technical efficiency with chi-square is the closest to the medium level of technical efficiency. On the contrary, the lowest technical efficiency in general reach branches within Košice and Banská Bystrica region. Their chi-square distance from the lowest level of achieved technical efficiency is in fact the lowest. This fact is confirmed also by the contingency 3, which implies that up to 15 branches of the analysed bank from 24 branches located in Košice region reach the score of technical efficiency within range 0 -22, which corresponds to a quarter of the lowest scores of technical efficiency within the whole analysed group of commercial bank branches (lowest). Six branches are in a group with the highest score of technical efficiency between the first quartile and median (low), thus, in interval 0.221 - 0.3. In other words, 21 from 24 branches of commercial bank located in Košice region have technical efficiency lower than 30% compared to technically efficient branches of the analysed bank. Unfavourable results when assessing technical efficiency reach also branches located in Banská Bystrica region, while up to 11 of them reached technical efficiency of the lowest quartile of technical efficiency and 8 of them recorded technical level in the second quartile of technical efficiency. When evaluating the quality of offered services the order of success of commercial bank branches is different from the previous evaluations of technical efficiency and employee profitability. The highest quality of offered services recorded branches in Prešov and Banská Bystrica region, where 9 branches in Banská Bystrica region and 8 branches in Prešov region reached service quality with chi-square distance closest to the highest level of service quality. Conversely, the worst rating of service quality had branches in Bratislava region and in Nitra region. Table 3 The Adopted Notation for the Districts of Slovakia Abbr. District Abbr. District Abbr. District Abbr. District BN Bánovce nad Bebravou KK Kežmarok NM Nové Mesto nad Váhom SO Sobrance BB Banská Bystrica KN Komárno NZ Nové Zámky SN Spišská Nová Ves BS Banská Štiavnica KS Košice - okolie PE Partizánske SL Stará Ľubovňa BJ Bardejov KE I Košice I PK Pezinok SP Stropkov BA I Bratislava I KE II Košice II PN Piešťany SK Svidník BA II Bratislava II KE III Košice III PT Poltár SA Šaľa BA III Bratislava III KE IV Košice IV PP Poprad TO Topoľčany BA IV Bratislava IV KA Krupina PB Považská Bystrica TV Trebišov BAV Bratislava V KM Kysucké Nové Mesto PO Prešov TN Trenčín BR Brezno LV Levice PD Prievidza TT Trnava BY Bytča LE Levoča PU Puchov TR Turčianske Teplice CA Čadca LM Liptovský Mikuláš RA Revúca TS Tvrdošín DT Detva LC Lučenec RS Rimavská Sobota VK Veľký Krtíš DK Dolný Kubín MA Malacky RV Rožňava VT Vranov nad Topľou DS Dunajská Streda MT Martin RK Ružomberok ZM Zlaté Moravce GA Galanta ML Medzilaborce SB Sabinov zv Zvolen GL Gelnica MI Michalovce SC Senec ZC Žarnovica HC Hlohovec MY Myjava SE Senica ZH Žiar nad Hronom HE Humenné NO Námestovo SI Skalica ZA Žilina IL Hava NR Nitra SV Snina Source: The authors. The spatial distribution of the technical efficiency and quality service indicator of the bank's branches is more accurately demonstrated in the cartograms of Figures 2 and 3. 145 For each of the 79 districts of Slovakia, they display simple averages of either indicator computed for the bank's branches residing in the respective district. Figure 2 The Spatial Distribution of the Technical Efficiency* of the Bank's Branches in the Districts of Slovakia Note: ' Captured by SB M efficiency scores. Source: The authors For each of the 79 districts of Slovakia, graphs display simple averages of either indicator computed for the bank's branches residing in the respective district. The cartogram in Graph 2 reveals significant differences between the districts of Slovakia in the average level of the technical efficiency of branches. The lowest average level of technical efficiency in the provision of banking services was attained by the bank's branches in the districts Veľký Křtíš ("VK") and Bardejov ("BJ"), in either case lower than 0.15. These two districts alongside some other districts with a low average level of technical efficiency such as Kežmarok ("KK"), Trebišov ("TV"), Detva ("DT"), Sobrance ("SO"), Medzilaborce ("ML"), Revúca ("RV"), Košice - okolie ("KS") and Gelnica ("GL") belong to the least developed parts of Slovakia. On the contrary, the highest average level of technical efficiency was recorded by the branches in the districts Senec ("SC"), Bratislava I ("BA I") and Žarnovica ("ZC"), in each case higher than 0.80. The districts in which branches with a higher level of technical efficiency prevail are concentrated in Western Slovakia as the most economically developed part of Slovakia. Figure 3 The Spatial Distribution of the Service Quality indicator* of the Bank's Branches in the Districts of Slovakia Note: ' Captured by service quality indicator. Source: The authors The cartogram in Graph 3 demonstrates significant differences between the districts of Slovakia in the average level of the quality service of bank's branches. The lowest 146 average level of quality service was attained by the bank's branches in the districts Zlaté Moravce ("ZM"), lower than 84 points out of 100. Moreover, Bratislava II ("BA II"), Bratislava III ("BA III") and Bratislava I ("BA I") with the highest employee profitability is the most numerous in the category of the lowest evaluation of quality of their services along with branches of the Piešťany ("PN") and the Prešov ("PO"). On the contrary, the highest average level of quality service was recorded by the branches in the districts Medzilaborce ("ML"), Košice region ("KS"), Gelnica ("GL"), Žiar nad Hronom ("ZH") and Senica ("SE"), in each case higher than 82 points. 4 Conclusions This paper represents a results of the spatial analysis applied on the technical efficiency and quality of services by the means of correspondence maps and cartograms. It is a part of ongoing study on the multi-criteria assessment of the production process of bank branches for one of the largest bank operating in the Slovak Republic. Taking into consideration 185 bank's branches clustered from two points of view, the size of the branch and region in which the branch operates, we can state that the results are rather heterogeneous. Moreover, it was not proved that the technical efficiency of the bank branches is associated with the high service quality standards. It was proved that while comparing the technical efficiency and service quality indicator, the order of success of commercial bank branches is different. An outcome of this research enables: to provide important information to both senior and branch management; to improve multi-dimensional performance evaluation; to strengthen adjustment of organizational culture differences across bank's branches. It is evident that there is a large space for an improvement of managerial skills of bank's branch management. Acknowledgement The support of the grant schemes VEGA 1/0757/15 Augmentation of the theoretical construct of the SCP paradigm and of the efficient structure hypothesis in banking and insurance by the aspect of risk and their empirical validation in the conditions of the Slovak Republic and VEGA 1/0859/16 Dynamics of nonlinear economic processes is gladly acknowledged. References Ahn, H., Le, M. H. (2014). An insight into the specification of the input-output set for DEA-based bank efficiency measurement. Management Review Quarterly 64, pp. 3-37. Berger, A.N. (2007). International Comparisons of Banking Efficiency. Financial Markets, Institutions & Instruments, vol. 16(3), pp. 119-144. Berger, A.N., Humphrey, D.B. (1997). Efficiency of Financial Institutions: International Survey and Directions for Future Research. European Journal of Operational Research, vol. 98(2), pp. 175-212. Bod'a, M., Zimková, E. (2015a). Efficieny in the Slovak banking industry: a comparison of three approaches. Prague Economic Papers, 24(4), pp. 434-451. Bod'a, M., Zimková, E. (2015b). How Non-radiality Matters - Pareto-Koopmans Technical Efficiency in Production of Branches of a Slovak Commercial Banks. Procedia Economics and Finance, vol. 30, pp. 100-110. Bod'a, M., Farkašovský, V, Zimková, E. (2016). Technical Efficiency and Profitability in Retail Production of Bank Branches. In: Palečková, I. and Szarowská, I. (eds.) Proceedings of the 15th International Conference on Finance and Banking. Karviná: Silesia n University, 2016. pp. 1-13. 147 Fethi, M.D., Pasiouras, F. (2010). Assessing Bank Efficiency and Performance with Operational Research and Artificial Intelligence Techniques: A Survey. European Journal of Operational Research, vol. 204(2), pp. 189-198. Horvátová, E. (2014). Development and characteristics of Slovak banking system. In Political sciences, law, finance, economics and tourism, vol II. Book series: International Multidisciplinary Scientific Conferences on Social Sciences and Arts, pp. 201-208. Horvátová, E. (2013). Analysis of the problem of pro-cyclicality in the Eurozone and pro-cyclicality solutions in Basel III. In: European Financial Systems 2013. Proceedings of the 10th International Scientific Conference, Brno: Masa ryk University, pp. 126-133. Lawson, C, Zimková, E. (2009). The Credit Crisis: What lesson for Visegrad? Prague Economic Papers, 18(2), pp. 99-113. doi: http://dx.doi.Org/10.18267/j.pep.344. Paradi, J.C., Zhu, H. (2013). A Survey on Bank Branch Efficiency and Performance Research with Data Envelopment Analysis. Omega, vol. 41(1), pp. 61-79. Soteriou, A.C., Stavrinides, Y. (1997). An internal customer service quality data envelopment analysis model for bank branches. International Journal of Operations & Production Management, vol. 17(8), pp. 780-789. Tone, K. (2001). A Slacks-Based Measure of Efficiency in Data Envelopment Analysis. European Journal of Operational Research, vol. 130(3), pp. 498-509. Zimková, E. (2015). Retailový produkčný process v komerčnom bankovníctve a jeho hodnotenie. Bratislava: Wolters Kluwer. Zimková, E. (2014). Technical Efficiency and Super-Efficiency of the Banking Sector in Slovakia. In Enterprise and Competitive Environment 2014. Proceedings of the 17th International Scientific Conference, Brno: Mendel Univ, 2014. Book Series Procedia: Economics & Finance, vol. 12, pp. 780-787. 148 The Access to Instrument of Countercyclical Capital Reserves in the European Union and the USA Lubos Fleischmann1 1 University of Economics, Prague Faculty of Finance and Accounting, Department of Banking and Insurance W. Churchill Sq. 1938/4, 130 67 Prague, The Czech Republic E-mail: lubos.fleischmann@gmail.com Abstract: Repeating financial crisis brings new actions in regulatory rules of the financial markets each time. Regulatory authorities seek to increase the resilience of the banking sector through modified and brand new instruments, especially against cyclical risks associated with fluctuations in credit activity. The same is happening in the post-lehman approach to macro-prudential tools. This paper deals with the impact of countercyclical capital reserves, one of the new supplementary capital reserves, which introduces the Third Basel Accord (Basel III). The paper analyzes the objectives of the countercyclical capital reserve instrument and the method for determining its amount. A comparison is performed among rates set in other European Union countries and approach to the countercyclical capital reserve in the USA. The method also carries out a comparison between the European and American approach. The primary question of the paper is answered using empirical data that relates to the effectiveness of the chosen instrument to achieve the objective, which is to ensure the stability of the banking sector during the crisis and what rate of the countercyclical capital reserve matches the above objectives. The paper also aims to assess whether the tool can fulfill the high expectations of regulatory authorities. Keywords: countercyclical capital reserve, Basel III, regulatory capital, cyclical risk capital ratio J EL codes: CI 5, E 37, E43, G20, G21 1 Introduction Repeating financial crisis brings new actions in regulatory rules of the financial markets each time. Regulatory authorities seek to increase the resilience of the banking sector through modified and brand new instruments, especially against cyclical risks associated with fluctuations in credit activity. The same is happening in the post-lehman approach to macro-prudential tools. This paper deals with the impact of countercyclical capital reserves (CCyB), one of the new supplementary capital reserves, which introduces the Third Basel Accord (Basel III). The paper analyzes the objectives of the countercyclical capital reserve instrument and the method for determining its amount. A comparison is performed among rates set in other European Union countries and approach to the countercyclical capital reserve in the USA. The method also carries out a comparison between the European and American approach. The aim of this paper is an attempt to answer the primary question, which concerns the efficiency and impact of CCyB tools to achieve the goal of ensuring stability of the banking sector. Appropriate conclusions are drawn at the end of the paper. Capital adequacy of financial institutions Capital adequacy is based on will to cover all future losses of banks or investment companies that are subject to risk with shareholders' equity (ie. internal resources of the particular joint-stock company). Potential losses of financial institutions should bear primarily their shareholders and not the clients. The market regulation authority sets the minimum amount of these capital requirements. It is also necessary to find a compromise between the costs associated with the possible lapse of the bank. The actual capital base of banks is in practice often over the permitted minimum, which on the other 149 hand increases their credit ratings and allows them (eg.) getting cheaper loans. (Cipra, 2015) The capital adequacy represents the ratio of bank's equity capital and risk-weighted assets. Adequacy of internal resources represents maintaining a minimum amount of regulated capital due to volume and riskiness of its own assets. The value of the bank's capital should cover future potential losses from the current risks of financial institution. (Vokorokošová, Kočišova, 2009). Globally enhanced regulatory systems set up primarily so-called Basel Accords: Basel I, Basel II and Basel III. Within the European Union apply also financial regulation known as the CRR / CRD with the corresponding serial number (Capital requirements dire*ctive), which is primarily based on the above mentioned Basel Accord. Table 1 Development of the Basel Reform Effective since Risk approach Characteristics Basel I 1993 Inadequate risk management Very simple application. Easy reduction of regulatory capital without significant restrictions or risk transfer (easy regulatory arbitrage). Basel II 2006 Higher risk sensitivity Effect on a change in the behavior of banks. A number of gaps allowing banks to evade unpleasant consequences of regulation._ Basel III 2019 (full implementation) Very high risk sensitivity Removing gaps of Basel II. Significant impacts on business portfolio, liquidity and balance sheets of the banks. A significant increase in the qualitative and quantitative regulatory capital requirements. Source: Cipra, T. (2015): Riziko ve financích a pojišťovnictví: Basel III a Solvency II. Ekopress, s.r.o. ISBN: 978-80-87865-24-8 2 Methodology and Data New capital reserves introduced in Basel III The financial crisis hit financial markets worldwide in 2008. The banking sector run into problems in liquidity and capital adequacy. It turned out that the rules set by the Basel II directive had been inadequate. The Basel III rules require a higher volume, quality, consistency and transparency of capital. The directive improves the quality and volume of Tier 1 as the predominant component of regulatory capital. It simplifies and reduces Tier 2, cancels Tier 3 and tightens requirements for hybrid instruments. Basel III introduces three additional capital buffers: a capital conservation buffer CCB, a countercyclical buffer CB and a systemic risk buffer. The capital adequacy ratio (CAR) is a measure of a bank's capital. It is expressed as a percentage of a bank's risk weighted credit exposures. Also known as capital-to-risk weighted assets ratio (CRAR), it is used to protect depositors and promote the stability and efficiency of financial systems around the world. Two types of capital are measured: tier one capital, which can absorb losses without a bank being required to cease trading, and tier two capital, which can absorb losses in the event of a winding-up and so provides a lesser degree of protection to depositors. (Investopedia, 2016) 150 Tier One Capital + Tier Two Capital CAR =_-_-_ Risk Weighted Assets Countercyclical capital buffer CRD IV directive (in response to Basel III regulatory concept) introduced an important macro-prudential tool into regulatory practice in the EU, which is counter-cyclical capital buffer (CCyB). The purpose of this tool is to increase the resilience of the financial system to risks associated with the behavior of the banking sector during the financial cycle, especially with strong fluctuations in the credit dynamics that amplify cyclical fluctuations in economic activity. The banks should create this buffer according to the guidelines of regulatory authorities in the period of excessive credit growth, when (due to high credit expansion) usually increase financial imbalances that lead to accumulation of systemic risk. In contrast, the created reserve should be dissolved and used by banks as a capital cushion to cover losses during the downturn in economic activity accompanied by increased financial strain and rising credit losses. It is necessary to prevent the decline in the supply of credit and bank transfer of an additional shock from the financial sector to the real economy. (CNB, 2016) Based on public data from the Czech National Bank, we can observe an increasing trend of mandatory capital requirements for banks and credit companies in the Czech banking sector (see. Table 2). Despite this trend, the members of the CNB Bank Board decided to establish a countercyclical capital buffer from 1 1st 2017. Table 2 Minimum Capital Reserves in the Czech Banking Sector (in Mil. CZK) Period Bank and credit companies reserves Mandatory reserves Free reserves 31.12.2015 75.27 66.46 8.81 31.12.2014 63.8 60.9 2.9 31.12.2013 59.86 57.13 2.73 31.12.2012 55.59 54.73 0.86 31.12.2011 52.9 51.29 1.61 31.12.2010 52.95 50.63 2.32 31.12.2009 50.35 49.8 0.55 31.12.2008 45.73 47.22 -1.49 31.12.2007 41.47 42.18 -0.72 31.12.2006 36.28 36.59 -0.31 31.12.2005 33.32 33.41 -0.09 31.12.2004 31.62 31.2 0.42 31.12.2003 29.38 29.42 -0.04 31.12.2002 28.51 28.34 0.17 31.12.2001 28.25 28.12 0.13 31.12.2000 26.95 26.85 0.1 31.12.1999 25.49 25.52 -0.03 31.12.1998 84.69 84.28 0.41 31.12.1997 103.63 101 2.63 31.12.1996 113.61 115.46 -1.85 31.12.1995 75.84 75.69 0.15 31.12.1994 56.39 56.34 0.04 31.12.1993 39.39 39.27 0.13 Source: CNB, processed by the author The following Figure 1 shows the main indicators for calculating the rate of countercyclical capital buffer and Table 2 lists the deviations from the long-term trend in relation to the main indicators. 151 Figure 1 The Main Indicators for Calculating the Rate of CCyB 5 000 500 LT1 ID ľ- oa oa ai o Sft S> N> s* -Í5 Source: Data from www.cnb.cz and www.bnm.md Republic of Moldova —Czech Republic Contemporary Moldovan banking system includes three basic elements: the Central Bank, commercial banks and specialized financial-credit institutions, comprising both banking organizations as well as non-banks. The structure of the branch network of banks operating in Moldova is determined by administrative-territorial division of the country, being organized as follows: central bank; subsidiaries; agencies and representative offices. Market lacks specialized banks, even if the banks working in the field give priority to one or another market segment. In early 2006, in Moldova were operating 15 commercial banks. It was opened a branch of the famous Austrian bank Raiffeisen Bank. But in 2015 the number of banks decreased, 4 banks were bankrupt. Of the total number of banks, four banks have a capital formed of foreign investments (CB „Mobiasbanca - Groupe Societe Generale", CB "EXIMBANK-Gruppo Veneto Banca", CB ProCredit Bank, RCB Chisinau, 2 of which are subsidiaries of foreign banks (Agora Redaction 2014). Trends in the evolution of banking sector In recent years both in Moldova as well as in Czech Republic, defective managed banks, undercapitalized have undergone several changes and have become a functional and competitive banking sector. In Moldova still continue the transformation, passing over some obsatcoles as: restrictions imposed by Russia which have caused deterioration of the financial situation of exporters of farming products, as a result, these economic agents reimburses bank loans with greater difficulty, but the Moldovan banking system 168 remains a fairly stable yet, despite the impact of the crisis in the eurozone crisis that has slowed economic boom in Moldova. Following are presented (Table 1) the evolution of trends in the banking sector of the Republic of Moldova and the Czech Republic. The case of the Czech Republic is interesting because, among other things, its banking sector remained stable during the crisis and did not suffer the tremors that affected many other European countries (CNB: Basic Trends in the Czech Banking Sector 2014). Table 1 Trends in the Evolution of the Banking Sector Republic of Moldova Czech Republic The share of foreign investments in the banks' capital in 2015 constituted 82.9%, maintaining the same level as in 2014. Highly concentrated banking sector (TOP4 = 57% of the sector's assets). • 92% of bank assets foreign-owned. Assets in the banking system increased by 15% compared to the beginning of 2015. The sector has been very profitable in the EU context. The long-term liquidity ratio constituted 0.7, down by 0.1 p.p. compared 2014. Current liquidity ration constituted 41.5%, increasing compared to the end of 2014 Total impairment losses decreased significantly by 24.7% in 2015 as a result of a decrease in impairments on loans and receivables (by 23.1%). The share of non-performing loans in total loans increased, accounting for 9.95% as of 2015. Credit costs decreased in absolute and relative terms. 2015 both in The average capital adequacy maintain further at a high level (24%). Financial assets held for trading comprised 6% of total assets and declined by 75.3% year on year. Interest rates are declining, although their level is still well above the European average, from 22% in 2014 to 15%- 2015 The evolution of base rate is stable, 0.05% in 2013 till now. Improvements in loan portfolio quality. Uncertainties about the credit quality and bank capital in 2014 will continue to persist and now._ Maintenance of sufficient capital coverage. The revenues of banks are greatly focused on the provision of loans_ Source: National Bank of Moldova (2012), Annual Report 2012 and Pavlat, V. (2016), Biggest Czech Banks In The Mirror Of Annual Reports, Ecoforum, Volume 5, Issue 1 (8), 2016 The financial situation of the banking sector According to the 2016 Index of Economic Freedom, Moldova ranks 117th globally with the overall score of 57.4, and Czech Republic ranks 21th globally with the overall score of 73.2. The real economic growth rate builds onto the economic growth rate by taking into account the effect that inflation has on the economy. The statistic shows the growth in real GDP in Czech Republic and Republic of Moldova from 2006 to 2015 (Figure 2). Grow potential of Czech Republic remains constrained by impediments to the geographical mobility of workers, skills mismatches and weaknesses in the business-legal environment. Despite a pickup in growth, inflation stayed low, remaining on average below 2% compared to Moldova where inflation rate is above 5%. 169 Figure 2 Economic Growth: the Rate of Change of Real GDP (%) Source: Author's own work based on data from investopedia.org Table 2 Basic Rate (Rate Applied on the Main Short-term Monetary Policy Operation) 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Republic Moldova, of % 14,5 15 17 9 6 9,5 4,5 3,5 6,5 22 Czech Republic, % 2,5 3,5 2,25 1 0,75 0,75 0,05 0,05 0,05 0,05 Source: Author's own work based on NBM and CNB of Czech Republic reports In Republic of Moldova the base interest rate are declining, although their level is still well above the European average, from 22% in 2015 to 15% in this year (Table 2). In Czech Republic in comparison with Moldova the base interest rate is at lower level, and mentains this level from 2012, that has a good impact on bank's activity. Figure 3 Banking Non Performing Loans to Total Gross Loans (%) 20% -i- o% H—i—i—i—i—i—i—i—i—i—i Source: Author's own work based on World Bank data From figure 3 can be observed that in Republic of Moldova persists the problem of non performing loans, this indicator affects the entire banking sector, in comparison with Czech Republic where this indicator is below 6% and mentains stable. The worsening of quality of the loan portfolio is due to a bad management, high base interest rate, currency depreciation and other factors. The Czech banks have the highest level of assets as % of GDP than Moldovan banks from 90% (2006) increased to over 120% (2015). Bank Assets as % of GDP in Republic of Moldova are below 80%. Efficiency of Moldovan and Czech banking sector Banking efficiency study will allow us to compare the performance of the banking system in Moldova with the Czech Republic. The banking sector in Moldova against the Czech Republic, remained poorly integrated into the national economy, placing it on the last places in the region at the Chapter share of lending in GDP (Stavarek D. 2003). Taking the example of the Czech Republic banking sector is stabilised, it shows good financial results and has sufficient capital to cover the assumed risks (Barta and Singer, 2006). Assessing the profitability of commercial banks rely on bank profitability indicators. The main indicators of this analysis are: Return on equity (ROE), Return on assets (ROA), 170 Equity Multiplier (EM), and Net profit rate indicator (Cociug, V. (2008) "Banking management, collections of problems"). The typical size of ROA is 0.5 - 1%. More than a half of the Czech banks have reached ROA over 1%, which is considered as a success value in the banking industry. The typical size of the ROE in developed countries is approximately 10-12%. A higher rate of return on financial capital could be the effect of a low or elevated expression of ability to obtain through debt, additional resources, as in the case of Moldova in 2006-2007 and 2015. Table 3 The Value of Bank Efficiency Indicators: ROA, ROE 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Republic of ^ Moldova, O % 3,44 3,94 3,49 -0,39 0,54 1,9 1,1 1,5 0,85 2,1 Czech Republic, 0,9 1,1 1,2 0,87 1,1 1,19 1,3 1,23 1,19 1,2 % Republic Moldova 20,64 24,16 19,9 "2'12 3,04 10,6 5,6 9,4 5,86 12,78 O % _ * Czech Republic, 12,1 15,3 16,7 13,8 14,4 9,9 11 12,8 7,28 12,9 %_ Source: Author's own work based on reports of NBM and CNB of Czech Republic We take an example for 2014 and calculate the profitability indicators (table 4) as follow: - Equity multiplier (EM) - Show indebtedness of the bank, how many times are greater assets than equity _Total assets Equity - Net profit rate (Rpn) - shows the relation between expenses and operating income, and effectiveness of expenditure, resource and management and ensure the functioning of the bank: Npr= Netprofit *100% (2) Operating income Table 4 The Financial and Profitability Indicators Country Total assets Equity Net profit Operating income EM Npr Republic of Moldova, mil MDL 69,095 11,765 713 6848 5,87 10,41 Czech Republic, mil CZK 214686 512332 64264 378693 0.41 16,9 Source: own calculation based on data from "Banking management, collections of problems", ASEM 2008 - 137 p. Matoušek and Taci (2005) examined the efficiency of the Czech-banking system and they found that results indicated that foreign banks were on average more efficient than other banks, although their efficiency was comparable with the efficiency of'good' small banks in the early years of their operation. Based on these results, it was argued that the early privatization of state-owned commercial banks and a more liberal policy towards foreign banks in the early stages of transition would have enhanced efficiency in the banking system. Stavárek and Polouček (2004) conclude that to achieve greater efficiency, a bank should be large, well-known, easily accessible and offer a wide range of products 171 and services, or if small, must focus on specific market segments, offering special products (Palečková (Řepková), Iveta, (2013), Estimation of Banking Efficiency in the Czech Republic: Dynamic Data Envelopment Analysis). The problems in the banking sector. For example in Czech Republic this situation occured after 1990 when of the total of created 18 banks, in the case of nine banks the license was withdrawn or went into special administration procedure and other situation occured in 1994, when 3 banks went bankrupt (Spulbar C. 2012). The same case was in Republic of Moldova, when in 2015, Moldova's banking sector went through a qualitative and quantitative reorganization process. The National Bank of Moldova ruled to revoke the licenses of financial activity of three commercial banks, with the liquidation process being launched. In table 5 are represented some of problems that faces the banking system from both countries: Table 5 Problems in the Banking Sector Republic of Moldova Czech Republic Insufficient regulation and supervision; Lack of secure facilities to use funds in the currency of the Republic of Moldova; Qualitative deterioration in the structure of loans provided to customers, growing share of poor quality loans to the detriment of good one; The problem of access to credit is an extremely actual; High real interest rate; The involvement of banks in financial service such as capital markets, leasing, factoring, or insurance is low; Tax evasion and a lack of trust in the banking system are partly to blame. The national currency went down by 26% in nominal terms against US Dollar and 13% against Euro. The high level of concentration and relatively low degree of competition in the banking market. • The overall "political" environment and subsequent pressure for the banks to provide financing for the privatized companies; • Serious flaws in the credit screening processes adopted by many banks implies that the problem of non - performing loans in bank portfolios may become more serious; • The oligopolistic structure of the Czech banking sector during those years, where just a few banks (including the formerly state-owned and still state-controlled banks) played a predominant role; • The high costs of operation of banks; • Excessively high involvement of the banks of a single state can pose a risk of greater instability, especially at times of economic swings and insufficiently diversified portfolios. • The CNB has reservation about enlarging the power and responsibility to inspect nonregular enteties within a group. Source: Tomsik, V. (2011), Financial System in the Czech Republic: The Story of Two Decades, National Bank of Moldova Conference 4 Conclusions This research has enabled a broader analysis of the situation in the banking sector of the Republic of Moldova and the Czech Republic, after which we can deduce following conclusions: Moldova could benefit enormously from the experience of the social and economic transformations of Czech Republic, achieving the transition to a market economy, maintaining broad access to basic services and establish rule of law. In the transition from a centrally planned economy to a market economy and one-party political system to a democratic regime, the Czech Republic has advanced faster than Moldova. Given that in some cases both countries followed the same path to a market economy, some experiences, positive and negative, can be a lesson for the future development of 172 Moldova. Although, Czech experience can not be total applied to Moldovan, there are some base lessons which must be appropriated. In conclusion, macroeconomic stabilization achieved is one of the most significant successes of Moldova in its way to a market economy, but still more efforts must be made to reduce the risks of inflation, even more because inflation is not only a macroeconomic indicator, but an indicator of the economy's competitiveness. As shown in the Czech and Moldovan experience, politically dependent judiciary can significantly undermine banking sector climate. Another important lesson, which is required to be learned from the Czech Republic is the need to build an institutional infrastructure to support the efficient Moldovan banking sector. Banking system policy should be simple and transparent as possible. The chaos in the banking system did not bring economic progress anywhere, so it is necessary to maintain monetary and banking stability. It should be further promoted free and fair competition entering the market: the failed banks must be subject to a modern bankruptcy laws that would protect the rights of creditors. The banks that went bankrupt should not be saved by using state funds, but by privatizing them in transparent auctions with an emphasis on strategic investors with experience. The financial sector should not be forced to provide subsidized loans or lending based on political preferences (bankruptcies banks are very expensive, as shown by the Czech experience). References Agora Redaction (2014), Analysis of banking system of Moldova, assessments, forecasts and recommendations. Retrieved from: http://agora.md/analize/14/sistemul-bancar-din-republica-moldova-~evaluari-prognoze-si-recomandari. Bárta, V., Singer, M. (2006), The banking sector after 15 years of restructuring: Czech experience and lessons, BIS Papers no. 28. ČNB (2016). Annual reports on banking supervisory activities (2006-2015). Prague: Czech National Bank. Retrieved from: http://www.cnb.cz/en/index.html. Cociug, V., Cinic, L, Timofei, O. (2008). "Banking management, collections of problems", ASEM Ed. Chisinau. Economic Research, Federal Reserve Bank of St. Louis: Bank's Return on Equity for Czech Republic. Retrieved from: https://research.stlouisfed.org/fred2/series/. IMF (2008) Republic of Moldova: Financial System Stability Assessment—Update, Country Report No. 08/274, Washington D.C. IMF (2010), Czech Republic: 2010 Article IV Consultation Concluding Statement. Retrieved from: https://www.imf.org/externa 1/np/rosc/cze/ba nk. html. Matoušek, R., Taci, A. (2005). Efficiency in Banking: Empirical Evidence from the Czech Republic. Economic Change and Restructuring, vol. 37, pp. 225-244. NBM (2016). Information on financial and economic activities of banks. Retrieved from: http ://bnm. md/bdi/pages/reports/drsb/DRSBl. xhtml ?id=08Jang=en. Palečková (Řepková), I., (2013), Estimation of Banking Efficiency in the Czech Republic: Dynamic Data Envelopment Analysis, DANUBE: Law and Economics Review, vol. 4, pp. 261-275. Pavlat, V. (2016), Biggest Czech Banks In The Mirror Of Annual Reports, Ecoforum, Volume 5, Issue 1 (8). Spulbar, C. (2012). Comparative Analysis of Banking Systems. Sitech ed. Craiova. Statistical Data Warehouse (2016), Financial sector indicators for Czech Republic and Republic of Moldova. Retrieved from: http://sdw.ecb.europa.eu/. 173 Stavarek, D. (2003), Banking Efficiency in Visegrad Countries Before Joining the European Union, Workshop on Efficiency of Financial Institutions and European Integration October 30 - 31, Technical University Lisbon. Stavärek, D., Poloucek, S. (2004). Efficiency and Profitability in the Banking Sector. In Poloucek, S. (ed.), Reforming the Financial Sector in Central European Countries. Hampshire: Palgrave Macmillan Publishers. Tomsik, V. (2011), Financial System in the Czech Republic: The Story of Two Decades, National Bank of Moldova Conference. World Bank (2013), Bank non performing loans to total gross loans 2006-2015. Retrieved from: http://data.worldbank.org/country/czech-republic. 174 Competitiveness Assessment of Slovak Republic Regions Beáta Gavurová1, Tatiana Vagašová2, Viliam Kováč3 technical University of Košice Faculty of Economics, Department of Banking and Investment Nemcovej 32, 04001 Košice, Slovakia E-mail: beata.gavurova@tuke.sk 2 Technical University of Košice Faculty of Economics, Department of Finance Nemcovej 32, 04001 Košice, Slovakia E-mail: tatiana.vagasova@tuke.sk 3 Technical University of Košice Faculty of Economics, Department of Finance Nemcovej 32, 04001 Košice, Slovakia E-mail: viliam.kovac@tuke.sk Abstract: The issue of competitiveness is directly related to international trade. The theories explain the factors of international trade in such a way that there is visible indication of a need to change an approach to understand competitiveness in economic policy. Despite the fact competitiveness is discussed in many studies, there is still no clear definition of this term in the literature. Competitiveness can be viewed from multiple angles. Therefore, it can be analysed at the international level and at the national level too. Besides this introduction, the theoretic part of the paper offers an overview of the approaches applied to not only describe competitiveness as part of trade, but also to measure it and to evaluate it. The special attention is paid to the analysis devoted to the Slovak regions. Local targeting lies in observation the data from the particular regions. One of the approaches understands competitiveness in form of gross domestic product. Our analysis takes it in this way. There are the eleven indicators selected that are able to model the level of competitiveness. These indicators are based on distribution of population, patents, households' income, employment, tertiary educated employed population, unemployment, economic activity of population, health insurance, criminality, municipal waste, and road network. Keywords: competitiveness, region, Slovak Republic JEL codes: C32, C33, R10 1 Introduction Nowadays, competitiveness is a frequently used concept which involves various approaches relating mainly to the international trade and welfare. In fact, it is concerned with the each economic subject - country, region, enterprise or individual. The most famous world institutions dealing with the competitiveness of the national economy are the International Institute for Management Development and the World Economic Forum. There are many definitions of competitiveness at the national, regional, and local level. According to the European Competitiveness Report, the economy is competitive when the country's inhabitants have enjoyed the high and rising standard of living, and high employment on a sustainable basis (European Commission Enterprise and Industry Directorate-General - Unit B4, Economic Analysis and Evaluation, 2010). More precisely, the level of economic activity should not cause an unsustainable external balance of the economy, or threaten the well-being of future generations. Storper (1997) defined regional competitiveness as the ability of the region to attract and retain companies with stable or increasing market share while achieving a stable or rising standard of living for those who contribute to it. The most frequently cited author in the discussions on competitiveness, Porter, pointed out that the best measure of the competitiveness of the region relative to other regions is their productivity. Competitiveness is then measured by productivity (Porter, 1999). 175 Delgado et al. (2012) define competitiveness as the expected level of performance per employee which is supported by the overall quality of the country representing a business location. Their research points out the three interrelated drivers of core competitiveness: social infrastructure and political institutions, monetary and fiscal policies, and microeconomic environment. They also find a positive and separate influence of each driver on output per potential worker. Ilzkovitz et al. (2008) analyse the structural changes that could explain the performance of trade inside countries and between countries too in the euro area. The research revealed a positive relationship between business performance and the degree of technological development. Structural competitiveness is the field of interest for Mothe and Paquet (1998) who argue that the overall performance of the national economy, flexible structure of its industry, the amount and structure of capital investments, technical infrastructure, and other factors are affected by externalities, namely economic, social and institutional framework of a country. However, many researchers address the regional competitiveness within one country. Fontagne and Santoni (2014) state that competitiveness is a multidimensional concept within the overall level of the country and is the result of the interaction of several independent drivers where individual companies perform the key players. The main determinants of regional competitiveness can be considered agglomeration and their growth, the existence of local institutions, regional policy, infrastructure and public finance. Martin (2005) defined the key determinants of regional competitiveness, namely productive capital, human capital, infrastructure, competitiveness and adaptability of companies, ability to innovate and their interaction with other factors. For comparison of regional competitiveness, he proposes to use an indicator of long-term sustainable economic growth of the region compared to other regions. Many Czech and Slovak researches devote to the analyses of regional competitiveness and its determinants. Provazníkova et al. (2009) describe the regions of the Slovak Republic, the Czech Republic, Poland, Hungary, Lithuania, Latvia, and Estonia. They analysed the role of the regions in these countries. In addition to basic macroeconomic indicators, as gross domestic product, labour productivity, total employment are, they also follow specific factors like age and educational structure of the population, transport infrastructure, investment rate, average wage, disposable income of households, spending on research and development in the regions, employment in science and research, amount of foreign investment. Some studies (Petr et al., 2010; Petr et al., 2011) describe the set of the composite indicators in terms of regional competitiveness using factor analysis and cluster analysis. A group of Slovak regions reached the worst results. Since 1993, the Slovak Republic has gone through the extensive reforms. These reforms have influenced every aspect of economic and social development. The transformation from a centrally planned economy to a market economy caused changes in understanding of economic and regional policy. This transformation revealed the structural weaknesses of the Slovak economy. These days, the Slovak Republic is a part of the European single market and the euro area. The economic growth, which has been reached in recent years, is reflected in the uneven development of the Slovak regions. This implies that the existing structural barriers have their natural origin at the regional level. The several analyses deal with the economic and demographic factors that impact on global or national competitiveness (Šikula, 2006; Hošoff and Hvozdíková, 2009; Šikula, 2010). For example, corporate sector, labour market, and foreign direct investment belong to the economic factors. Human potential performs as an instance of demographic factor. Analysis of the selected structural characteristics is described by the Frank (2014) who defines the main trends and barriers in the Slovak regions throughout the main macroeconomic indicators, demographic trends, income and educational level, and the impact of the implementation of cohesion policy on regional competitiveness. Morovská et al. (2009) argue that regional competitiveness is closely related to the four main aspects, namely the structure of economic activities, the level of innovation, the degree of accessibility of the region, and the level of education. Lábaj (2014) points out 176 that the increasing interdependence of national economies leads to necessity to revise the traditional view on comparative advantages and competitiveness of the economies at the national or regional level. 2 Methodology and Data The dataset involves the eleven variables.^ It was obtained from the database of the Statistical Office of the Slovak Republic (Štatistický úrad Slovenskej republiky) which provides primary source these statistics. Data As the elementary input data for the regression modelling process, the mid-year state of all the variables is applied from 1993 to 2012. The eleven elementary fields covering contents of regression analysis are: • distribution of population; • patents; • households' income; • employment; • tertiary educated employed population; • unemployment; • economic activity of population; • health insurance; • criminality; • municipal waste; • road network. Methodology We have applied the regression analysis for the panel dataset. The selected methodology of the regression analysis is a pooling approach. This choice was done according to the executed tests that show usage of this methodology as the most appropriate manner to examine such a dataset. After choice of the pooling approach, we performed the Baltagi-Li joint test to find out presence of random effects. Difficult issue of this test is that it can reveal influence of random effects also with presence of serial autocorrelation. The model has x2 test statistics at level of 39.503 with 2 degrees of freedom and p-value at 3,276 . 10"10. This outcome means rejection of the zero hypothesis, which signals absence of random effects or serial autocorrelation. That is why, it should involve random effects. To confirm the outcome of the Baltagi-Li joint test, it is appropriate to perform another test revealing random effects themselves. The Bera-Sosa-Escudero-Yoon locally robust test investigates a presence of random effects within the regression model. Its forte lies in revealing local serial correlation itself. The model test statistics stands at 38.767 with one degree of freedom and p-value at level of 4.774 . 10"10. This test gives the result in terms of rejecting the zero hypothesis stating random effects are present in form of sub-random effects. It means their presence is confirmed, but they do not have such a big influence on explaining the explained dimension. The subsequent step is continuation of the previous finding - taking the Wooldridge test to explore found random effects. Its z test statistics reaches value of -1.0238 with p-value at level of 0.3059. This demonstrates that the zero hypothesis cannot be rejected. It means individual effect is observed. Model Specification The above mentioned dimensions are represented by the ten variables involved in the regression model. Each one is evaluated individually. Hence, the regression analysis has the subsequent equation: 177 GDP = fi0+filPS + fi2TEEPTP + fi3MHI + p4EP + p5TEEP + fi6UR + fi7EAP + + j8&HI + fi9C + j810MW + pnBN (1) where involved variables mean: • GDP - gross domestic product • B0 - a constant value; • Bi - regression coefficient of the PS variable; • PS - a population share of a self-governing region; • B2 - regression coefficient of the TEEPTP variable; • TEEPTP - a tertiary educated employed population to a number of patents ratio; • B3 - regression coefficient of the MHI variable; • MHI - households' income per capita per month in euro; • B4 - regression coefficient of the EP variable; • EP - employed population; • B5 - regression coefficient of the TEEP variable; • TEEP - tertiary educated employed population without constantly preparing for profession; • B6 - regression coefficient of the UR variable; • UR - unemployment rate; • B7 - regression coefficient of the EAP variable; • EAP - economically active population; • B8 - regression coefficient of the HI variable; • HI - expenditures on health insurance in thousands of euro; • B9 - regression coefficient of the C variable; • C - a number of malfeasances; • Bio - regression coefficient of the MW variable; • MW - a weight of municipal waste per capita in tons; • Bu - regression coefficient of the RN variable; • RN - a length of newly built expressways in kilometres. 3 Results and Discussion The estimated regression coefficients of the variables involved in the model and their related p-values are displayed in the Table 1 Regression Model Attributes. Table 1 Regression Model Attributes Variable Estimated regression coefficient p-value Statistical significance ßo 1.0642 . 104 2.133 . 10"4 =(==(==(==(= PS 2.5053 . 105 4.156 . 10"11 =(==(==(==(= TEEPTP 1.2047 . 107 0.0268961 =(==(= MHI 14.426 0.0024199 =(==(==(= EP 32.201 0.0703163 =1= TEEP 45.548 0.0355422 =(==(= UR 241.16 6.615 . 10"4 =(==(==(==(= EAP 0.082 1.449 . 10"6 =(==(==(==(= HI 0.018 8.405 . 10"7 =(==(==(==(= C 0.2253 3.356 . 10"7 =(==(==(==(= MW 0.0092 0.1171078 RN 15.813 0.0201003 =(==(= Source: Own elaboration by the authors 178 Legend for the Table 1 - the boundaries for the statistical significance levels are set to be as follows: • the first level - marked * - p-value is lower than 0.1 including but higher than 0.05; • the second level - marked ** - p-value is lower than 0.05 including but higher than 0.01; • the third level - marked *** - p-value is lower than 0.01 including but higher than 0.001; • the fourth level - marked **** - p-value is lower than 0.001 including. There are six positively influencing variables and five variables which affect negatively the explained indicator quantified in the regression model. Tertiary educated employed population, households' income, criminality, economic activity of population, health insurance, and municipal waste increase the explained indicator, whilst patents, distribution of population, unemployment rate, road network and employed population decrease the main estimated variable. The biggest influence in a positive way tertiary educated employed population has and in a negative way patents have. The most statistically significant variable is distribution of population, but there are four others in the highest significance level where criminality, households' income, economic activity of population and unemployment rate belong. The second level of statistical significance is reached by health insurance. Road network, tertiary educated employed population, and patents are significant at the third level. From all the statistically significant variables, the less important employed population is. The sole variable is insignificant according to the common threshold of ten-per-cent level - municipal waste. Although, there is to note that its p-value only very slightly oversteps this boundary and stands at level of 0.1171. Therefore, it can be considered to be statistically significant enough for this model. To sum it all up, all the comprised variables are significant in terms of purpose of this model and their regression coefficients are interpretable. The regression model equation looks like: GDP = 10642 + 250530 PS +12047000 TEEPTP + UA26MHI + 32.201 EP + + 44.548TEEP + 241A6UR + 0.082EAP + 0M8HI + 0.2254C + 0.0092MW + (2) + 15M3RN The regression coefficients determine the final value of the explained indicator. The basement is induced by a constant value of 10642. Subsequently, the biggest positive impact is caused by tertiary educated employed population. Each increment of this variable by one unit - one person in this case - brings an addition of 44.55 to the explained variable. If households' income rises by one unit - one euro, it increases gross domestic product by 14.43. Criminality effects also positively and this influence can be considered paradoxically. Each increment of criminality by one unit - by one malfeasance - causes a 0.2254 growth of the explained variable. Economic activity of population measured as a number of economically active inhabitants enhances gross domestic product by 0.082 per every newly economically active person. Each further euro paid in a form of health insurance means enlarge of the explained variable by 1.802 . 10"5. The last positive indicator is municipal waste, whose increment by one unit - in this case one ton - causes a 9.221 . 10"3 increase. The biggest negative impact patents have - at level of -12047100, so each increment of this ratio increases gross domestic product by this value. Distribution of population decreases the explained variable by 250526. As it is calculated as a population share of the particular self-governing region, the higher number it is, the higher decrease happens. Therefore, the most uniform distribution of population among all the self-governing regions is appropriate. Reduction of unemployment rate by one unit - by one per cent - causes a 0.0241 cut to the explained variable. Road network as the influencing indicator behaves very surprisingly. Its increase brings diminish of gross domestic product. Each additional unit - a newly built kilometre of expressway - brings reduction by 0.1581. This unexpected point probably lies in the fact that a number of new-built expressways does not involve all the types of 179 motorways built. We had to take into consideration just right this number, because aggregate sum of all the motorways also comprise roads, which are not available in full profile and by this fact they do not fulfil the factual requirements to become expressways. The least negatively influencing variable is employed population. Everyone additional employed inhabitant decreases gross domestic product by 3.22 . 10"4. Again, this is not expectable outcome, as employment should be one of driving force of economy. The explanation is little bit structured. Firstly, influence of this indicator is very weak and moves on the fourth decimal point. Secondly, unemployment rate effects gross domestic product in a positive way, so a true share of employed people gives the right outcome. Thirdly, employed population does not involve all the working inhabitants, because person employed by work performance agreement outside employment relationship does not count the number of this indicator. Hence, we declare it as a component of the regression model despite its influence. 4 Conclusions The paper discusses the structural aspects of competitiveness at the level of self-governing regions of the Slovak Republic. The aim of this paper is to identify the impact of monitored variables on gross domestic product per capita using the dataset reflecting the figures of the self-governing regions and to confirm influence of these dimensions on the creation of gross domestic product. The purpose of this analysis is to make a basic scheme for the further regional and national benchmarking. Most countries in the world have intended to become more dynamic and competitive economy capable of sustainable economic development. A uniform approach to define and measure competitiveness has not been introduced so far. Nevertheless, increasing competitiveness of the particular region is often presented as one of the fundamental objectives of regional policy. Whatever incorrect assessment of competitiveness of a country may have negative impacts on its social and political stability. On the contrary, negative assessment of the competitiveness can boost economy, mobilise the internal reserves and use certain market gaps. It is able to possess not only economic stimulus, but also psychological influence. Then, this is reflected in the restoration of confidence in the capital markets as well as production growth rate, or other types of outputs with adequate positive consequences in the export increase, employment growth and the consequent growth of domestic consumption. As the executed regression analysis has shown, there is a plenty of indicators which can influence quantification of competitiveness - from the demographic factors through the socio-economic aspects to the infrastructure elements. All these factors interact altogether. Therefore, it is difficult to mark the most significant ones. Now, according to this analysis, they are recognisable only by their statistical significance. To determine it, it should be a question for further analysis. There are also other specialised types of analyses that suit to demonstrate not yet revealed linkages among the explored dimensions. In the recent years, issue of regional competitiveness have gained great popularity due to the fact that one of the accompanying effects of advancing globalisation is increasingly important role of regions in the economic development of a country. The annually published reports of the International Institute for Management Development and the World Economic Forum have achieved high support from government level or business sphere. Consequently, they are considered the most authoritative to evaluate competitiveness. However, it is very important to make analyses by the national team of researchers because they know the strengths and the weaknesses of their countries in detail and such a look is essential in creation of regional policies. 180 Acknowledgments This work is supported by the Scientific Grant Agency of the Ministry of Education, Science, Research and Sport of the Slovak Republic as part of the research project VEGA 1/0986/15 Proposal of the Dimensional Models of the Management Effectiveness of ICT and Information Systems in Health Facilities in Slovakia and the Economic-Financial Quantification of Their Effects on the Health System in Slovakia. References Delgado, M., Ketels, Ch., Porter, M., Stern, S. (2012). The Determinants of National Competitiveness. National Bureau of Economic Research Working Paper 18249. Retrieved from: http://www.nber.org/papers/wl8249.pdf. European Commission Enterprise and Industry Directorate-General - Unit B4, Economic Analysis and Evaluation (2010). European Competitiveness Report 2009. Luxembourg: Publications Office of the European Union. Retrieved from: http://bookshop.europa.eu/en/european-competitiveness-report-2009-pbNBAK09001/downloads/NB-AK-09-001-EN- C/NBAK09001ENC_002.pdf?FileName = NBAK09001ENC_002.pdf&SKU = NBAK09001ENC_P DF&CatalogueNumber=NB-AK-09-001-EN-C. Fontagne, L., Santoni, G. (2014). Theoretical and policy aspects of competitiveness at different aggregation levels. Retrieved from: http://mapcompete.eu/wp-content/uploads/2015/05/D5_l_ComSurvey.pdf. Frank, K. (2014). Analýza vybraných štrukturálnych charakteristík slovenských regiónov. Ekonomický ústav Slovenskej akadémie vied Working Papers 68. Retrieved from: http://ekonom.sav.sk/uploads/journals/265_wp68frankl.pdf. Hošoff, B., Hvozdíková, V. (2009). Analýza konkurencieschopnosti krajín V4 z pohľadu atraktivity pre PZI. Ekonomický ústav Slovenskej akadémie vied Working Papers 21. Retrieved from: http://ekonom.sav.sk/uploads/journals/WP21.pdf. Ilzkovitz, F., Dierx, A., Galgau, O., Leib, K. (2008). Trade Performance and Structural Competitiveness Developments in the Euro Area: Are Member States Equipped to Meet the Globalisation Challenges of the 21st Century? Retrieved from: https://research.stlouisfed.org/conferences/integration/Galgau-paper.pdf. Lábaj, M. (2014). Štrukturálne aspekty ekonomického rozvoja: Slovenská ekonomika v globálnych súvislostiach. Bratislava: Ekonomický ústav Slovenskej akadémie vied. Retrieved from: http://www.ekonom.sav.sk/uploads/journals/255_labaj-monografia-2.pdf. Martin, R. L. (2005). A study on the factors of regional competitiveness. University of Cambridge. Cambridge. Cambridge: Cambridge Econometrics, University of Cambridge. Retrieved from: http://ec.europa.eu/regional_policy/sources/docgener/studies/pdf/3cr/competitiveness. p df. Morovská, L, Butoracová Šindleryová, L, Gburová, J. (2009). Analýza regionálneho rozvoja vybraných krajov SR s podporou marketingového inštrumentária a zvyšovanie konkurencieschopnosti regiónu. In: Zborník vedeckých prác Katedry ekonómie a ekonomiky ANNO 2009. Retrieved from: http://www.pulib.sk/elpub2/FM/KotuliclO/pdf_doc/16.pdf. Mothe, 1, Paquet, G. (1998). Finance and the Technology-Trade Nexus. Technology in society, vol. 20(4), pp. 441-468. Retrieved from: http://www.gouvernance.ca/publications/98-08.pdf. Petr, P., Křupka, 1, Provazníkova, R. (2010). Statistical Approach to Analysis of the Regions. In: 10th World Scientific and Engineering Academy and Society International 181 Conference on Applied Computer Science. Retrieved from: http://www.wseas.us/e-library/conferences/2010/Ja pan/ACS/ACS-43.pdf. Petr, P., Křupka, 1, Provazníkova, R. (2011). Multidimensional Modeling of Cohesion Regions. International Journal of Mathematical Models and Methods in Applied Sciences, vol. 5(1), pp. 150-158. Retrieved from: http://www.naun.org/main/NAUN/ijmmas/19-659.pdf. Porter, M. E. (1999). The Competitive Advantage of Nations. New York: Free Press. Provazníkova, R., Křupka, J., Kašparová, M. (2009). Modelování konkurenceschopnosti regionů v podmínkách globalizace. Scientific Papers of the University of Pardubice -Series D, Faculty of Economics and Administration, vol. 14, pp. 113-124. Storper, M. (1997). The Regional World: Territorial Development in a Global Economy. New York: The Guilford Press. Šikula, M. (2006). Konkurenceschopnost' slovenskej a českej ekonomiky - stav a perspektívy. In: Zborník príspevkov zo slovensko-českej vedeckej konferencie konanej pod záštitou podpredsedu vlády Slovenskej republiky pre európske záležitosti, ľudské práva a menšiny Pála Csákyho v Bratislave 14. marca 2006. pp. 264. Šikula, M. (2010). Krízové a pokrízové adaptačné procesy a nové nároky na konkurenceschopnost'. In: Zborník príspevkov zo slovensko-českej vedeckej konferencie konanej pod záštitou podpredsedu vlády Slovenskej republiky pre európske záležitosti, ľudské práva a menšiny Pála Csákyho v Bratislave 14. marca 2006. 182 Keynesian Model in Small and Medium Enterprises Development: Puzzling Case of Russian Regions Maria Ginzburg1, Nadezhda Yashina1, Elena Ivanova2 1 Lobachevsky State University Institute of Economics and Entrepreneurship, Finance and Credit Department Gagarin ave., 23, 603950 Nizhny Novgorod, Russia E-mail: ginzburg@iee.unn.ru, yashina@iee.unn.ru 2The Russian Presidential Academy of National Economy and Public Administration Nizhny Novgorod Institute of Management, Finance and Credit Department Gagarin ave., 46, 603950 Nizhny Novgorod, Russia E-mail: ivanova457@yandex.ru Abstract: This study explores the potential effects of Keynesian demand stimulation for small and medium-size businesses in Russian regions. The authors tested the following hypotheses: (1) is there a connection between small and medium-size business development in any given region of Russia and regional output (Gross Regional Product -GRP); (2) the higher the share of electric power consumption in GRP of a given region, the more regional product is created in this region and the higher the share of this regional product can be redistributed by means of small businesses. To test these hypotheses, the authors developed three indices related to the development of small and medium-size enterprises and the GRP, produced in the provinces of the Russian Federation. One index (SMEDI) summarizes the level of development of small and medium-size businesses in a region; another (GRPI) - the level of Gross Regional Product, produced in a given region; and finally (EPCI) - the level of electric power consumption in the GRP of a corresponding region. Correlation analysis was used to analyze data for 83 federal subjects of the Russian Federation for 2010-2014. The results indicate a moderate relationship between SMEDI and GRPI, and no relationship between SMEDI and EPCI. We conclude that the level of development of SMEs in the regions of Russia in 2010-2014 depends on the level of GRP to a certain degree. The level of development of SMEs is directly proportional to increases in GRP. But Keynesian model of demand stimulation, if applied to the Russian economy, stumbles upon the problems of fiscal federalism and the excessive centralization of economic governance, and this fact does not create effective stimulus for economic growth and small business development in Russia. JEL codes: G38, H77, L26, 052, Rll 1 Introduction Regional- and local-level entrepreneurship is an important engine of sustainable economic development and growth in the current knowledge-based economy (Baumol, 2004; Grudzinsky and Bedny, 2013; Grossman and Helpman, 1994; Shane, 2000; Kirzner, 1997). Specifically in the case of Russia, the share of revenue in the form of taxes collected from small businesses, according to the Treasury of the Russian Federation (RF) for 2014, is rather low - 3.61% (OFFICIAL STATISTIC, 2015). But the number of people employed in small and medium-size businesses in Russia has reached 24% of the economically active population (Small and medium-size businesses in Russia, 2014), which signals the importance of this segment, both for the economy and for the society as a whole. Following Doh and Kim (2014), Bateman (2000), this study acknowledges a crucial role played by small- and medium-size enterprises (SME) around the world, providing sources for most new jobs and platforms. Some Russian scholars have observed that the existing model of fiscal federalism does not create incentives for SME growth because it does not leave the regions with significant discretionary spending capacity and authority (Asadullina et al., 2015). Karpov et al. (2015), Yashina et al. (2016) identify several key problems in the mechanism of small and medium-size business development. They include the asymmetry of subsidies in the federal budget, 183 delays with the receipt and distribution of subsidies at the regional level, faulty program evaluation of the state SME support, as well as ineffective methodology for grant awards. At the same time, Tereshchenko and Sherbakov (2015) report significant positive correlation between major institutional indicators, on the one hand, and the indicators of economic development in the regions, on the other. They determined that the strengthening of the institutions responsible for the protection of the rights of ownership and the promotion of entrepreneurial activity can produce important synergies for growth in separate regions and at the national level. The purpose of our study is to explore the possibility and the potential effectiveness of the Keynesian model implementation of demand stimulation for small and medium-size business development in the Russian regions. Keynesian model is interesting to scientists worldwide. Murota and Ono (2015) analyze possibilities of demand-stimulation, suggest that countries that have lapsed into long-run stagnation should expand government spending that directly creates employment in order to reduce the deflationary gap. Costabile (2009) argue that the structural reform and the technical provisions proposed by the "Keynes Plan" may provide useful remedies to curb both inflationary and deflationary pressures on the world economy. Ostapenko (2014) claimed that the multiplier value tends to be the highest in Keynesian structural macro-econometric models. Hernandez de Cos and Moral-Benito(2013)provide some evidence that output growth might affect the fiscal tightening process so that fiscal consolidations are not exogenous to economic growth. Once they allow for feedback effects from economic growth to fiscal adjustments, they find that expansionary effects disappear and recover the typical Keynesian effect of fiscal adjustments. 2 Methodology and Data In our study, we relied on the Keynes research (1936), he supposed that demand stimulating among consumers, and inside the industry, creates its multilevel distribution structure. The industrial production that is present in the region, stimulates consumer demand, and it supports small and medium-sized businesses, and allows them to grow in the region. We also used the results obtained by us in the pilot study (Yashina et al., 2016), where we revealed the dependence of the level of development of small and medium-sized businesses in the region from the level of financial conditions of this particular region. The better the financial status of the region is, the higher it the level of development of SMEs. Relying on the Keynes provisions (1936), we believe that small and medium-sized businesses realizes through a distribution function of finance. As a consequence, actively development is possible where industrial potential is. The industry creates the national income, which redistributes by SMEs. Therefore, we hypothesized that: (1) there is a link between the level of development of small and medium-sized businesses in particular Russian region and the level of gross regional product (GRP), created in the region. We believe that the higher the level of GRP of the region, the higher the level of development of SMEs in the region. Our second hypothesis connected with the fact that industry in the Russian Federation has a "heavy" structure, characterized by a high proportion of fixed assets and high level of energy consumption. So, 59% of fixed assets concentrated in "heavy" industries such as mining, manufacturing, power generation and distribution, etc. (OFFICIAL STATISTIC, 2015c). Accordingly, the GRP growth connected with increased electric power consumption. And regions, created more value added than others, consume more electricity. Therefore we hypothesized that: (2) the higher the proportion of electric power consumption in GRP of the region, the more regional product creates in this region. And the more regional product may be distributed through small businesses. That should entail an increase in the development of SMEs in the Russian regions. 184 To test our hypotheses, we relied on the methodology contained in our pilot study (Yashina et al., 2016). To operationalize the key variables, we started by constructing three measures that aggregated distinct explanatory factors or, in the case of the dependent variable, outcome indicators into unique indices. Aggregating the influence of several factors into a single index allows obtaining parsimonious measure for the analyzed phenomena. For example, the level of SME development in the regions of the Russian Federation is a complex economic category, which cannot be sufficiently described by any one indicator. Altman (1968) in his work suggests that combinations of ratios can be analyzed together to remove possible ambiguities and errors observed in the earlier studies. The strength of regional SMEs has been estimated by examining changes in the number of people employed in this sector as well as changes in the share of SME tax contributions to the total tax income of a region: the higher levels of SME development are reflected in the higher proportion of SME taxes in the tax income of the federal subject. The absolute numbers in official records are not comparable and fail to take into account changes over time. Thus, the second step of the study entailed index standardization (normalization) in order to obtain comparable estimates. As the third step of the study, we used the tools of linear correlation analysis to detect possible relationships and explore their characteristics. The indicators for composite indices have been selected from a larger set of economic indicators, following the principles elaborated in Ginzburg et al. (2015). We relied on: the principle of data reliability - the use of credible statistical sources to minimize the probability of incorrect information; the principle of data relevance - the use of indicators that are directly (rather than indirectly) related to the subject of the study; the principle of versatility -the use of indicators that characterize distinct dimensions of the phenomenon in order to avoid "lopsided" analysis. In the study, we used the data of the Federal State Statistics Service, Federal Treasury, Ministry of Economic Development of the Russian Federation for the period 2010-2014 years for all 83 federal subjects of the Russian Federation (excluding the affiliated territories of the Republic of Crimea and Sevastopol for 2014). In accordance with these principles, the indices were computed using the following sets of statistics: The SME Development Index (SMEDI) comprises: the share of the tax collected from different forms of SMEs in the tax revenue of the region, (%); total employed labor force in the Russian Federation, thousands of people; the number of employed in SMEs by region, thousands of people. The index is calculated by the following formula: (NESME\ CMFm _lr (TA)jn t { SPE Ijn ^ , ..n *MbUlJn - - t™uf((rA);) + ma^N^E^h U) where: TA - the share of total income tax amount collected from different forms of SMEs; NESME - number of employees in SMEs in the region of the Russian Federation; SPE- the region's population, employed in Economics, T - a period of time (from 2010 to 2014), j - the number of the relevant region in the list of regions of Russian Federation, n - the number of the year. SMEDI changes in the interval from 0 to 1. SMEDI growth means increase the contribution of small and medium-sized enterprises in Gross Regional Product (GRP) and/or the growth of employment in small and medium business compared to similar indicators in other years of the period analyzed. The growth index is considered to be a positive factor, the decline is considered to be a negative factor. Next we computed two value-added indexes: Gross Regional Product Index and Energy Consumption Index. When formed these indexes we did not include data on the mining and transportation of minerals, oil extraction, we included only the electricity 185 consumption, because resources are mines in several regions of the country, but electricity consumes everywhere. The Gross Regional Product Index (GRPI) comprises: gross regional product in current prices, (mln. rubles).We believe that the higher the GRP level in the region is, according to the Keynesian model, the higher the level of the development of SMEs in the region should be. The index is calculated by the following formula: GRPIjn = m^GKPjy (2) where GRP - Gross Regional Product, T - a period of time (from 2010 to 2014), j - the number of the relevant region in the list of regions of Russian Federation, n - the number of the year. GRPI changes in the interval from 0 to 1. The higher the value in a specified interval in the GRPI, the higher is the possibility for Value-Added creation which is demonstrated in given region of the Russian Federation. The growth index is considered to be a positive factor, the decline is considered to be a negative factor. The Electric Power Consumption Index (EPCI) comprises: the share of the energy consumption in the corresponding region, in the gross regional product, (%). This index will allow us to identify the relationship between the level of SME development and the level of electric power consumption in the region. We believe that the higher the industrial potential of the region, the higher is the share of electric power consumption in GRP. We formed this index to reverse the regional disproportions presented in the previous index (for example, in the year 2014 gross regional product of city of Moscow amounted to 21.8% from GRP, produced by all the regions of the Russian Federation together. While the level of electric power consumption in the city was 5.18% of the total consumption of the Russian Federation) (OFFICIAL STATISTIC, 2015d). The index is calculated by the following formula: Zr^ljn- max((EPC i \~>J Where EPC - electric power consumption, mln. kilowatt-hours, GRP - Gross Regional Product, T - a period of time (from 2010 to 2014), j - the number of the relevant region in the list of regions of Russian Federation, n - the number of the year. EPCI changes in the interval from 0 to l.The higher the value of EPCI in a specified interval, the higher is the share of electric power consumption in GRP. Growth of this index indicates that the power consumption increases in the region as a consequence of the industrial production growth. The growth index is considered to be a positive factor, the decline is considered to be a negative factor. When the indexes were formed, we explored the relationship between them. 3 Results and Discussion Table 1 reports the SME development index. A very low level of SME development characterizes the fiscal transfers-dependent regions, such as Dagestan (0,1926 in 2014),Ingushetia (0,0977) and Chechnya (0,1136). Table 1 The SME Development Index (SMEDI) for Years 2010-2014 (Fragment) 2010 2011 2012 2013 2014 City of Saint-Peters burg 0.3415 1 0.9433 0.9735 1 City of Moscow 0.4390 0.9427 1 1 0.9946 The Krasnodar Territory 0.5324 0.8962 0.8634 0.8766 0.9255 Republic of Dagestan 0.0848 0.1490 0.1560 0.1477 0.1926 Republic of Chechnya 0.0246 0.0648 0.0904 0.1009 0.1136 Republic of Ingushetiya 0.0554 0.0509 0.0478 0.0691 0.0977 Source: calculated by authors. Based on OFFICIAL STATISTICS (2015a,b,c,d), (2016) 186 It needs to be pointed out though that SMEDI increased over 2010-2014. More robust levels of SME development are observed in Krasnodar Territory (SMEDI2014 = 0.9255) and federal centers - city of Saint-Peters burg (SMEDI2014 = 1) and Moscow (0.9946). In these regions, too, the values of SMEDI grew from 2010 to 2014. Table 2 demonstrates that the largest GRP is in the Federal Center - city of Moscow, while industrial regions lagging behind it several times. Table 2 The Gross Regional Product Index (GRPI) for Years 2010-2014 (Fragment) 2010 2011 2012 2013 2014 City of Moscow 1 1 1 1 1 The Tumen Region 0,3942 0.4134 0.4336 0.4190 0.4043 Khanti-Mansi Autonomous Area 0.2354 0.2453 0.2535 0.2310 0.2206 Republic of Tatarstan 0.1196 0.1313 0.1347 0.1313 0.1305 Krasnoyarsk Territory 0.1260 0.1177 0.1109 0.1064 0.1111 Republic of Severnaya Osetia 0.0090 0.0086 0.0091 0.0100 0.010 Source: Calculated by authors. Based on OFFICIAL STATISTICS (2015a,b,c,d), (2016) Thus, Tumen region GRP Index in year 2014 was almost 2.5 times lower than has the Federal Center (the city of Moscow) (GfiP/2014 = 0.4043),Khanti-Mansi GRPI laged in 4.5 times, GRP index of Republic of Tatarstan - in 7.7 times. While the listed regions are the major industrial regions of Russia. In Table 3 we present data on The Electric Power Consumption Index for several Russian regions. Table 3 The Electric Power Consumption Index (EPCI) for Years 2010-2014 (Fragment) 2010 2011 2012 2013 2014 Republic of Khakassia 1 1 1 1 1 The Irkutsk region 0.5545 0.6073 0.5812 0.6145 0.6064 The Kranoyarsk Territory 0.2814 0.3031 0.3366 0.3697 0.3655 The Chelyabinsk region 0.3000 0.3197 0.3191 0.3559 0.3596 The Tumen region 0.1529 0.1525 0.1499 0.1705 0.1807 City of Moscow 0.0346 0.0363 0.0371 0.0406 0.0421 Source: Calculated by authors. Based on OFFICIAL STATISTICS (2015a,b,c,d), (2016) The regions with the highest GRPI in 2014, have are quite low The Electric Power Consumption Index. The city of Moscow (EPCI2014 = 0.0421), The Tumen region (EPCI2014) = 0.1807). The relationship between the Small and Medium-sized Enterprises Development Index (SMEDI) and Gross Regional Product Index (GRPI), between SMEDI and Electric Power Consumption Index (EPCI) was tested by means of linear correlation analysis. The correlation coefficients were computed by the following formulas: _ *Lj=1{SMEDl-SMEDr){GRPl-GRPl} . . rSMEDl-GRPl = [ ,,. ,,. i (4) S™_ ^SMEDI-SMEDI) 2 J,J= 1(grp/ - GRPI)2 _ T,j=1(SMED1-SMED1)(EPC1-EPC1) ._. ^SMEDI—EPCI ~ i , V^J T,jL t(SMEDI-smed1)2J^=1{epc1-epc1)2 where SMEDi-the index, value of the SME development level, grpi- the index, value of the Gross Regional Product level, EPCI - the index, value of Electric Power Consumption level, all three averaged over the set of regions; N - the number of studied regions. Calculations according to the formulas (4), (5) yielded the following results (Table 4): 187 Table 4 Correlation Analysis between SMEDI - GRPI, SMEDI - EPCI for Years 2010-2014 2010 2011 2012 2013 2014 rSMEDl-GRPl 0.194918 0.419228 0.446969 0.442711 0.414872 rSMEDl-EPCl -0.114418 -0.077492 -0.056294 0.012258 -0.011530 Source: Calculated by authors. Based on OFFICIAL STATISTICS (2015a,b,c,d), (2016) We can see that the correlation between SME development and Gross Regional Product (^medi-grpi)gets notably stronger, starting with 2012. Low levels of correlation between the two indices in 2010 can arguably be attributed to the lingering effects of the 2008-2009 crisis. The correlation coefficients between SME development and Electric Power Consumption (rsMEDi-EPci) indicate there is no connection between the two indices. This can be caused by a number of reasons, for instance, the absence of a relationship or non-linear connection between the variables. To check for the lagged effects (one year) of GRP growth on SME development, energy consumption on SME development, we calculated lagged correlations in which the SMEDI index for year n is correlated with the GRPI/EPCI indexes for the preceding year (n-1). The results are as follows (Table 5). Table 5 Lag Correlation for SMEDI - GRPI, SMEDI - EPCI for Years 2010-2014 SMEDl2011 — GRPI2oio SMEDI2012 - GRPI2011 SMEDI2013 — GRPI2012 SMEDI2014 — GRPI2oi3 r'SMEDI-GRPI 0.416868235 0.445839414 0.443910586 0.414264501 SMEDI2011 - EPCI2010 SMEDI2012 - EPCI2011 SMEDI2013 - EPCI2012 SMEDI2014 - EPCI2013 rSMEDl-EPCl -0.07391109 -0.087640102 0.010378072 -0.00756268 Source: Calculated by authors. Based on OFFICIAL STATISTICS (2015a,b,c,d,), (2016) There seems to be evidence of lagged effects for all years: a higher gross regional product of a region in a year n corresponds to a more robust SME development in the following year. As we can see, correlation results as well as lagged correlation support hypothesis (1) - SME development in a certain region of Russian Federation depends on level of GRP of this region, that support findings of our pilot study (Yashina et al., 2016). As for energy consumption effect on SME development, we conclude that there is no strong evidence of lagged effect: energy consumption does not translate into higher levels of SME development in the following year. Correlation results do not support hypothesis (2) - SME development in Russia's regions during the period from 2010 to 2014 years do not associated with an increase in energy consumption in the Russian regions. Regression analysis involves the effect of approximation of real dependency values of some analytic function that is selected in accordance with several criteria (Fisher, etc.), the function must be a good match with the real dependency. This, in the face of a lack of data and their inaccuracies, is rather difficult to describe. The consequence of small correlation coefficients is the graph, more similar to stain than graph- this geometry is hard to approximate. To talk about regression analysis makes sense, starting with the correlation coefficients > 0.7. Therefore, at this stage, we cannot use this tool, but it is the goal of our future work on this topic. 4 Conclusions Our test supports hypothesis (1) - SME growth appears to respond to the Gross regional product of the region increase, as proposed by Tereshchenko and Sherbakov (2015), Bondareva and Zatrochova (2014). And totally reject hypothesis (2) - the higher the level of electric power consumption in the region is, so most likely, the higher industrial 188 production in the region, and, as a consequence - the level of development of SMEs. Relationship between the level of SMEs development and GRP growth, is moderate. We believe that this situation is a consequence of the current model of fiscal federalism in Russia, when the cash flows generated in the regions redistribute in favor of the Federal Centre at the expense of regions. As well as models of centralization of the economy when companies working in Russia's regions, have registrations for tax purpose in Moscow, and, accordingly, pay taxes to the Federal Center, leaving regions where their main activities are, alone with their unresolved social problems. In such circumstances, there is no direct relationship between the index of small business development and value creation indexes (in our case are Gross Regional Product Index and Electric Power Consumption Index). The consequence is the attraction and the concentration of capital in the metropolitan area and regional underdevelopment, spreading to the most sensitive and flexible element-small business. This is consistent with the findings obtained by Asadullina et al. (2015) that the Russian model of fiscal federalism does not create effective incentives to economic growth. This model belongs to the class of low competition models, distinguished by the fact that all levels of the budgetary system, both vertically and horizontally, are in hard-deterministic relationships, not leaving the territories to compete among themselves, thereby extinguishing the incentives for economic growth in territories. We share the view of Aleshin et al. (2013) that it needs to change the mechanism of fiscal redistribution of tax revenues towards its decentralization. Now the Russian capital is not distributed in Russian small business area, tend to consolidation, and can be considered as an integration part of the European financial system, aimed at expansion. Thus, Keynesian demand stimulation model for small and medium business development in Russia's regions faces with features of Russian fiscal federalism, that does not allow it to be implemented in full size. References Aleshin, V. A., Ovchinnikov, V. N., Chelysheva, E. A. (2013) Fiscal federalism: problem and development prospect today. Journal of Economic Regulation, vol. 1, Iss.4, pp. 115-123. Altman, E. I. (1968). Financial Ratios, Discriminant Analysis and the Prediction of Corporate Bankruptcy. The Journal of Finance, vol.23, Iss.4, pp. 589-609. Asadullina, A. V., Asylguzhin, B. V., Truhov, A. I. (2015). Competitive fiscal federalism and its role in the economic development of the country. Finance and Credit, vol. 13 (637).pp. 44-53. Bateman, M. (2000) Neo-Liberalism, SME Development and the Role of Business Support Centres in the Transition Economies of Central and Eastern Europe. Small Business Economics, vol. 14, Iss. 4, pp. 275-298. Costabile, L. (2009) Current global imbalances and the Keynes Plan: A Keynesian approach for reforming the international monetary system. Structural Change and Economic Dynamics, vol. 20, Iss.2, pp. 79-89. Doh, S., Kim, B. (2014) Government support for SME innovations in the regional industries: The case of government financial support program in South Korea. Research Policy, vol. 43 (9), pp. 1557-1569. Ginzburg, M. Y., Yashina, N. I., Litovsky, I. A. (2015). Standartised integro-differential index as an indicator, that uniquely characterizes socio-economic situation of a state at the world stage. Economics and entrepreneurship, vol. 5, Iss. I, pp. 360-368. Hernandez de Cos, P., Moral-Benito E. (2013). Fiscal Consolidations and Economic Growth. Fiscal Studies, vol. 34, Iss. 4, pp. 491-515. Keynes, J. M. (1936) The general theory of employment, interest and money, New York: Palgrave Macmillan. 189 Murota, R., Ono, Y. (2015) Fiscal policy under deflationary gap and long-run stagnation: Re interpretation of Keynesian multipliers. Economic Modelling, vol.51, pp. 596-603. OFFICIAL STATISTICS (2015). Labor force. Federal Statistic. Retrieved from: http://www.gks.ru/wps/wcm/connect/rosstat_main/rosstat/ru/statistics/wages/labour_fo rce/#. OFFICIAL STATISTICS (2015a). National Account. Gross Regional Product. Retrieved from: www.gks.ru/wps/wcm/connect/rosstat_main/rosstat/ru/statistics/accounts/. OFFICIAL STATISTICS (2015b). Consolidated information on the performance of the budgets of the regions of the Russian Federation. Retrieved from: http://old.roskazna.ru/byudzhetov-subektov-rf-i-mestnykh-byudzhetov/. OFFICIAL STATISTICS (2015c). Fixes Assets. Retrieved from: www.gks.ru/wps/wcm/connect/rosstat_main/rosstat/ru/statistics/enterprise/fund/. OFFICIAL STATISTICS (2015d). Manufacturing Output. Retrieved from: http://www.gks.ru/wps/wcm/connect/rosstat_main/rosstat/ru/statistics/enterprise/indust rial/. OFFICIAL STATISTICS (2016). Materials of the Federal Small and Medium Enterprises web-site. Retrieved from: http://smb.gov.ru/statistics/officialdata. Ostapenko, V. M. (2014). Fiscal Multipliers: Theory and Empirical Estimates. Economics and Entrepreneurship, vol. 5-2 (46-2), pp. 127-134. SMALL AND MEDIUM-SIZED BUSINESSES IN RUSSIA (2014). Statistical Digest. Federal service of State Statistics (Rosstat). Moscow. Yashina, N., Ginzburg, M., Emelyanova, O., Litovsky, I. (2016). Small and Medium Enterprise Development in Russia: Public Expenses - Are they effective? Proceedings of the 20th Scientific Conference "Current Trends in Public Sector Research",. Brno, Masaryk University, Faculty of Economics and Administration, Department of Public Economics, pp. 212-223. 190 Pension-related Application of the Cohort Life Table Ján Gogola1, Ondřej Slavíček2 1 University of Pardubice Faculty of Economics and Administration, Institute of Mathematics and Quantitative Methods Studentská 84, 532 10 Pardubice, Czech Republic E-mail: jan.gogola@upce.cz 2 University of Pardubice Faculty of Economics and Administration, Institute of Mathematics and Quantitative Methods Studentská 84, 532 10 Pardubice, Czech Republic E-mail: Ondrej.Slavicek@upce.cz Abstract: Longevity risk, the risk that people will live longer than expected, weighs heavily on those who run pension schemes and on insurers that provide annuities. Hence the prediction of future mortality rates is an issue of fundamental importance for the insurance and pensions industry. Our analysis focuses on mortality at higher ages (65-95), given our interest in pension-related applications where the risk associated with longer-term cash flow is primarily linked to uncertainty in future rates of mortality. The Lee-Carter model became one of the most applied models and it is used to forecast age-specific death rates. The main goal of this paper is to apply the Lee-Carter model to construct the so-called "cohort life tables" for calculation of a 30-year annuity to a person aged 65 in 2015. We use data on deaths and exposures for the Czech Republic from the Human Mortality Database (HMD). The HMD provides evidence that life expectancy is increasing. We have shown that if the today rate of increase will continue, it will at age 65 concluded (after calculation) to increase the present value of pension liabilities in defined-benefit schemes about 5 % if we use cohort life table instead of period life table. Probability statements derived from the use of a single model and parameter set should be treated with caution. Hence, there is a need for awareness of model risk when assessing longevity-related liabilities. Key words: longevity risk, annuity, stochastic mortality, life table, Lee-Carter model JEL Classifications: C53, G22, J11, J32 1 Introduction Benjamin Franklin said: "In this world nothing can be said certain, except death and taxes." The death is certain, but the timing is much less certain. The mortality of the population in developed countries has improved rapidly over the last thirty years and this has important financial implications for the insurance industry, since several important classes of liability are sensitive to the direction of future mortality trends. This uncertainty about the future development of mortality gives rise to longevity risk. Longevity risk, the risk that people live longer than expected, weighs heavily on those who run pension schemes and on insurers that provide annuities. The risk that the reserves established for the payment of benefits (retirement, widowhood, orphan hood, disability, dependency,...) are inadequate for that purpose if they are based on life tables (or mortality tables) with lower survival hypothesis than real. Longevity risk plays a central role in the insurance company management since only careful assumptions about future evolution of mortality phenomenon allow the company to correctly face its future obligations. Longevity risk represents a sub-modul of the underwriting risk module in the Solvency II framework. The most popular and widely used model for projecting longevity is the well-known Lee-Carter model. This paper follows on articles Gogola, J. (2014), Gogola, J. (2014a), Gogola, J. (2015), Jindrová, P., Slavíček, O. (2012), Pacáková, V., Jindrová, P. (2014) and Pacáková, V., Jindrová, P., Seinerová, K. (2013). They deal with the development and the prediction of 191 life expectancy in selected European countries (Czech Republic, Slovakia, Finland and Spain) by applying Lee-Carter model and the Quantification of Selected Factors of Longevity. Most stochastic mortality models are constructed in a similar manner. Specifically, when they are fitted to historical data, one or more time-varying parameters (x^) are identified. By extrapolating these parameters to the future, we can obtain a forecast of future death probabilities and consequently other demographic quantities such as life expectancies. They are important for quantifying longevity in pension risks and for constructing benchmarks for longevity-linked liabilities. The main goal of this paper is to apply the Lee-Carter model to construct the so-called "cohort life tables" and use them for calculation of a 30-year annuity to a person aged 65 in 2015. 2 Methodology and Data We use data of the total population, males and females' deaths and exposure to risk between 1950 and 2014 for the Czech Republic (CR) from the Human Mortality Database (www.mortality.org). We consider the restricted age range from 0 to 95. Let calendar year t runs from exact time t to exact time t+1 and let dxtbe the number of deaths aged x last birthday in the calendar year t. We suppose that the data on deaths are arranged in a matrixD - (dx,)■ In a similar way, the data on exposure are arranged in a matrixec ={ext) where ext is a measure of the average population size aged x last birthday in calendar year t, the so-called central exposed to risk. We suppose that (dxt) and (ext) are each na xny matrices, so that we have na ages and n years. We denote the force of mortality (or hazard rate) at exact time t for lives with exact age x byjuxt. The force of mortality can be thought as an instantaneous death rate, the probability that a life subject to a force of mortality juxt dies in the interval of time (t,t + dt) is approximately juxt -dt where dt is small. The force of mortality juxt for human populations varies slowly in both x and t and a standard assumption is that juxt is constant over each year of age, i.e., from exact age x to exact age x+1, and over each calendar year, i.e., from exact time t to exact time t+1. Thus, Mw+V=M„ forO (3) or in a matrix form m = , that means element-wise division in R. Ec We also consider the mortality rate qxt. This is the probability that an individual aged exactly x at exact time t will die between t and t+1. We have the following relation between the force of mortality and the mortality rate: 192 9x,t 1-exp \-fix+SJ+s ds -l-e^'. (4) Vo J We use the following conventions for our model: • The ocx,f5x coefficients will reflect age-related effects • The Kt coefficients will reflect time-related effects Our models are fitting to historical data. The Lee-Carter model was introduced by Ronald D. Lee and Lawrence Carter in 1992 with the article Lee, R. D., Carter, L. (1992). The model grew out of their work in the late 1980s and early 1990s attempting to use inverse projection to infer rates in historical demography. The model has been used by the United States Social Security Administration, the US Census Bureau and the United Nations. It has become the most widely used mortality forecasting technique in the world today. Lee and Carter proposed the following model for the force of mortality: ^gmxt = ax + /5x-Kt, (5) with constraints IX =1, (6) x=l I>,=0- (?) The second constraint implies that, for each x, the estimate for ax will be equal (at least approximately) to the mean over t of the log mxt. Let (Z> represent the full set of a parameters and the notation for juxt is extended to juxt(0), to indicate its dependence on these parameters. For our model the log-likelihood is: KfcD,E) = Y,(dxy^g[exrjUxtm-exrjUxt(^-log(dJ)), (8) and estimation is by maximum likelihood (MLE). By the equation (5) the log of the force mortality is expressed as the sum of an age-specific component ax that is independent of time and another component that is the product of a time-varying parameter Kt reflecting the general level of mortality and an age-specific component f5x that represents how rapidly or slowly mortality at each age varies when the general level of mortality changes. Interpretation of the parameters in Lee-Carter model is quite simple: exp(ax)'\s the general shape of the mortality schedule and the actual forces of mortality change according to overall mortality index Kt modulated by an age response f5x (the shape of the (5x profile tells which rates decline rapidly and which slowly over time in response of change inKt ). For practice the fitting of a model is usually only the first step and the main purpose is the forecasting of mortality. For forecasting-time series we use Random Walk with Drift. 193 The estimated age parameters, ax,(3x, are assumed invariant over time. This last assumption is certainly an approximation but the method has been very thoroughly tested in Booth, H., Tickle, L, Smith, L. (2005) and found to work. We assume that trend observed in past years can be graduated (or smoothed) and that it will continue in future years. By the Random Walk with Drift the dynamics of jct follows =Kt_,+e + et_l, (9) with i.i.d standard Gaussian distribution et~ N(0; o] ). Value at future time t+h can be written as h-l (10) which has Gaussian distribution N(at, + h-9\o~l ■ h). Hence the best point estimate for future value at time t+h is Kt + h-0, and the 95% confident interval is (Kt+h-d-\,96-a£-4h\Kt + h-d + \,96 a£-4h) ■ (11) 3 Results In Figure 1 we have plotted the maximum likelihood estimates for the parameters of the Lee-Carter model, using CR total population data, aged 0-95. Model fitting was done in R, which was also used for graphs (Figure 1). Note that estimated values for fix are higher at the lowest ages, meaning that at those ages the mortality improvements are faster. The decreasing trend in jct reflects general improvements in mortality over time at all ages. Figure 1 Estimated Parameters of the Lee-Carter Model for Population of CR 1950 1960 1970 1980 1990 2000 2010 Year Source: Own processing 194 We will now simulate the Kt up to 2060 according to equation (9). These results in case of the total population are plotted in Figure 2. The dashed curves in plot show the 2,5-th and 97,5-th percentile of the distribution of jct resulting in a 95 % confidence interval. By forecasted Kt we get the predictions for the force of mortality juxt -exp(ax + f5x ■ Kt), which lead us by equation (4) to mortality rates qX/t. Figure 2 Predicted jct for Total Population with 95 % CI and Few Simulations Kappa_t -200 Year Source: Own processing To avoid underestimation of the relevant liabilities we use dynamic mortality model. Cohort or dynamic life table provide a view on the future evolution of mortality rates and it implies the diagonal arrangement in a projecting life table (see Table 1). Table 1 Period Life Table vs. Cohort Life Table Qx,t 2014 2015 2016 2017 2018 2019 2020 65 0.014699 0.014505 0.014314 0.014125 0.013938 0.013754 0.013573 66 0.015832 0.015618 0.015406 0.015197 0.014991 0.014788 0.014587 67 0.017191 0.016954 0.016721 0.016491 0.016263 0.016039 0.015818 68 0.018574 0.018311 0.018051 0.017795 0.017543 0.017294 0.017048 69 0.020037 0.019744 0.019456 0.019172 0.018892 0.018615 0.018343 70 0.021675 0.02135 0.021029 0.020714 0.020403 0.020097 0.019795 71 0.023349 0.02299 0.022637 0.022289 0.021946 0.021609 0.021276 Source: Own calculations Finally by equations (12)-(15) we find the present values of the annuities such as term immediate annuity ax.n\, term annuity-due äx.n\. We will also consider annuities payable m-times per year. 195 aX:n\ = 2T=X- tPx (12) T| - "x:n| + 2m «S = ^:n|+^-(l-^- nP*) (UDD) (13) «x:n| = XU1^- tPx (14) «S = ^:n|-^-(l-^- nPX) (UDD) (15) (where (UDD) means the assumption of Uniform Distribution of Deaths). Take an individual aged 65 in 2015 (birth year = 1950) who wants to purchase a 30 years annuity. For calculation annuities first we use the Period table, which contains the last available mortality rates. In our case it is year 2014 (the second column of Table 1). Then we use the diagonal values (Cohort table) for the cohort aged 65 in 2015 (born 1950) who are still alive in year 2015+t. Table 2 gives present values of 30 years annuities for the individual aged 65 from the whole population of the Czech Republic with interest rate of 2 % p.a. In appendix (Table 3 and Table 4) we show present values of annuities separately for both genders. Table 2 Present Values of Annuities for the Total Population in the Czech Republic (x=65, r>=30, /=0,02) ax:n\ (12) Ux:n\ ..(12) Ux:n\ ax:n\ Period table 14.04 14.31 14.74 15.01 Cohort table 14.75 15.01 15.43 15.69 Relative change 5.01% 4.85% 4.69% 4.54% 2.5% 14.13 14.40 14.83 15.09 0.65% 0.61% 0.60% 0.57% 97.5% 15.34 15.60 16.01 16.27 9.26% 8.99% 8.64% 8.39% Source: Own calculations 4 Conclusions National governments and the WHO announce life expectancies of different populations every year. To financial institutions, life expectancy is not an adequate measure of risk, because all it does not give any idea about how mortality rates at different ages vary over time. On the other hand, indicators of longevity risk cannot be too complicated. An indicator that is composed by a huge array of numbers is difficult to interpret and will lose the purpose as a "summary" of a mortality pattern. We have presented stochastic models to analyse the mortality and shown how they may be fitted. Afterwards we can turn to the industry requirement to forecast future mortality. We have shown that if the today rate of increase will continue, it will at age 65 concluded (after calculation) to increase the present value of pension liabilities in defined-benefit schemes cca. 4,5-5 % if we use cohort life table instead of period life table. But forecasting of mortality should be approached with both caution and humility. Any prediction is unlikely to be correct. There is a need for awareness of model risk when assessing longevity-related liabilities, especially for annuities and pensions. The fact that parameters can be estimated does not imply that they can sensibly be forecast. 196 Such forecasting should enable actuaries to examine the financial consequences with different models and hence to come to an informed assessment of the impact of longevity risk on the portfolios in their care. Acknowledgments This research could be performed due to the support of the University of Pardubice student project grant no. SGS_2016_023 „Ekonomický a sociální rozvoj v soukromém a veřejném sektoru" (Faculty of Economics and Administration). References Booth, H., Tickle, L, Smith, L. (2005). Evaluation of the variants of the Lee-Carter method of forecasting mortality: a multi-country comparison. New Zealand Population Review, vol. 31, pp. 13-34. Gogola, J. (2014). Lee-Carter family of stochastic mortality models. In: Sborník 7. mezinárodní vědecká conference: Řízení a modelování finančních rizik. VŠB-TU Ostrava, pp. 209 - 2017. Gogola, J. (2014). Stochastic Mortality Models. Application to CR mortality data. Mathematical and Statistical Methods for Actuarial Sciences and Finance, Springer, pp. 113-116. Gogola, J. (2015). Comparison of selected stochastic mortality models, International Journal of Mathematical Models and Methods in Applied Sciences. INTERNATIONAL JOURNAL OF MATHEMATICAL MODELS AND METHODS IN APPLIED SCIENCES, vol. 9, pp. 159-165. Human Mortality Database. University of California, Berkeley (USA), and Max Planck Institute for Demographic Research (Germany). Retrieved from: www.mortality.org. Jindrová, P., Slavíček, O. (2012). Life expectancy development and prediction for selected European countries. In: 6-th International Scientific Conference Managing and Modelling of Financial Risk proceedings. VŠB-TU Ostrava, pp. 303-312. Lee, R. D., Carter, L. (1992). Modelling and forecasting the time series of U.S. mortality. Journal of the American Statistical Association, vol. 87, pp. 659-671. Pacáková, V., Jindrová, P. (2014). Quantification of Selected Factors of Longevity. In: Proceedings of the 2014 International Conference on Mathematical Methods in Applied Sciences (MMAS'14). Saint Petersburg State Polytechnic University, pp. 170 - 174. Pacáková, V., Jindrová, P., Seinerová, K. (2013). Mortality Models of Insured Population in the Slovak Republic. In: Proceedings of the 7th Professor Aleksander Zelias International Conference on Modelling and Forecasting of Socio-Economic Phenomena. 7. - 10. 09. 2013. Zakopané, pp. 99-106. R Development Core Team (2005). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna. Retrieved from: http://www.R-project.org. 197 Appendix Table 3 Present Values of Annuities for Males (x=65, n=30, /=0,02) in the CR _x-n\ ux:n\_ x:n\_x-n\ Period table 12.68 12.95 13.38 13.66 Cohort table 13.29 13.55 13.99 14.25 Relative 4.83% 4.65% 4.54% 4.38% change_ 2.5% 12.72 12.99 13.43 13.70 0.37% 0.33% 0.36% 0.32% 97.5% 13.85 14.11 14.54 14.80 9.26% 8.95% 8.69% 8.40% Source: Own calculations Table 4 Present Values of Annuities for Females (x=65, r?=30, /=0,02) in the CR _ „(12) ..(12) Ux:n\ ux:n\ CLx,m CLx.n\ Period table 15.14 15.41 15.83 16.10 Cohort table 15.92 16.18 16.58 16.84 Relative change 5.13% 4.98% 4.78% 4.64% 2.5% 15.27 15.53 15.95 16.21 0.83% 0.79% 0.76% 0.72% 97.5% 16.54 16.79 17.18 17.44 9.23% 8.97% 8.59% 8.36% Source: Own calculations 198 Influence of selected environmental factors on the efficiency of commercial insurers Eva Grmanová1, Peter Hošták2 Alexander Dubcek University of Trend n Faculty of Social and Economic Relations Studentská 3, 911 50. Trenčín, Slovak Republic E-mail: eva.grmanova@tnuni.sk 2Alexander Dubcek University of Trencin Faculty of Social and Economic Relations Studentská 3, 911 50. Trenčín, Slovak Republic E-mail: peter.hostak@tnuni.sk Abstract: Paper aims to analyze the specifics of the efficiency of Czech and Slovak commercial insurers arising from the impact of environmental factors. Models of data envelopment analysis (DEA) are used to analyze the efficiency of insurance companies. Efficiency is evaluated based on the efficiency score. DEA models allow for the evaluation of factors which in the short term are not under the influence of the analyzed entities. These are called environmental factors. Impact of environmental factors is studied based on an approach proposed by Charnes, Cooper and Rhodes (1981). This procedure divides the analyzed subjects into and in each subgroup efficiency scores are determined. Target values corresponding to the full efficiency are calculated for inefficient subjects. In the next step the efficiency score is expressed on the basis of above mentioned target values and a statistical hypothesis testing is used to determine the statistical significance of the difference between the efficiency scores in groups. This approach allows to compare differences in the efficiency of various groups of insurance companies. Keywords: environmental factors, efficiency, insurance companies JEL codes: G22, C52 1 Introduction The insurance market as one of the components of the financial market is of great importance in the national economy. It contributes significantly to the economic growth. It is a place where supply meets demand for various insurance products. Insurance is closely linked to the risk. Risk realization leads to loss. The size of the risk is the result of two characteristics -frequency and severity of damage. (Vávrová, 2015). Importance of insurance can be assessed both in terms of insured businesses and citizens, as well as in terms of the whole society. Insurance stabilizes the economic situation of the company in the event of an insurance claim. Costs in the critical period are distributed more evenly in comparison to the case of self-insurance. Availability of insurance to citizens stabilizes living standards and mitigates the impact of adverse consequences in the event of an adverse event. From the macroeconomic perspective, positive impact of the insurance industry for the national economy includes accumulation and redistribution of available funds, promotion and the development of the tertiary sector and the employment as well as support for the protection of values (Chovan, 2000). The insurance market in the European Union is undergoing significant changes. Establishment of a common regulatory environment in the EU and the process of creating a single insurance market of the EU aim towards development of a common integrated insurance space. Central and Eastern European countries are significantly lagging behind in this area (Brokešová, Ondruška and Pastoráková, 2015). Insurance market in Slovakia and the insurance market in the Czech Republic are part of this space. These national insurance markets are linked through their history and the mutual resemblance. Commercial insurance companies operate in these markets in an increasingly competitive environment and their results are affected by many factors. Management of insurance 199 companies directly coordinates some of these factors in a way that supports the achievement of desired business objectives. However, there are factors that influence the activities of insurance companies and are not under the control of management. These include purchasing power, unemployment rate and others. Thus the efficiency of insurance company is directly related to its own activities as well as the factors that the insurance company does not directly manage. The management of insurance companies must adapt to these factors and seek an optimal response in order to ensure maximum efficiency. 2 Methodology and Data In our analysis we will focus on the impact of selected environmental factors on the efficiency of commercial insurance companies. Our methodology builds upon established theoretical models that provide basis for the execution of the analysis and comparison of the efficiency of commercial insurance companies in Slovakia and the Czech Republic, with particular emphasis on the analysis of the impact of selected factors on the respective efficiency scores. We will assess the efficiency of insurance companies by score of technical efficiency, expressed in DEA models. Individual entities are within the DEA analysis referred to as Decision Making Units (DMUs). These models enable us to analyze the efficiency of transformation of multiple inputs to multiple outputs. DEA models are based on non-parametric approach. They use linear programming methods to construct envelopment of data. (Grmanová and Jablonský, 2009) They assess the efficiency score for each analyzed subject. Efficiency scores are calculated as relative values to the envelopment of data. DEA models are based on the assumption that all the DMUs to be analyzed have the same inputs and outputs, and they operate in the equal environment. In practice, the equal-environment assumption is not satisfied. Operation of insurance companies is affected by various factors, which cannot be directly affected by the insurance company managers. In our analysis, we refer to these factors as environmental factors. Thus, we understand the environmental factors as factors that are not directly influenced by management. Such factors can include macroeconomic as well as microeconomic indicators. Macroeconomic environmental factors may include, for example, GDP per capita, unemployment rate in a country where insurance companies operate, and more. Micro-environmental factors can include age structure of employees, total capital, and more. Among the environmental factors, however, we may also include territorial division, division by frequency of occurrence of floods or storms, demographic indicators and the like. In our analysis we examine the relationship of territorial specifications and technical efficiency of commercial insurance companies. Methods using environmental factors when assessing the efficiency can be, according to Fandel (2001), divided into four groups. Each method has some drawbacks. The first method assumes that values of environmental factors can be sorted according to their impact on technical efficiency ranging from best to worst. In this case, it is common to use the approach proposed by Banker and Morey (1986). The efficiency of a particular DMU is compared only against those DMUs that have values of environmental factors that are equal to or worse than the value of environmental factors of the analyzed DMU. Another method that is not based on assumptions of previous method proposed Charnes, Cooper and Rhodes (1981). The method comprises of the following steps: • all DMUs are divided into subgroups according to environmental variable, • in each of the sub-groups a measure of technical efficiency is expressed for all DMUs belonging to the group, • for each DMU is determined value of inputs respectively outputs at its projection on the production possibilities frontier of the group (projected values) 200 • the values of the technical efficiency score are calculated using the projected values. • using appropriate statistical tests a statistical significance of differences in mean values of the score of technical efficiency is determined within established groups. Another method according to Fandel is based on the assumption that environmental factors are part of the task of mathematical programming. They are classified as input, output or neutral variable. The fourth method consists of two steps. In the first step, linear programming is used to express the technical efficiency score using traditional non-environmental variables. In a second step the regression analysis is employed to determine the correlation of the technical efficiency and environmental factors. The technical efficiency scores have values in the range (0,1) as they originate in the input-oriented DEA model and therefore a special type of regression analysis, known as Tobit regression, is used. In our analysis, we use a method proposed by Charnes, Cooper and Rhodes (1981). To express the technical efficiency score of insurance companies we use the input-oriented BCC model. BCC model We assume that we have n homogeneous DMUs and we monitor m inputs x, and s outputs yif then assuming variable returns to scale model expressing technical efficiency in input-oriented model has form mm 0?-f(eV+eV), (1) subject to xi + s =0„xn, (2) YI-s+=y9, (3) eTl = l, (4) l,s+,s >0 (5) (Jablonský and Dlouhý, 2004) Efficiency score 6 of the q-th DMU takes values in the (o,l) range. The projected values for the inputs and outputs that are necessary to achieve the technical efficiency can be obtained in one of two ways 1. x9'=xť, y9'=Yť, (6) where I* is the vector of optimal values of weights calculated by model or where variables marked * represent vector of optimal values of variables in the input-oriented BCC model. The optimization process takes place in two phases. In the first phase is realized a maximum possible reduction of inputs with the help of value 6q. In the second phase is realized a shift that takes effect with the help of variation variables s+,s . The objective of this model is therefore to reduce the inputs x as much as possible, but so that they remain within an acceptable set of inputs. This reduction provides a projection of the DMU to the frontier, which is a linear combination of efficient DMUs. To test the hypothesis of equal distribution of technical efficiency scores, we use the non-parametric Mann-Whitney U test based on the order of values. Evaluation of hypothesis testing is done based on the values of U statistics and p-values calculated in the program Statistica. 201 Our analysis includes 29 commercial insurance companies in Slovakia and the Czech Republic. Of this number, there were 14 commercial insurance companies based in Slovakia and 15 commercial insurance companies based in the Czech Republic. Inputs in the model include operating expenses and claims incurred. The outputs in the model include earned premium and income from financial investments. Data on insurance companies in Slovakia are collected from the respective annual reports. Data on insurance companies in the Czech Republic come from the results of individual members of the Czech Insurance Association. The aim of our paper is to use the method of Charnes, Cooper and Rhodes (1981) in order to assess whether there is a statistically significant difference in mean scores of technical efficiency of commercial insurance companies on a common insurance market according to selected factors. Selected factors include territorial jurisdiction, and the size of the insurance company, expressed as its market share in terms of earned insurance premiums. Our basic premise reflects the fact that insurance companies in the Czech Republic operate in the larger insurance market, and at the same time the Czech Republic is characterized by lower unemployment rates and greater purchasing power. On this basis, we assume that the insurance companies in the Czech Republic have greater average score of technical efficiency compared to insurance companies in Slovakia. We also assume that insurance companies belonging to group with a greater market share in terms of earned premiums have higher average score of efficiency compared with insurance companies belonging to group with smaller market shares in terms of earned premiums. 3 Results and Discussion At the beginning of the analysis, we expressed the basic descriptive statistics of the two inputs and two outputs. The data are in Table 1. Table 1 Descriptive Statistics of Inputs/Outputs Mean Median Standard Coefficient of (EUR) (EUR) deviation (EUR) variation (%) Operating costs 54807.3 27651.0 62288.6 113.7 Claims incurred 135465.7 73187.0 174675.1 128.9 Earned premiums 201089.5 119984.0 241977.6 120.3 Income from financial investments 45119.8 17225.0 70912.7 157.2 Source: Own processing in Statistica The parameter earned premiums has the highest mean and median values and the parameter operating cost has the smallest mean and median values. The mean of all analyzed parameters was greater than the median i.e. more than half of insurance companies had parameter values that are below the arithmetic mean. The parameter income from financial investments has the highest coefficient of variation. In the next step we have expressed a correlation matrix of analyzed parameters. The values of correlation coefficients are reported in Table 2. There is a strong linear dependence between all pairs of analyzed parameters, as all correlation coefficients are statistically significant. The strongest linear dependence is between the pair of earned premiums and claims incurred. The lowest value of linear dependence is reported between income from financial investments and claims incurred. 202 Table 2 Correlation Matrix Claims incurred Operating costs Earned premiums Income from financial investments Claims incurred 1 0.91 0.96 0.78 Operating costs 0.91 1 0.94 0.80 Earned premiums 0.96 0.94 1 0.86 Income from financial investments 0.78 0.80 0.86 1 Source: Own processing in Statistica Efficiency of commercial insurance companies in Slovakia and Czech Republic Next, we divided the 29 analyzed insurance companies into two groups. In one group were insurance companies from Slovakia and in the other group were insurance companies from Czech Republic. We used the EMS program to determine in each group the technical efficiency score according to relations (1) to (5). These values allow to project in each of the two groups the inefficient DMUs on the efficiency frontier in the respective group. Descriptive statistics for the technical efficiency scores in the two groups are presented in Table 3. Table 3 Descriptive Statistics - Technical Efficiency Score (Territorial) Mean Median Min Max Standard deviation Insurance companies in Slovakia 0.9299 1 0.7309 1 0.1167 Insurance companies in the Czech Republic 0.7960 0.8237 0.2914 1 0.2452 Source: Own processing in Statistica The mean of the technical efficiency scores in the Slovak Republic is greater than in the Czech Republic. The mean of the technical efficiency scores of insurance companies in Slovakia and the Czech Republic is less than the median. The standard deviation of the technical efficiency scores of insurance companies in the Czech Republic is much greater than the standard deviation of the technical efficiency scores of insurance companies in Slova kia. In each of these groups, we adjusted the values of indicators in accordance with equation (7) and this way we have obtained the projected values. Projected values were combined into a single file and we have expressed their technical efficiency scores. In the next step, we expressed the descriptive statistics for the technical efficiency scores of the projected values. The data are presented in Table 4. Table 4 Descriptive Statistics - Technical Efficiency Scores of Projected Values Mean Median Min Max Standard deviation Insurance companies in Slovakia 0.6632 0.6442 0.3874 1.0000 0.2074 Insurance companies in the Czech Republic 0.7900 0.8237 0.2914 1.0000 0.2457 TOTAL 0.7288 0.7393 0.2914 1.0000 0.2331 Source: Own processing in Statistica 203 The mean of the technical efficiency scores of commercial insurance companies in Slovakia is smaller than the mean of the technical efficiency scores of commercial insurance companies in the Czech Republic. In the case of insurance companies in the Slovak Republic the mean is greater than the median i.e. the majority of reported values is below the mean. The mean technical efficiency score for the insurance companies in the Czech Republic as well as the mean technical efficiency scores for all analyzed insurance companies is less than the median i.e. majority of the reported values exceeds the mean. Insurance companies in the Czech Republic have greater standard deviation of technical efficiency scores compared to insurance companies in Slovakia. By using the non-parametric Mann-Whitney U test for two independent sets in the Statistica program we tested the null hypothesis 1H0. The null hypothesis 1H0 is stated as: There is not a statistically significant difference in the mean of technical efficiency scores of commercial insurance companies in the Slovak Republic and the mean of technical efficiency scores of commercial insurance companies in the Czech Republic. Alternative hypothesis lHi is stated as: There is a statistically significant difference in the mean of technical efficiency scores of commercial insurance companies in the Slovak Republic and the mean of technical efficiency scores of commercial insurance companies in the Czech Republic. Sum of the ranks, U-statistic and p-value are reported in Table 5. Table 5 Mann-Whitney U Test Sum of the Sum of the U- P- ranks (SK) ranks (CZ) statistics value The difference in mean values 173 262 68 0.1063 Source: Own processing in Statistica Based on the Mann-Whitney U test, we can conclude that the null hypothesis cannot be rejected. At the significance level of 0.05 there is not a statistically significant difference in the mean values of technical efficiency scores of commercial insurance companies in Slovakia and in the Czech Republic. It follows that the analyzed territorial specification does not affect the efficiency of commercial insurance companies. Efficiency of commercial insurance companies with large and small share in the insurance market. Next, we divided the 29 analyzed insurance companies into two groups according to their share of earned premiums on earned premiums of all insurance companies. In the first group are insurance companies with earned premiums that exceed 2.4% of earned premiums for all insurers. This group consists of 13 commercial insurance companies, of which there are 3 insurance companies based in Slovakia and 10 insurance companies based in the Czech Republic. In the second group are insurance companies with earned premiums of less than 2.4% of earned premiums for all insurers. This group consists of 16 commercial insurance companies, of which there are 11 insurance companies based in Slovakia and 5 insurance companies based in the Czech Republic. Descriptive statistics of technical efficiency scores in the two groups are presented in Table 6. Table 6 Descriptive Statistics - Technical Efficiency Score (Size) Mean Median Min Max Standard deviation Large insurance companies 0.8043 0.8479 0.3448 1 0.2231 Small insurance companies 0.9261 1 0.5769 1 0.1303 Source: Own processing in Statistica The mean of the technical efficiency scores of small and large insurance companies is less than the median. The standard deviation of technical efficiency scores of large 204 insurance companies is greater than the standard deviation of technical efficiency scores of small insurance companies. In each of these groups, we adjusted the values of indicators in accordance with equation (7) and this way we have obtained the projected values. Projected values were combined into a single file and we have expressed their technical efficiency scores. The descriptive statistics for the technical efficiency scores of the projected values is presented in Table 7. Table 7 Descriptive Statistics - Technical Efficiency Scores of Projected Values Mean Median Min Max Standard deviation Large insurance companies 0.9145 0.9998 0.5383 1.0000 0.1454 Small insurance companies 0.7382 0.7720 0.4059 1.0000 0.2424 TOTAL 0.8172 0.8891 0.4059 1.0000 0.2203 Source: Own processing in Statistica The mean of the technical efficiency scores of small insurance companies is smaller than the mean of the technical efficiency scores of large insurance companies. In the case of small insurance companies, large insurance companies as well as all insurance companies the mean is smaller than the median i.e. the majority of reported values is above the mean. Small insurance companies have greater standard deviation of technical efficiency scores compared to large insurance companies. In the next step, we tested the null hypothesis 2H0. The null hypothesis 2H0 is stated as: there is not a statistically significant difference in the mean of technical efficiency scores of insurance companies with a small share in the insurance market and the mean of technical efficiency scores of commercial insurance companies with a large share in the insurance market. Alternative hypothesis 2Hi is stated as: there is a statistically significant difference in the mean of technical efficiency scores of insurance companies with small share in the insurance market and the mean of technical efficiency scores of commercial insurance companies with large share in the insurance market. Sum of the ranks in both groups, U-statistic and p-value are reported in Table 8. Table 8 Mann-Whitney U Test Sum of the Sum of the ranks ranks (insurers with (insurers with U p-level large market small market _share)_share)_ The difference in mean values 231.5 203.5 67.5 0.1095 Source: Own processing in Statistica Based on the Mann-Whitney U test, we can conclude that the null hypothesis cannot be rejected. At the significance level of 0.05 there is not a statistically significant difference in the mean values of technical efficiency scores of commercial insurance companies with a small share in the insurance market and commercial insurance companies with a large share in the insurance market. 4 Conclusions This article uses the method of Charnes, Cooper and Rhodes (1981) for determining the influence of environmental variables on efficiency scores. Our predictions regarding the impact of selected factors on the efficiency of commercial insurance companies were not confirmed. The analysis implies that there is not a statistically significant difference in the 205 mean of technical efficiency scores of insurance companies in Slovakia and the mean of technical efficiency scores in the Czech Republic; and that there is not a statistically significant difference in the mean of technical efficiency scores of small insurance companies and the mean of technical efficiency scores of large insurance companies. Our research, however, has certain limitations. Cooper, Seiford and Tone (2006) reported that the number of analyzed DMUs should be at least three times the sum of the number of inputs and outputs. Our analysis includes four parameters, and therefore each group must have at least 12 DMUs i.e. for our sample, we can create a maximum of two groups. It was therefore not possible to use finer classification criteria, for example to classify size of insurance companies as small, medium and large. Acknowledgements This article is part of a research project funded by the Scientific Grant Agency of the Ministry of Education, Science, Research and Sport of the Slovak republic and the Slovak Academy of Science VEGA 1/0208/14 Insurance Market and Insurance Companies Efficiency. References Banker, R. D., Morey, R. C. (1986). The Use Categorical Variables in Data Envelopment Analysis. Management Science, vol. 1986(4), pp. 513-521. Brokešová, Z., Ondruška, T., Pastoráková, E. (2015). Economic and Demographic Determinants of Life Insurance Industry Development. In: Proceedings of the 12th International Scientific Conference European Financial Systems 2015. Brno: ESF MU, pp. 61-65. Charnes, A., Cooper, W. W., Rhodes, E. (1981). Evaluating Program and Managerial Efficiency: An Application of Data Envelopment Analysis to Program Follow Through. Management Science, vol. 27(6), pp. 668-697. Chovan, P. (2000). Theory and Practice of Insurance. Bratislava: SÚVAHA. Cooper, W. W., Seiford, M. L., Tone, K. (2006). Introduction to Data Envelopment Analysis and Its Uses. Boston: Springer. Czech Insurance Association. (2013). Individual Results of Members. Retrieved from: http://www.cap.cz. Fandel, P. (2001). Environmental factors in the assessment of efficiency in agriculture. In: Proceedings of the Economic and management aspects of sustainable agriculture. International Scientific Days. Nitra: Slovak University of Agriculture in Nitra. Grmanová, E., Jablonský, J. (2009). Analysis of the Efficiency of Slovak and Czech Insurance Companies Using Data Envelopment Analysis Models. JOURNAL OF ECONOMICS, vol. 57(9), pp. 857-869. Insurance Companies. (2013). Annual Reports of Slovak Insurance Companies. Jablonský, J., Dlouhý, M. (2004). Models for Evaluation of Efficiency of Production Units. Praha: PROFESSIONAL PUBLISHING. Vávrová, E. (2015). Insurance Solutions of Covering Agricultural Risks: The Case of the Czech Republic. In: Proceedings of the 12th International Scientific Conference European Financial Systems2015. Brno: ESF MU, pp. 663-669. 206 Holdings of Government Bonds by Commercial Banks during the Financial and Debt Crisis in Europe Vladimír Gvozdják1, Božena Chovancová2 university of Economics in Bratislava Faculty of National Economy, Department of Banking and International Finance Dolnozemská cesta 1, 852 35 Bratislava, Slovak Republic E-mail: vgvozdjak@gmail.com 2University of Economics in Bratislava Faculty of National Economy, Department of Banking and International Finance Dolnozemská cesta 1, 852 35 Bratislava, Slovak Republic E-mail: bozena.chovancova@euba.sk Abstract: The government bonds of the EU countries were considered a safe investment amongst investors and asset managers. The main motive for their holding by commercial banks was to achieve additional revenue from investing spare funds and to store liquidity in them since they can be quickly sold on the secondary market or used as a collateral in refinancing operations. In this working paper we focus on the current issues related to government bond holdings by commercial banks. With the advent of financial and debt crisis some countries failed to meet their liabilities connected with bonds towards investors. This changed the global view of sovereign bonds as the safest form of investment. Based on the analysis of data provided by the Statistical Data Warehouse of the European Central Bank we focus on the evolution of government bond holdings by commercial banks during the debt crisis in Europe in 2010 - 2014 in selected EU countries. By using the panel data regression we will try to estimate the impact of some of the significant determinants which influenced the volume of the holdings - namely credit rating of the sovereign bonds and their interest rates. This regression suggests that there is a correlation between these independent variables and the holdings by commercial banks. Keywords: government bonds, commercial banks, debt crisis JEL codes: G12, G21 1 Introduction Government bonds of the EU countries were considered a safe investments. They are an inherent part of all portfolios of securities in banks, insurance companies and other financial institutions. They were used in order to stabilize portfolios with their secure and predictable cash-flow generated by coupons and face value repaid at the maturity. The current financial and debt crisis has changed this view of investors regarding European sovereign bonds as the most secure asset. Some Eurozone countries were not able to meet their liabilities and that not only led to decreased credit quality of bonds but contributed to the instability of the financial system as well. The aim of this working paper is to analyse the evolution and impact of government bond holding by commercial banks caused by the selected factors which were affected by the debt crisis - i.e. credit rating of the sovereigns and the interest rates of bonds. The issue of government bond holdings during financial crisis is dealt with in more working papers. For example Gennaioli, Martin and Rossi (2015) carried out a research in more than 20.000 banks in 191 countries taking into account 20 sovereign defaults over 1998 - 2012. According to their finding, banks hold on average 9 % of their assets in state bonds. More than 75 % of them are domestic bonds. During sovereign defaults, exposure to government bonds increases. As it will be stated later, this finding is consistent with results coming from our model. In view of the above-mentioned authors there are more reasons why banks increase their sovereign holdings during default: their risk appetite, regulation or financial repression issues. They conclude that there are two 207 main hypotheses for the determinants of bank bondholdings: the "liquidity view" - banks buy bonds during regular business activity because they store liquidity or are used as a collateral in short-term lending. The second one is the "risk-taking view", which means that banks maintain or even increase their bondholdings precisely when they are risky. This is in view of banks' reaching for yield or because of bailout guarantees. The risk-taking view has been most emphasized during the recent European debt crisis, where the large increase in bondholdings has been attributed to banks' search for yield and to moral suasion. This also might include liquidity extensions to banks, and direct purchases of government bonds or conditional commitments to purchase them by the central banks. As the authors showed in their previous works, governments are more willing to repay their debts when domestic banks hold a lot of their bonds. Another reason why commercial banks buy government bonds not fully taking into account their risk level is the preferential regulatory treatment by the current Capital requirement regulation and directive (CRR/CRD IV), according to which the sovereigns are assigned 0 % risk weights. According to Acharya and Steffen (2013) banks hold lower quality government bonds, allowing them to both gain from preferential regulation and to gain high returns without internalizing the systemic consequences of doing so. The preferential treatment of government bonds as a reason supporting their holdings by commercial banks is documented by Bonner (2014), who by using unique transaction-level data suggests that preferential treatment in both capital and liquidity regulation increases banks' demand for government bonds beyond their own risk appetite. The rationale behind favourable treatment in financial regulation is the view that government bonds are risk-free assets making them a reliable source of liquidity and collateral. He also states that regulation leads to a longer-term increase in government bond holdings. At the same time he claims that there is very little evidence of whether regulatory treatment is truly the main driver of banks' large holdings of government bonds or whether this is not rather caused by banks' own targets and risk management process. To distinguish whether a change in banks' government bond holdings is caused by regulation or by its funding and liquidity needs, one would need detailed information on banks' targets used in their internal risk management frameworks, and such data are not available in a structural form. The fact that banks increase the government bond holdings even if their credit quality is worsening brings potential problem: government bonds on the balance sheets of banks are the main transmission channel through which weak government finances may affect the banking system and can constitute a systemic risk. This issue is sometimes referred to as the "doom loop" between governments and their respective banking sector. Euro-area sovereign bonds accounted for just over 10 percent of banks' assets in the currency area, or 2.73 trillion euros, at the end of 2015, which is an increase of about 300 billion euros compared to the previous year, based on the ECB data. According to a report of European Political Strategy Centre (Issue 03/2015, 9 November) the largest share of government bonds is in most cases held in the form of domestic government bonds. Based on the EBA Stress test data in 2014, the share of sovereign debt held by domestic banks in Eurozone varies between countries from more than 10% (Latvia) to over 90 % (Malta). This reflects some facts like the size of the existing stock of national public debt and its attractiveness to foreign banks. The average for the euro area is very high at 57 % and has been increasing since the beginning of the crisis. To a certain degree, the excessive demand for government bonds during the debt crisis was partly supported by the ECB, which released trillions of euro via long-term refinancing operations to commercial banks which used this cheap money for buying higher-yielding bonds issued by their national governments in so-called "carry trades". According to Christopher Thompson from Financial Times (2013) in view of these operations, over the course of two years from October 2011, the Spanish banks increased government bond holdings as a proportion of their total assets from 5 % to 9,4 %, Italian banks from 6,4 % to 10,3 %, Portuguese banks from 4,6 % to 7,8 % and Slovenian banks from 7,8 % to 10 %. In Germany the banks increased this proportion from 3,8 % to 4,5 %, French and Austrian banks by 1 % respectively. The majority of 208 the sovereign bond holdings consisted of banks' own domestic government bonds. In view of Groendahl and Black (2016), the treatment of sovereign bonds as risk-free assets exacerbated the debt crisis in the euro area because the balance sheets of banks in countries including Greece, Spain and Portugal were laden with bonds of their individual sovereigns. Bundesbank supports the introduction of risk weights for government bonds as well as exposure limits, which would reduce the preferential treatment and excessive demand for government bonds. Supporting demand for sovereign bonds by preferential regulation treatment in Basel II and Basel III is confirmed in the work of Lang and Schroder (2015), who proved that the demand is substantially driven by government net issue of securities, and both Basel II (i.e. impact of credit default probability on risk weights) and Basel III (i.e. enhanced capital and new liquidity requirements) have a strong positive impact on banks' for domestic marketable sovereign debt. The primary goal of the Basel framework is to improve the capitalization of the banks and to increase liquidity buffers. Existing bank regulation incentivises banks to purchase more government debt in order to meet these requirements. According to Asonuma, Bakhache and Hesse (2015), there are more factors which contributed to the "home bias" (i.e. the preference of domestic banks for holding domestic sovereign debt instruments compared to other sovereign debt instruments). These were preferential regulatory treatment with a zero risk-weighting. In this context, however, risk weights on other assets, including foreign sovereign debt, might differ significantly between countries which potentially could contribute to cross-country variations of home bias. The increase in home bias during and after the recent crisis period across many countries benefited from the higher importance of domestic sovereign debt for central bank collateral (as well as market funding). The supply of public debt has also increased specially in many advanced economies and led to domestic banks absorbing much of new sovereign debt issuances, when there was a foreign investor retrenchment. In particular, this occurred in an environment of increased global risk aversion. Structural factors, e.g. the availability of other investment opportunities relative to the size of the banking sector could as well affect domestic banks' holding of domestic sovereign debt. Based on the ECB data from the first half of 2013 banks in some countries held excessive volumes of bonds in relation to the Core Tier 1, thanks to preferential treatment and zero weights risk approach (e.g. Germany 214 %, Italy 204 %, Spain 156 %). 2 Methodology and Data In our analysis we focused on the bondholdings by commercial banks during the years 2010 - 2014 in 23 selected countries of the European Union. The main aim of our research is to analyse the evolution and changes in holdings during the financial and debt crisis. We try to estimate the impact of explanatory variables - in our case the sovereign credit rating, which reflects the creditworthiness of a particular state, and long-term interest rates of bonds as a yield. Government bond holdings represent in our case the dependent variable. As for the credit rating we used the rating of Standard & Poor's in particular year. Since these values are expressed not as numerals, we used a linear scale transformation following the working paper Sovereign Credit Ratings and Financial Markets Linkages (Alfonso, Furceri and Gomes, 2011). In line with this scale, the best rating AAA is 17 and the worst rating D is 1. Credit rating should reflect the ability of a particular state to meet its debt towards investors in time, and thus, generally speaking, should be a factor which influences the decisions of banks to invest or not to invest into this type of assets. Another independent variable in our model is interest rates of 10-year government bonds as a factor reflecting the yield of this type of asset. In the following figure we can see the evolution of 10-year government bonds interest rates in 2010 - 2014. We can see the 209 gradual decline in interest rates for each state, which might be caused by the fall of market interest rates in line with the declining ECB key rates. Figure 1 Interest Rates of 10Y Government Bonds of Selected EU Countries in 2010 - 14 ■ 2010 1 2011 12012 1 2013 12014 Source: Own processing based on ECB Statistical Data Warehouse The dependent variable in our model is the government bondholdings of monetary and other financial institutions in selected countries of the European Union in 2010 - 2014. The data is provided by the Statistical Data Warehouse of the European Central Bank. We focus only on holdings of domestic government bonds by domestic banks, so that we can analyse the impact of only domestic credit rating and sovereign rates changes during the years of the European debt crisis. Figure 2 Domestic Government Bond Holdings by Commercial Banks in Selected EU Countries in 2010 - 2014 1400 1 200 1000 1 ll III . | 1IH.11 .... iiii .. .■ i II 1 . lliliill^MinK, ... l 200 if J ■ 1010 ■1011 ■ 1011 «1013 B1014 Source: Own processing based on ECB Statistical Data Warehouse As seen in the graph, the volume of sovereign bond holdings had mostly an increasing trend in the selected EU countries with only a few exceptions (Belgium, Czech Republic, Germany and Slovakia). We can see that the highest volume of holdings by commercial 210 35 banks is in France, Germany, Italy, Poland and Spain. France, Italy and Spain increased their holdings despite the fact that the sovereign bond ratings of these countries downgraded during this period. Model Specification In our analysis we used the panel regression with random effects. This method is used for analysing panel data, which in our case represent the evolution of bond holdings, rating and interest rates in 23 countries during 2010 - 2014. The fitness of this model was confirmed by the Hausman test, by means of which we rejected the null hypothesis that both fixed-effects model and random effect model are consistent, suggesting that the random-effect model would be appropriate in our case (H = 2,5475 with p-value = prob(chi-square(2) > 2,5475) = 0,27978). 3 Results and Discussion Based on the data of independent variables (credit ratings and interest rates of 10Y government bonds) and dependent variable (government bond holdings by commercial banks) in 23 countries during 2010 - 2014 we got these results of the regression: Table 1: Panel Regression of Data (Random Effects Model) coefficient std. error t-ratio p-value const 362975 89055,6 4,076 8,60e-05 *** Rating_transf -9823,22 2762,59 -3,556 0,0006 *** Y interest rate -5429,43 2680,19 -2,026 0,0452 ** Mean dependent var 223893,8 S.D. dependent var 385102,8 Sum squared resid 1,76E+13 S.E. of regression 395003,7 Log-likelihood -1644,134 Akaike criterion 3294,268 Schwarz criterion 3302,503 Hannan-Quinn 3297,61 Within' variance = l,60E+09 'Between' variance = 1,56E+11 theta used for quasi-demeaning = 0,954728 corr(y,yhat)/v2 = 0,0404669 Source: Own processing Following the regression we can see that intercept and coefficients for credit rating and interest rates of 10-year government bonds are statistically significant. In both cases the coefficients are negative. In case of rating it means that the lower the rating, the higher the government bond holdings. As for the interest rates it would mean the lower the rates, the higher the government bonds holdings. The interpretation of these results might be difficult, because they contradict the logical motivation for buying these assets, i.e. the higher are the rating and rates, the higher are the government bond holdings. In our case it is quite the opposite. This is in our view the result of the financial and debt crisis which changed the view of basic investor's logic. In view of Figure 3 in all selected countries with only few exceptions the commercial banks increased their holdings of bonds during the period in spite of the fact that the credit ratings were downgraded and interest rates declined (see Figure 2). In our opinion, there might be three different reasons which could explain this phenomenon: • lack of other safer investment opportunities, • preferential treatment of government bond holdings in line with the capital regulation of the EU (Capital Directive IV), where government bonds of the EU countries are considered risk-free. This means that sovereign bonds under these 211 conditions have zero risk weight and banks thus have zero credit risk exposure towards these assets and they don't have to increase their own capital, • moral suasion of commercial banks when they rely on the fact that if a sovereign defaulted, it would be bailed-out and the banks would be redeemed. This hazard could potentially jeopardize the stability of the financial sector. In line with our regression, the results are consistent with findings of authors mentioned at the beginning of the paper (Gennaioli, Bonner, Asonuma and others). It is quite interesting that there was no statistically significant relationship between sovereign bond holdings on the one hand and ratings and interest rates on the other hand in the period prior to the financial and debt crisis (years 2005 - 2009). We assume that the current capital regulation is responsible for the evolution of bondholdings during 2010 - 2014. Since the holdings increased in spite of growing default risk we can expect changes in the European capital regulation and preferential treatment of sovereign bonds in a short time in order to avoid potential systemic shock in the future. 4 Conclusions The aim of this working paper was to assess the impact of sovereign credit rating and 10Y bond interest rates on the bond holdings of commercial banks during 2010 - 2014 in selected 23 countries of the European Union by using panel regression. Based on our results there exists a statistically significant inverse relationship between these variables, which might contradict the basic investors' logic. It is most likely that this phenomenon is caused by preferential treatment of government bonds as risk-free assets supported also by moral suasion and a pledge of bail-out of defaulted sovereigns, especially in the Eurozone area. This might lead to an overexposure of commercial banks to domestic debt, bearing in mind that banks own on average almost 9 % of their assets in the form of domestic government bonds. In case a state defaulted, there could be a potential systemic shock of domestic commercial banks. These conditions call for a reform of the regulation. In an article by Francesco Guarascio (2015), the European Commission plans to review the banks' sovereign bond holdings to break excessive exposures to national debt, seen as a vulnerability of the Eurozone banking system. In the medium term, it may make sense to review the treatment of bank exposures to sovereign debt, for example by setting large exposure limits. Among the largest Eurozone countries, Italy and Spain would be likely to see the biggest impact on their bond markets, as banks in those countries hold significant amounts of their sovereign debt. A debate is continuing within the European Central Bank on whether a quantitative cap should be preferred to a limit based on the asset risk. There is a proposal of a 25 % equity cap, which would force Eurozone banks to sell bonds worth 1.1 trillion euros, according to a report by credit rating agency Fitch published last year. Banks can now easily offload some of their excessive bonds onto the ECB and in turn help its bond-buying stimulus programme, while increasing the financial stability of the Eurozone. Another possibility to avoid such a concentration of risk would be to buttress the bonds with capital reserves, which would at least make it possible to adequately contain the risk over the medium term. Acknowledgements This working paper is a result of the project VEGA (1/0124/14) "The role of financial institutions and capital market in solving problems of debt crisis in Europe". References Acharya, V., Steffen, S. (2014). The Greatest Carry Trade Ever? Understanding Eurozone Bank Risks. NBER working paper 19039. Retrieved from: http://www.nber.org/papers/wl9039. 212 Alfonso, A., Furceri, D., Gomes, P. (2011). Sovereign Credit Ratings and Financial Markets Linkages - Application to European Data. ECB Working Paper Series no 1347/June 2011. Retrieved from: https://www.ecb.europa.eu/pub/pdf/scpwps/ecbwpl347.pdf7af8437707498f9ee33f5da67 75c3460f. Asonuma, T., Bakhache, S., Hesse, H. (2015). Is Banks' Home Bias Good or Bad for Public Debt Sustainability?. IMF Working Paper WP/15/44. Retrieved from: https://www.imf.org/externa l/pubs/ft/wp/2015/wpl544.pdf. Bonner, C. (2014). Preferential Regulatory Treatment and Banks' Demand for Government Bonds. DNW Working Paper No.433/July 2014. Retrieved from: http://www.dnb.nl/binaries/Working%20Paper%20433_tcm46-310158.pdf. European Political Strategy Centre (2015). Severing the 'Doom Loop': Further Risk Reduction in the Banking Union. Five Presidents' Report Series. Issue 03/2015, 9 November. Retrieved from: http://ec.europa.eu/epsc/publications/series/5p_bankexposure_en.htm. Gennaioli, N., Martin, A., Rossi, S. (2015). Government Default, Bonds, and Bank Lending Around the World: What do the Data Say? Retrieved from: http://www.econ.upf.edu/~martin/gmr2.pdf. Groendahl, B., Black, J. (2016). Forget Basel IV: Bundesbank Says Beware of Banks' Sovereign Risk. Bloomberg. 18 January 2016. Retrieved from: http://www.bloomberg.com/news/articles/2016-01-18/forget-basel-iv-bundesbank-says-beware-of-banks-sovereign-risk. Guarascio, F. (2015). EU Plans to Review Rules on Banks' Exposure to Sovereign Bonds. Reuters. 24 November 2015. Retrieved from: http://uk.reuters.com/article/uk-eu-banks-sovereign-idUKKBN0TDlHO20151124. Lang, M., Schroder, M. (2015). What Drives the Demand of Monetary Financial Institutions for Domestic Government Bonds? - Empirical Evidence on the Impact of Basel II and Basel III. Frankfurt School of Finance & Management. Retrieved from: http://www.frankfurt-school.de/clicnetclm/fileDownload.do?goid=000000683018AB4. Thompson, Ch. (2013). European Banks Overexposed to Domestic Debt: Feedback Loop Between Banks and Sovereigns is a Concern. Financial Times. 23 December 2013. Retrieved from: https://next.ft.com/content/8290470c-6bl7-lle3-8e33-00144feabdc0. 213 Foreign Trade Finance: What is the Impact of the Global Financial Crisis of 2007-2009? Peter Haiss, Franz Binder, Kushtrim Hajzeri, Wadim Kalmykov1 1 WU University of Economics and Business Department of Global Business and Trade, Institute of Export Management Welthandelsplatz 1, Building Dl, A-1020 Vienna, Austria E-mail: peter.haiss@wu.ac.at Abstract: We analyze the impact of the global financial crisis of 2007-2009 on international trade finance. Two different perspectives are covered: the demand side and the supply side of trade finance. Additionally, the implications for trade insurance and Export Credit Agencies are investigated. Furthermore, the role of the state is discussed as the link between trade, finance and GDP growth is of utmost importance for economic policy actors. Our research draws on survey and financial market data and gives a global overview. We find that on the demand side, both the volume of finance and its mix changed. On the supply side, volume and prices were affected as well. The role of the state has risen in course of the crisis in terms of insuring trade via Export Credit Agencies (ECAs). Ever since, the market moved back to normal, and will be more digital soon. Keywords: trade finance, foreign trade, financial crisis, ECA JEL codes: E41, E51, F10, F34, G21 1 Introduction Trade finance refers to any kind of financing associated with trade credit, trade insurance and guarantees, typically using trade credit as collateral (Ahn, 2011). It covers a broad spectrum of payment arrangements between importers and exporters and is conducted not only through third-party financial institutions, but also through inter-firm transactions. The main objective of trade finance is to bridge the time gap between the delivery and acceptance of a good and thus it deals mainly in short-term maturities of up to 180 days (Auboin and Meier-Ewert, 2003). However it is also used for mid- and longer-term financing when finance for the production and sale of large capital goods is required. Trade finance is a driver for GDP growth, particularly for emerging economies (Bordo and Rousseau, 2011). According to the World Trade Organization (WTO) about 80 to 90% of global trade relies on trade finance (Auboin, 2009). It is one of many reasons, why this topic is highly interesting and practically important to a variety of institutions. Managers for example are in particular interested to know how the supply of trade finance changes during a crisis. Are banks willing to supply less? Does it become more costly to get trade finance? Banks on the other hand are interested to know whether exporters and importers will demand more or less trade finance or if during financial crisis companies switch from one option of trade finance to another. To governments it is especially during a crisis a highly relevant topic to focus on keeping trade and the economy going. Trade finance influences both variables and therefore policy makers are highly interested in the implications of trade finance and are motivated to encourage it. According to the International Chamber of Commerce (ICC) Trade Register, trade finance is predominantly a low risk business characterized by low default rates and a high level of collateralization (ICC, 2013b). From an institutional perspective, the major branches are inter-firm trade credit, bank-intermediated trade finance and export credit insurance. Inter-firm trade credit plays a particularly important role and accounts for about 60 to 65% of trade finance volume (ICC, 2013b). Especially in developing countries it is a relatively prevalent form of trade finance (Demirgüg-Kunt and Maksimovic, 2001). The two types of inter-firm trade credit transactions are open account arrangements on one side and cash-in-advance arrangements at the opposite side of the spectrum. 214 Open account transactions constitute the largest share of trade finance arrangements between importers and exporters. It involves the exporter supplying working capital to the importer by extending credit to the importer directly. The exporter then bears the risk in that transaction, such as the credit risk (risk of non-payment). By contrast, the importer has to bear the risk in cash-in-advance arrangements as he pays for the goods before they are shipped (Madura and Fox, 2014). Cash-in-advance financing is estimated to be only about half as common as open account financing (ICC, 2013b). A significant part of open account transactions are intra-firm (trade within the same corporate group), and therefore presumably do not need any bank-financing and insurance (Asmundson et al., 2011). Another significant part of open account transactions are insured by Export Credit Agencies (ECAs) as discussed below. Alongside importers and exporters, banks play a central role in trade finance. They act as intermediaries in the process and provide liquidity in the form of working capital to companies and supply payment mechanisms that reduce the payment and nonperformance risks (Love, 2011). Among the intermediated trade finance products, the two most commonly used are letter of credits (L/Cs) and documentary collections (Niepmann and Schmidt-Eisenlohr, 2014). The amount of capital a bank is willing to commit is subject on the bank's assessment of risk i.e. counterparty and country risk, and also on supervisory regulations (i.e. the regulatory minimum capital requirements) (Chauffour and Farole, 2009). In addition, in times of crisis banks tend to become more risk-averse in general and as a result cut down on lending activities altogether (Berman and Martin, 2011). It will be later analyzed to what extent this happened during the Financial Crisis of 2007-2009. Alongside bank intermediated trade finance, export credit insurance is a crucial component in international trade (Auboin and Meier-Ewert, 2003). As an alternative to bank-intermediated trade finance products, export credit insurances represent a different approach to mitigating the risk of non-payment. There are two main ways of obtaining these guarantees. On one hand, from private insurers, which are typically short term guarantees. On the other hand, long term export loans and guarantees can rather be obtained from public export credit agencies (ECAs). They collaborate with commercial banks and provide guarantees and liquidity, i.e. also have a complementary function (Asmundson et al., 2011). The Berne Union is an organization which consists of public and private companies. It was founded 1934 in Berne, Switzerland and has 80 members today. It also founded the Prague Club with the aim to support young export credit agencies. The Berne Union insured USD 1.9 trillion of exports and FDIs in 2009, which constitutes more than 10 percent of international trade (Berne Union, 2010). In the following we analyze the impact of the global financial crisis 2007 - 2009 on global foreign trade finance up to now. We focus on the following questions: • What were the implications on the demand and on the supply side for bank trade finance? • What were the changes for trade insurance and export credit agencies (ECAs)? The remainder of the paper is organized as follows. In the first part the theoretical foundations will be discussed. Thereafter, surveys and other data will be analyzed to answer the above research questions. 2 Methodology and Data To answer the research questions, we mainly make use of yearly survey data from the International Chamber of Commerce (ICC) as well as the IMF in corporation with the Bankers' Association for the Finance and Trade - International Financial Services Association (BAFT-IFSA). These surveys were conducted in the wake of the financial crisis of 2007-2009 as assessment of trade finance conditions would otherwise not have been possible (Curran, 2009). The ICC Global Trade and Finance survey has since become an important publication published yearly by the ICC. Still, in contrast to trade volume data, coherent data across countries for the trade finance market does not exist 215 (Asmundson et al., 2011), so we have to rely on the accuracy of the surveys to provide the insights needed to analyze the structure, volume and movements during the period under review. Alongside this, SWIFT data, WTO trade volume data and Berne Union data are the foundation for giving an in-depth analysis on the impact the financial crisis had on trade finance developments during and after the crisis. In addition we also include various literature sources and expert insights in our analysis to present a more detailed analysis. The research gives a broad global overview without focusing on any region specifically. Not only the direct short-term effects are analyzed, but also whether the crisis has had long lasting implications for trade finance up to the year 2015. 3 Results and Discussion The Financial Crisis and the Great Trade Collapse Starting in the U.S financial sector, the financial crisis quickly spread around the world and into all sectors that were dependent on credit or other financial products. The world went through a difficult re-balancing, with important economic, social, and political implications for almost all major nations (Frieden, 2009). Since most industries and businesses are financed through banks and equally affected capital markets, the crisis led to a tremendous slow down of the whole economy, especially in world trade. For decades the expansion of global trade had depended on reliable and cost-efficient sources of finance, backed by a deep, global secondary market of fluid and secured financing instruments, and credit insurance products provided by private and public institutions (Auboin, 2009c). Figure 1 Merchandise Trade Volumes 2005-2014 (Absolute Values in USD Bn) 16000 200 0 0 -j-1-1-1-1-1-1-1-1- 2005 2005 2007 2008 2009 2010 2011 2012 2013 2014 Source: Baldwin (2009), based on WTO Database Trade volumes of advanced, emerging, and developing economies were all growing until the sharp decline in the second half of 2008 and early 2009 (figure 1) during which world trade contracted by about 30% (Baldwin, 2009; Chor and Manova, 2012). Accordingly, this period is often referred to as The Great Trade Collapse. The importance of trade finance and credit insurance to support trade flows became apparent (Morel, 2011). After this period economies stabilized in 2009 and started recovering afterwards. However in 2010, in most economies trade was still lower than at the peak in 2008 (Asmundson et al., 2011). The surprisingly fast recovery slowed down in 2011. This development had many reasons for example the Arab Spring, Sovereign Debt Crisis in Europe or natural causes like the earthquake in Japan (ICC, 2012 and 2013a). These follow-up recovery, though required major institutional efforts. As many economies were facing severe problems in 2007/2008, fears of a new era of protectionism, including total trade bans, were rising. At that point it was crucial for the world to maintain and strengthen an integrated international trading system (Frieden, 216 2009). The fear of a trade collapse led to the first meeting of the Group of Twenty (G-20) in November 2008. They declared that it was critical to reject protectionism and avoid turning inward in face of a crisis (Tussie, 2012). Despite their pledge to eschew protectionism for at least 12 months, in 2009 international trade suffered its hardest drop in 70 years. G-20 governments together had implemented as many as 179 measures that harmed foreign trade, investments, workers, and intellectual property (Evenett, 2009). Those discriminatory instruments were mainly financial assistance packages (state aids, bailouts), anti-dumping, countervailing, safeguard actions and tariff increases (Curran, 2009). At the time the WTO recommended increasing the capacity of international financial institutions and exporting credit agencies (ECAs) to take some risks of the private sector partners. Subsequently, regional development banks and the International Financial Corporation (IFC), a member of the World Bank Group encouraging private sector development, doubled the on average capacity under trade facilitation programs. ECAs launched short-term lending programs for working capital and credit guarantees aimed at small and medium businesses. Several countries also used their central banks foreign exchange reserves to supply local banks and importers with foreign exchange through repurchasing agreements (Auboin, 2009). In order to counter the supply-decrease of trade finance the G-20 leaders proposed a new trade finance "package" at the 2009 London Summit. Its main point was to allow greater co-lending and risk co-sharing between banks and international and national institutions through guarantees against commercial and political risk. This included a reinforcement of the IFC's global trade finance facility through the introduction of a liquidity pool, allowing for a 40-60% co-lending agreement between the IFC and commercial banks. Additionally it was made possible for ECAs to provide more direct funding in the short-run (working capital lending; Auboin, 2009). Those measures introduced by the G-20 in collaboration with the WTO helped to restore the confidence and stabilized trade finance markets fairly rapidly. While the supranational intervention in the market was a major factor in rebuilding international trade itself (Auboin and Engemann, 2013), the crisis still left its imprint on trade finance with regard to volumes, prices and institutions as discussed in the following. Structural Changes of Trade Finance Triggered by the Crisis According to ICC surveys, 2/3 of banks reported that their trade credit volumes decreased 2007 and 2008 across the globe (ICC 2010). There was a strong spillover from interbank markets, as 40% of the banks decreased their credit lines for corporates but 52% for financial institutions. The ICC (2010) mainly quoted supply-side reasons for the decline in trade finance, particularly more stringent credit criteria, restrictions in capital allocation, market exits, and reduced inter-bank lending. The IMF-FAFT-IFSTA (2010) rather found demand-side causes, referring to the fall in underlying trade activity, a fall in the price of transactions alongside less credit availability and a shift towards open account transactions. Beginning in the second half of 2010, trade finance rebounded. From a pricing perspective, the crisis led to an upswing. In the wake of the banking liquidity crisis and global trade collapse, it is indicated in the IMF-BAFT (2010) survey that approximately 90% of banks raised their prices for trade related products (Dorsey, 2009; Asmundson et al., 2011). According to the IMF "the largest banks were much more likely to increase pricing and by larger average amounts..." (Asmundson et al., 2011). Short-term repercussions included in some cases the widening of spreads for letters of credit from 10-15 basis points to levels 250-500 basis points above LIBOR (Auboin, 2009). The ICC (2009) cites banks' higher funding costs, increased capital constrains and greater counterparty risk as the main reasons for the increase in pricing. With the recovery of trade and bank liquidity, markets started to quickly return to normal pricing conditions in 2010 over the next years, "...the average price for L/Cs in large emerging economies fell from 150-250 basis points in 2009 to 70-150 basis points in 2010" (ICC, 2011). The trend continued and resulted in tight prices for trade finance 217 products from in 2013 onwards due to intense competition and abundant short-term liquidity in global markets (Committee on the Global Financial System, 2014). According to Chauffeur and Farole (2009), "... increased risk aversion on the part of importers and exporters increased their willingness to pay for bank trade finance in spite of increased pricing". The increased pricing did not result in a reduced share of bank-intermediated trade finance to world trade. Bank-intermediated transactions gained in relative importance in the wake of the crisis. Their relative share rose form 33% in October 2007 to 36% in January 2009. Even more explicit is the relative reduction of open account transactions from 48% in October 2007 to mere 42% in January 2009. In addition, the ICC surveys (2009, 2010) indicate that the confirmation requests for L/Cs increased considerably between 2007 and 2009 in spite of increased fees for confirmation. Were these changes in demand preferences sustained? Probably not. According to the ICC (2015) the share of open account transactions has returned to its pre-crisis level as the crisis abated. Banks changed their lending practices, though. As consequence of increased risk and risk-aversion banks became more cautious and limited their exposure, particularly with certain sectors and countries (e.g. small and medium sized enterprises; Chauffeur and Malouche, 2011). In addition, "banks have also limited their own risk through expanded insurance, shorter maturities and stronger covenants, and higher cash deposits or other collateral from clients" (Asmundson et al., 2011). The stringent risk management practices have eased, but are likely to have remained somewhat stricter than in pre-crisis years (ICC, 2012 and 2014). Export credit insurers have already played an active role in supporting international trade prior to the crisis. As a result of the crisis the share of short and long term ECA-covered exports increased dramatically (Morel, 2011). Explanations for this were exiting and newly introduced ECA programs during the crisis (Asmundson et al., 2011). Figure 2 Short and Long-Term Export Credit Insurance (in % of Global Export Volumes) 1100% 10.50% 10 00°, 9.50% 9.00% 8,50% !,00% 2005 2006 2007 2008 2009 2010 2011 2012 2013 Source: Bernue Union (2010) based on WTO Database 2014 In the pre-crisis period 2005 to early 2008, an almost parallel development between credit insurance and exports could be observed (with a share of about 9%. In the course of the financial crisis the ratio changed fundamentally and increased to a level of 10.5%. According to the IMF "this suggests that ECAs may have played an important role in cushioning the downturn (Asmundson et al., 2011)." After the crisis the ratio of export credit insurance and export volumes dropped by more than one percent nearly reaching a pre-crisis level in 2011. Since then a clear upward trend can be observed. This development can be attributed to the fact that credit insurance as a tool for mitigating 218 risk in international trade has gained in appreciation, which has led to a lasting increase in demand (Morel, 2011). A similar development can be seen in the ratio between the short term claims and the turnover covered by Berne Union. Short term claims paid more than doubled, from $1.1 billion in 2008 to $2.4 billion in 2009. Many claims emerged in the end of 2008 and were paid in 2009 (Morel, 2011). The ratio between short term claims and covered export turnover surged, indicating higher risk during the period. Starting at the end of 2008 at a level of 0.08% it peaked in 2009 at a rate of 0.22%. Thereafter high claims declined in 2010 and stabilized at a level of approximately 0.12% in 2012 (Morel, 2011). The increased risk perception and increasing claims in 2008 also led to lower supply of private trade credit insurance. While credit limits were reduced and premiums raised by private trade credit insurers (Van der Veer, 2011), public-owned ECAs increased theirs. This was mainly a result of governmental measures supporting ECAs to fill the gap in export credit insurance supply (Berne Union, 2010). The share of ECAs in global short term supply increased in 2009 and 2010 in comparison to previous years. From 2006 to 2010 the share covered by ECAs rose from 15 to 28%. Annual reports of some European countries show the decline in supply of private trade insurers as means to reduce their exposure. Austria for example showed a private supply decline of 15-30% and Sweden of 20-30% (Van der Veer, 2011). Table 1 Credit Limits of Private and Public Trade Insurance 2006 to 2010 (in Percent) 2006 - 2008 2009 2010 Private insurers 85.00% 79.00% 72.00% ECAs 15.00% 21.00% 28.00% Source: Berne Union (2010) In the EU public ECAs even began to insure short term marketable risks, which under normal circumstances would have been covered by private trade insurance agencies as ECAs had left these risks to the private market years before. During the crisis ECAs helped especially small and medium-sized enterprises to continue foreign trade by providing trade credit insurance (Berne Union, 2010). While stepping in during the crisis public trade insurance could have had an important role by reassuring the private sector that governmental institutions are prepared to give support through challenging periods (Asmundson et al., 2011). Although credit insurance is not a source of liquidity in itself, it helped to unlock bank financing during the crisis and was able to ensure that liquidity was available for short term and medium to long-term finance (Morel, 2011). 4 Conclusions This paper provides an overview on the impact of the 2007-2009 global financial crisis on bank trade finance. This period had many implications both for companies on the demand side and banks and trade credit insurance agencies on the supply side. Even statutory regulations were changed as a result of the tremendous impact the crisis had on world trade. Demand for bank trade finance fell in absolute terms during the crisis as a result of reduced trade in the short term. However, relative to trade volumes there was more demand for bank-intermediated finance and letters of credit (L/Cs), and less demand for open account transactions. As trade recovered, demand in absolute terms also recovered. We could not find lasting changes in relative demand of companies for bank trade finance as a result of the crisis. The pre-crisis trend towards less L/Cs and more open account transactions is likely to have continued. 219 On the supply side, a decline in the quantity supplied and higher prices were observed in the short term. It seems that those were only temporary effects that subsided in most sectors and regions as the crisis abated. Sustained effects and implications of the crisis include Basel III as response to the crisis, and risk management practices that are likely to have remained somewhat stricter than in pre-crisis years. Implications for trade insurance and Export Credit Agencies (ECAs) were similarly severe. Short term and overall insurance volumes decreased in the crisis, though in relative terms they decreased less than trade. By contrast, medium to long term insurance increased both in relative and absolute terms during the crisis. One of the causes is the public sectors' intervention, which in turn led to an increased market share of ECAs during the crisis. As a result of the crisis (short term) trade credit insurance has become ever more important for companies and banks, which is reflected in the increasing volumes in the past years. In a forward-looking mode, global trade continues growing and banks as well as trade credit agencies are essential for this development to continue. New challenges are on the horizon, especially digitization is going to be one of the main trends in the trade finance market. This could lead to a replacement of paper document flows by electronic data flows. Also supply chain solutions, especially for large companies, will gain in importance. As demand and supply changes, the business of trade finance will evolve and support foreign trade and the economy at large. References Ahn, J. (2011). A Theory of Domestic and International Trade Finance, IMF Working Paper 2011/262. Retrieved from: https://www.imf.org/externa l/pubs/ft/wp/201 Vwpll262.pdf. Asmundson, I., Dorsey T., Khachatryan, A., Niculcea, I., Saito, M. (2011). Trade and Trade Finance in the 2008-09 Financial Crisis, IMF Working Paper 11/16. Retrieved from: https://www.imf.org/externa l/pubs/ft/wp/201 Vwplll6.pdf. Auboin, M., Meier-Ewert, M. (2003). Improving the Availability of Trade Finance during Financial Crises, WTO Discussion Paper XI-2003. Retrieved from: https://www.wto.org/english/res_e/booksp_e/discussion_papers2_e.pdf. Auboin, M. (2009). Trade Finance: G20 and follow-up, VOX CEPR's Policy Portal. Retrieved from: http://www.voxeu.org/article/trade-finance-g20-and-follow. Auboin, M., Engemann, M. (2013). Trade Finance in periods of crisis: What have we learned in recent year? WTO Staff Working Paper ERSD-2013-01, WTO Economic Research and Statistics Division. Baldwin, R. (2009). The great trade collapse: What caused it and what does it mean? In Baldwin, R., ed.: The Great Trade Collapse: Causes, Consequences & Prospects, pp. 1-14. Retrieved from: http://www.voxeu.org/sites/default/files/great_trade_collapse.pdf. Bordo, M., Rousseau, P. (2011). Historical Evidence on the Finance-Trade-Growth Nexus. NBER Working Paper No. wl7024. Retrieved from: http://ssrn.com/abstract=1833161. Berne Union (2010). Berne Union Yearbook 2010. Retrieved from: http://www.berneunion.org/wp-content/uploads/2013/10/Berne-Union-Yearbook- 2010.pdf. Chauffour, J., Farole, T. (2009). Trade Finance in Crisis Market Adjustment or Market Failure? The World Bank Policy Research Working Paper 5003, Retrieved from: https://openknowledge.worldba nk.org/bitstrea m/handle/10986/4195/wps5003.pdf?sequ ence=l. Chauffour, 1, Malouche, M. (2011). Trade Finance during the Great Trade Collapse, The World Bank, Washington DC. Retrieved from: http://econ.sciences-po.fr/sites/default/files/file/pmartin/Trade-Finance-finalpdf.pdf. 220 Chor, D., Manova, K. (2012). Off the cliff and back? Credit conditions and international trade during the global financial crisis. Journal of International Economics, vol. 87(1) pp. 117-133. Retrieved from: http://web.stanford.edu/~manova/crisis.pdf. Curran, L. (2009). The impact of the Crisis on EU Competitiveness in International Trade. In: Forum, The Impact of the Financial and Economic Crisis on World Trade and Trade Policy, pp. 264-268. Retrieved from: http ://www.intereconomics.eu/downloads/getfile.php?id=702. Demirguc-Kunt, A., Maksimovic, V. (2001). Firms as Financial Intermediaries: Evidence from Trade Credit Data, World Bank Working Paper 2696. Dorsey, T. (2009). Trade Finance Stumbles. Finance and Development, vol. 46(1), Retrieved from: http://www.imf.org/external/pubs/ft/fandd/2009/03/dorsey.htm. Evenett, S. 1 (2009). Crisis-era protectionism one year after the Washington G20 meeting. In: Baldwin, R., ed., The Great Trade Collapse: Causes, Consequences and Prospects, pp. 37-46. Retrieved from: http://www.voxeu.org/sites/default/files/great_trade_collapse.pdf. Frieden, J. (2009). Global trade in the aftermath of the global crisis. In: Baldwin, R., ed. The Great Trade Collapse: Causes, Consequences and Prospects, pp. 25-30, Retrieved from: http://www.voxeu.org/sites/default/files/great_trade_collapse.pdf. IMF-BAFT-IFSA (2010). A survey among banks assessing the current trade finance environment. Trade finance services: Current environment & recommendations: Wave 3. ICC (2009, 2010, 2011, 2010). Rethinking Trade Finance, ICC Annual Global Surveys on Trade Finance, International Chamber of Commerce. Retrieved from: http://www.iccwbo.org/. ICC (2013a). Rethinking Trade & Finance 2013: An ICC Private Sector Development Perspective, International Chamber of Commerce. Retrieved from: http://www.iccwbo.org/. ICC (2013b). Global Risks Trade Finance Report 2013, ICC Trade Register, International Chamber of Commerce. Retrieved from: http://www.iccwbo.org/. ICC (2014, 2015). Rethinking Trade & Finance 2014/2015: An ICC Private Sector Development Perspective, International Chamber of Commerce. Retrieved from: http://www.iccwbo.org/. Love, I. (2011). Trade Credit versus Bank Credit during Financial Crises. In: Chauffour, J., Malouche, M., eds.: Trade Finance during the Great Trade Collapse, The World Bank, Washington DC, pp. 27-40. Retrieved from: http://econ.sciences-po.fr/sites/default/files/file/pmartin/Trade-Finance-finalpdf.pdf. Morel, F. (2011). Credit Insurance in Support of International Trade: Observations throughout the Crisis, In: Chauffour, 1, Malouche, M., eds.: Trade Finance during the Great Trade Collapse, The World Bank, Washington DC, pp. 337-356. Retrieved from: http://econ.sciences-po.fr/sites/default/files/file/pmartin/Trade-Finance-finalpdf.pdf. Niepmann, F., Schmidt-Eisenlohr, T. (2013). International Trade, Risk, and the Role of Banks Federal Reserve Bank of New York Staff Reports No. 633, (Revised Nov. 2014). Retrieved from: http://www.newyorkfed.org/research/staff_reports/sr633.pdf. Tussie, D (2010). G20 and the multilateral trade impasse, FRIDE Policy Brief, ISSN: 1989-2667, FRIDE - A European Think Tank for Global Action. Retrieved from: http://fride.org/download/PB_G20_7_eng_The_G20_and_the_multilateral_trade_impass e.pdf. Van der Veer, K. J.M. (2011). Private Trade Credit Insurers during the Crisis: The Invisible Banks. In Chauffour, 1, Malouche, M., eds.: Trade Finance during the Great Trade Collapse, The World Bank, Washington DC, pp. 199-212. Retrieved from: http://econ.sciences-po.fr/sites/default/files/file/pmartin/Trade-Finance-finalpdf.pdf. 221 The Influence of the Size of the Region on the Financial Situation of Hospitals Taťána Hajdíková University of Economics, Prague Faculty of Management, department of management Jarošovská 1117/11, Jindřichův Hradec, Czech Republic E-mail: hajdikova@fm.vse.cz Abstract: The presented paper will focus on selected financial indicators. When selecting indicators, their application for the non-profit sector, particularly hospitals, will be taken into account. The study of literature suggests that there are opinions about the use of specific indicators in the healthcare system. In recent years, many European countries have undergone reforms in the management of the healthcare market to improve the quality of hospital services. Some of the taken measures resulted in the closure of hospitals, which were characterized by worse financial performance, particularly in rural areas. The literature also suggests that hospitals located in rural areas are more economically sensitive, especially to economies of scale. The aim of the research is to determine whether there are differences in the results of selected indicators of hospitals located in different areas. The assumption is that hospitals located in areas with smaller population are characterized by lower profitability, higher debt and by overall lower financial situation. In the paper data from public sources and from annual reports of hospitals for the year 2013 will be used. Hospitals will be chosen according to the division of the Institute of Health Information and Statistics of the Czech Republic. Keywords; public finance, hospitals, profitability, indebtedness, revenue JEL codes: H2 1 Introduction Modern economy is characterized by dynamic changes. Therefore, the forecasting of financial troubles of companies is difficult and has a high importance for all parties examining the company's financial situation. There are a number of indicators dealing with finances and measuring business performance. Synthetic indicators have been the subject of research since the sixties (Beaver, 1966), (Altman, 1968). Indicators should reflect the issues of the industry for which they are used. The health care sector is broad and dependent on many other industries, such as medical equipment manufacturing or pharmaceuticals. Few studies deal with forecasting and financial distress for specific businesses such as hospitals. Extensive research has been led by George H. Pink, et al. (Pink, 2007) at hospitals in Ontario. The collective searched for and determined 37 key indicators, with the help of a team of 30 researchers from the rank of scientists, physicians experienced in hospital management and professional public. Also Zelleret al. (Zeller, 1997) analyzed the financial factors that should define financial measures to prevent a critical condition. His study was conducted on the data of both profit and nonprofit hospitals and subsequently it led to identification of six characteristics of financial performance. Another view on the comparison of group of hospitals and group of industrial businesses has been conducted by the team led by Professor Chu (Chu, 1991). Their research determined the existence of the same financial ratios for the group of hospitals in comparison with the situation of group of businesses. With costs, efficiency and the profit efficiency of hospitals deal many studies (Hollingsworth and Street, 2006), (Shen et al., 2005), (Herr et al., 2010), (Hajdíková, 2014a), (Bern, 2015). Technical advances and demographic changes are the reason for the increasing demand for health care. Health care costs are growing faster than the sources of financing of the health care system. With considerable economic issues are dealing especially small hospitals (Augurzky, 2010). Hospital with less than 100 beds has a higher probability of closing (Williams et al., 1992), (Lillie-Blanton et al., 1992). An opposite view holds Simpson (Simpson, 1995), who carried out a study on smaller sample of California 222 hospitals in the 90s. McCue (McCue, 1997) finds that hospitals with fewer than 100 beds are less likely to ensure sustained positive cash flow. Augurzky (Augurzky, 2010) analyzed the financial performance of small German hospitals and his findings were that these hospitals have a higher one-year probability of failure and a lower EBITDA profit margin. Smaller hospitals in rural areas don't have worse outcomes than those in urban areas, where there are a number of other hospitals. In the Czech Republic, the hospitals have various legal forms, which, however, has no significant effect on the net profit of hospitals (Hajdfkova, 2013). However, the results show that the impact on the level of net profit may be affected by the size of the hospital of a given legal form (Hajdfkova, 2014b). The aim of this paper is to evaluate and compare the financial situation of hospitals with the following working propositions: PI: there are differences in the results of indicators of hospitals located in various areas P2: the financial situation of hospitals with fewer beds will be worse P3: hospitals with a low number of beds are located in rural areas 2 Methodology and Data The paper contains an analysis of selected financial indicators. These indicators were selected on the basis of scientific methods (table 1). Table 1 List of Financial Indicators for the External Evaluation of the Financial Situation of Hospitals Abbrev. of indicator Indicator Formula Ul ROA net income / total assets U2 current liquidity current assets / current liabilities U3 indicator of overall debt total debt / total assets U4 wage productivity Employee expenses/operating income (sales) Source: Author's own expertise To evaluate the selected indicators, the data from publicly available sources for the year 2013 have been used. Data was acquired from the profit and loss accounts, the balance sheets, from the annual reports of hospitals and from the web portal of Institute of Health Information and Statistics (UZIS, 2016) (MFCR, 2016). Given the objective of the article, only data from hospitals in the Czech Republic are processed. In the Czech Republic there were 10.5 million of inhabitants in 2013. The country is divided into 14 regions. Towards the end of 2013 the Czech Republic had 188 hospitals, which are, according to the bed capacity, owned by 50% by the state. The remaining 50% is owned by non-state sector, of which 25% is owned by counties and municipalities, and 25% by private owners. Hospitals are established with the legal form of contributory organizations, limited liability companies, joint stock companies, charitable organizations and church organizations (UZIS, 2014). Figure 1 Hospital in Individual Regions Source: MFCR (2016), own processing 223 Comparison of hospitals has been done using two multivariate methods - namely it was a simple sum of a sequence and a method of distance from a fictitious object. Simple sum of a sequence: for each criterion (here the indicators) the organizations are ordered based on a value of particular criterion. The hospital with the best value of given criterion is assigned the first position, the next in order is assigned second position, etc. This is done for all criteria under consideration. After that the assigned values are summed up and it holds that the lower the total sum, the better the result of the hospital. The method of distance from a fictitious object is based on measuring a distance of particular organization from a fictitious organization. The latter is constructed as the organization that reaches the best value in all criteria involved in an analysis. It means that in the given set of hospitals the best value for each indicator is identified, and with all of them put together the ideal object is created. Then the distance (di) is computed as dt = Jz^fay - vfJf (1) Where vy is a standardized variable of j-th indicator for i-th hospital, vfj is standardized variable of j-th indicator for fictitious organization; m is the number of indicators. An interpretation of results is based on the fact that the lower the distance the better the position of hospital. 3 Results and Discussion Data from 106 hospitals was used for comparison. The remaining data was excluded for the lack of accuracy or incompleteness. Minimum, maximum and average values of calculated indicators are shown in Table 2. Table 2 Average, Minimum, Maximum Values of Calculated Indicators Average Min Max Ul 0,0496 -0,3073 0,2941 U2 1,7606 0,4466 10,5309 U3 0,4401 0,0374 1,3666 U4 0,5354 0,2034 0,7138 Source: Author's own research and calculations Hospital Rehamedica Žacléř from the Hradec Kralove region reached the lowest in the indicator Ul (ROA). The value of the indicator is negative, which suggests the debts of the hospital are not covered. The highest value in return on assets has reached the Hornická Hospital and Polyclinic Ltd. from the Vysočina region. Minimum and maximum values of the U2 indicator (current liquidity) diverge from common values. The U3 indicator (overall debt) is being kept on the minimum value in the Zlín region by the hospital in Uherský Brod, Ltd. The maximum value is being reached by the PP hospitals, Ltd. in Brandýs nad Labem. The U4 indicator (wage productivity) is maintained on approximately the same value by all hospitals. Hospitals were tested using the multivariate methods. The table 3 shows best four hospitals with the best and four hospitals with the worst ranking. These hospitals were assigned the number of beds and number of inhabitants in the district in which the hospital are located. The largest district is Prague with 1.246 milions of inhabitants and the smallest is Jesenik in the Olomouc Region with only 40 thousands of residents. 224 Table 3 Order of Best and Worst Ranked Hospitals Number of Simple sum of Number residents in Hospital the sequence of beds the district Hospital Polička (Pardubice region) 17 120 104971 Hospital Podlesí (Moravskoslezský region) 23 167 212448 P-P Klinika Kladno (South Bohemia region) 24 /\ 159984 Mediterra Sedlčany (Central Bohemia region) 31 103 113905 Kladno hospital (Central Bohemia region) 69 622 159984 Karlovy Vary hospital (Karlovy Vary region) 72 1134 117868 Hospital of st. Zdislava (Vysočina region) 76 106 118646 Rehamedica Žacléř (Hradec Králové region) 99 119900 Source: Simple sum of the sequence - author's own calculations; Number of beds (UZIS, 2013); Number of residents in the district (CZSO, 2013) 4 Conclusions The paper analyzed the financial indicators of hospitals in the Czech Republic regarding the return on assets (ROA), current liquidity, overall debt and wage productivity. These indicators have been quantified due to a lack of data for 106 hospitals in 2013. A uniform ranking of hospitals in terms of worst and best result was chosen. It could be assumed that hospitals with the best ranking will score almost the best for all indicators and vice versa. The best hospital, regarding the order of all indicators, ranked 17th and hospital with the worst record ranked the 99th place. It may be due to the selection of indicators or due to the financial situation that might have occurred in the hospital in the period under review. Results of indicators were tested for the hospital's location in a region. No close match of resulting values of hospitals located in one region has been found. Assumptions of differences existing in the indicator's outcomes of hospitals located in different regions were confirmed. In the research, the regions in the Czech Republic and selected size of the area by population in the district were compared. Mild consensus, however, occurred when comparing the values of the indicators of hospitals to the number of beds. Hospitals with more beds were arranged rather between hospitals with poorer financial results. In contrary, hospitals with fewer beds showed better value indicators. Hospitals with fewer beds are spread throughout the Czech Republic and are located even in districts with a high population density. The surprising finding is that hospitals with fewer beds have satisfactory results. The result is similar to research of Simpson (Simpson, 1995). This finding will be confirmed by further research using other research methods on a longer period of time. The number of hospitals in the surveyed region will be added to the relation with the assessed indicators. It is also necessary to mention the limitations of the research, because some hospitals are grouped in the holding company (South Bohemia Hospitals Holding) and health care concepts of hospitals in this group may be different from a hospital deciding separately with the same number of beds and similar region. Acknowledgments This paper was funded from the resources of the project 12/2016, a negotiated contract between the researchers and IGA (Internal Grant Agency), University of Economics in Prague. References Altman, E. I. (1968). Financial Ratios, Discriminant Analysis and the Prediction of Corporate Bankruptcy. The Journal of Finance, vol. 23(4), pp. 589-609. Augurzky, B., Schmitz, H. (2010). Is there a Future for Small Hospitals in Germany? Ruhr economic papers, vol. 198. 225 Beaver, W. H. (1966). Financial Ratios as Predictors of Failure. Journal of Accounting Research, vol. 4, pp. 71. Bern, A., Michalskí, G. (2015). Hospital Profitability vs. Selected Healthcare System Indicators. In: Central European Conference in Finance and Economics (CEFE2015). Košice: Technical University of Košice, pp. 52-61. Chu, D. K., Zollinger, T. W., Kelly, A. S., Saywell, R. M. (1991). An Empirical Analysis of Cash Flow, Working Capital, and the Stability of Financial Ratio Groups in the Hospital Industry. Journal of Accounting and Public Policy, vol. 10(1), pp. 39-58. CZSO (2014). Statistical Yearbook of the Czech Republic - 2013. Retrieved from: https://www.czso.cz/csu/czso/statistical-yearbook-of-the-czech-republic-2013- 653m9thdnl. Hajdíková, T., Komárkova, L., Pirožek. P. (2013). Financial and operational performance of hospitals in the Czech Republic. In: Proceedings of the 10th International Scientific Conference. Brno: Masarykova univerzita, pp. 120-125. Hajdíková, T., Komárkova, L., Pirožek. P. (2014a). The Issue of Indebtedness of Czech Hospitals. In Deev, O., Kujarová, V., Krajíček, 1, ed., European Financial Systems 2014. Brno: Masarykova univerzita, pp. 230-235. Hajdíková, T., Komárkova, L., Pirožek. P. (2014b). Komparace právních forem nemocnic působících v prostředí zdravotnických služeb v ČR. In: Sborník recenzovaných příspěvků z mezinárodní vědecká konference Hradecké ekonomické dny 2014, Díl I., Ekonomický rozvoj a management regionů. Hradec Králové: Gaudeamus, pp. 254-261. Herr, A., Schmitz, H., Augurzky, B. (2010). Profit efficiency and ownership of German hospitals. Health Economics, vol. 20(6), pp. 660-674. Hollingsworth, B., Street, A. (2006). The market for efficiency analysis of health care organisations. Health Economics, vol. 15(10), pp. 1055-1059. Lillie-Blanton, M., Felt, S., Redmon, P., Renn, S., Machlin, S., Wennar, E. (1992). Rural and urban hospital closures, 1985-1988: Operating and environmental characteristics that affect risk. Inquiry, vol. 29, pp. 332-344. McCue, M. J. (1997). Small hospitals with positive cash flow: Why are they winners? Medical Care Research and Review, vol. 54, pp. 32-60. MFCR (2016). ARES - ekonomické subjekty. Retrieved from: http ://wwwi nfo. mfcr.cz/a res/a res_es. html. cz. Pink, G. H., Daniel, I., Hall, L. M., Mckillop, I. (2007). Selection of Key Financial Indicators: A Literature, Panel and Survey Approach. Healthcare Quarterly, vol. 10(1), pp. 87-96. Shen, Y. C, Eggleston, K., Lau, 1, Schmid, C. (2005). Hospital ownership and financial performance: A quantitative research review. NBER Working Papers 11662. Simpson, J. (1995). A note on entry by small hospitals. Journal of Health Economics, vol. 14(1), pp. 107-113. ÚZIS (2013). Kardexy 2013. Retrieved from: http://www.uzis.cz/cr-kraje z 2013. ÚZIS (2014). Demografická situace v České republice v roce 2013. Retrieved from: http://www.uzis.cz/rychle-informace/demograficka-situace-ceske-republice-roce-2013. Williams, D., Hadley, J., Pettengill, 1 (1992). Profits, community role, and hospital closures: An urban rural analysis. Medical Care, vol. 30, pp. 174-187. Zeller, T. L, Stanko, B. B., Cleverley, W. O. (1997). A New Perspective on Hospital Financial Ratio Analysis. Healthcare Financial Management, vol. 51(11), pp. 62-66. 226 Analysis of Various Entrepreneurial Activities and their Development in the Czech Republic from 2008 to 2015 Eva Hamplová1, Jaroslav Kovárník2, Pavel Jedlička3 1 University of Hradec Králové Faculty of Informatics and Management, Department of Economics Rokitanského 62, 500 03 Hradec Králové, Czech Republic E-mail: eva.hamplova@uhk.cz 2 University of Hradec Králové Faculty of Informatics and Management, Department of Economics Rokitanského 62, 500 03 Hradec Králové, Czech Republic E-mail: jaroslav.kovarnik@uhk.cz 3 University of Hradec Králové Faculty of Informatics and Management, Department of Economics Rokitanského 62, 500 03 Hradec Králové, Czech Republic E-mail: pavel.jedlicka@uhk.cz Abstract: This contribution analyses the number of newly established as well as closed down entrepreneurial units in the Czech Republic between 2008 and 2015 with regard to the legal form of business as well as the field of business activity. The aim is to use the non-public sources of Czech Statistical Office and assess the trend in the number of entrepreneurial units in the Czech Republic. The methods of comparative analysis and trend analysis were used to monitor individual types of business activities as for the number of established and closed entrepreneurial units. The types of economic activities with high entrepreneurial activity were determined by means of net balance and average annual growth rate. The contribution reacts to the conclusions of a European Commission study Flash Eurobarometer (2012), which assesses the Czech Republic as one of four EU countries where a growing number of people prefers self-employment to employment. The authors consider the differences among various entrepreneurial activities and among legal forms of business to be a starting point for a discussion about support to particular fields of business, about private investment and forms of financial flows on both national and international levels. Keywords: business, business environment, national economy JEL codes: L25, L26, O10 1 Introduction Entrepreneurship satisfies unfulfilled needs of people, supports regional development and builds human and intellectual capital of individuals involved in entrepreneurial activities (Uhlaner, 2010). Moreover, entrepreneurship contributes to employment creation, productivity and economic growth, thus corroborating the relevance of entrepreneurship for the world's economies (Lukes, et al., 2013). It is common to define entrepreneurship as the occupational choice to work for one's account and risk. At the individual level researchers have examined a variety of questions such as: Who is more likely to become an entrepreneur? Are there particular personality traits, childhood, adolescent or adult experiences of the person, or specific motives that are associated with entrepreneurship (Rauch & Frese, 2007; Gorgievski, 2010)? Why are some entrepreneurs more successful than others? Are entrepreneurs different than other people? Do they have different personalities? Do they act differently? Entrepreneurs pursue their dreams of developing successful new ventures regardless of all the obstacles and barriers to be faced. One of the interesting questions raised in research carried out on entrepreneurship is, what helps people to undertake entrepreneurial activity, and to continue on with it to achieve its successful outcome, even if it is such a demanding task? Recent analyses suggest that the personality of the 227 entrepreneur is important for successful business start-up and growth (Laguna, 2010). The entrepreneur needs to be alert, sensitive to market needs and inefficiently used resources, and have the courage to make the decision to seize the opportunity he has just noticed (Lukes, Jakl, 2012). Since 2008, Europe is suffering from the consequences of the most serious economic crisis that has not been experienced over the past 50 years. For the first time in Europe, there are more than 2.5 million unemployed and small and medium enterprises in most Member States are still unable to get back to its pre-crisis level (Entrepreneurship 2020 Action Plan, 2013). Since 2004, the percentage of people who prefer self-employment to the relationship of employment decreases in 23 of the 27 EU Member States (Entrepreneurship 2020 Action Plan, 2013). While three years ago, 45 % of Europeans preferred the self-employed, now this figure has fallen to 37 % (Eurobarometer Survey on Entrepreneurship, 2012) unlike in the US and China, where the percentage is much higher: (51 % or 56 %). Countries where preferences of self-employment increased during the period 2004-2012 are the Czech Republic (from 30% to 34%), Latvia (from 42% to 49%), Lithuania (from 52% to 58%) and Slovakia (30 to 33%) (Eurobarometer Survey on Entrepreneurship, 2012). The authors consider the differences among various entrepreneurial activities and among legal forms of business to be a starting point for a discussion about support to particular fields of business, about private investment and forms of financial flows on both national and international levels. The aim is to support both national and European interest based on the European Commission action plan - Reigniting the entrepreneurial spirit in Europe. 2 Methodology and Data Internal database of Czech Statistical Office was used during the preparation of this article. Records about the number of SMEs in 2008 - 2015 were obtained thanks to Information Services Unit of the Czech Statistical Office and Business Statistics Coordination and Business Cycle Surveys Department of the Czech Statistical Office. Obtained data about the number of newly established and closed down entrepreneurial units are arranged into time series and consequently evaluated by the analysis of average growth rate and by index of net increase (decrease) in the number of newly established entrepreneurial units during analyzed period 2008 - 2015. For identification of newly established entrepreneurial units has been used an abbreviation EST (established), and for identification of closed down units an abbreviation CLO (closed). For mutual evaluation has been used BALANCE (EST - CLO). For identification of different legal forms of business has been used data in following order: • Self-Employed Person (SEP) o (A) Private entrepreneurs in business under the Trade Act o (B) Agricultural entrepreneurs - natural persons o (C) Others • Legal Person o LLC o Joint-stock companies o Limited partnerships o General commercial partnerships o Others With respect to various entrepreneurial activities have been used statistical nomenclature CZ-NACE on the A - T levels. The methodology used to tackle the subject has the nature of basic theoretical research which is focused on the analysis of the structure, linkages and relations of the studied subject. From the methodology point of view trend, system and qualitative analysis is used to achieve the objective. 228 3 Results and Discussion The aim of this paper is to describe business activity of entrepreneurial units in the Czech Republic. The legal form criterion has been used for the analysis in the first part of this article. The legal form is the base for the discussion, whether entrepreneurs establish new units as self-employed persons or as legal persons. As far as self-employed persons are concerned, the huge advantage in the conditions of the Czech Republic is quick and both administratively and financially establishment. The disadvantages are the number of the owners, the size of the unit, and the liability, where the entrepreneur has to be liable also with own personal property, which includes the property of the family too. This risk seems to be really high, which is probably the reason why the number of newly established units of this legal form has significantly lower growth rate than in the case of legal persons. Figure 1 shows that the development of the number of newly established units has relatively steady progress, in contrast with the development of the number of closed down units. Figure 1 Number of Newly Established and Closed Down Units between 2008 - 2015 in the Legal Form Self-Employed Person (SEP) and Annual Growth Rate of SEP -60000 Source: Own research based on Czech Statistical Office (2015) Higher activity occurred in 2010, but the number of newly established entrepreneurial units has been decreasing since that year. The average growth rate of newly established units of this legal form presents - 2.18% in analyzed period. With respect to the number of closed down self-employed persons, the development is significantly different. There are higher amounts of closed down units in the years 2009 and 2013, where this development has influenced the average growth rate on 20.41% in the period 2008 -2015. With regard to different growth rates of newly established entrepreneurial units and of closed down entrepreneurial units, it is possible to see decreasing development of the balance of net newly established entrepreneurial units. However, this development has not yet impact on the decrease of total amount of entrepreneurial units, because net index of the establishment of new entrepreneurial units for all SEP is 111.51% (Table 1). Nevertheless, this situation can change in the following years. The reality of the establishment of new entrepreneurial units in the legal form of Self-Employed Persons is supplemented by the Figure 2, where it is possible to analyze the development of the most numerous group of SEP in the Czech Republic, namely Private entrepreneurs in business under the Trade Act. This group presents more than 90% of total SEP and its index of net growth is 126.28%. This can be explained that for every closed down unit has been created 1.25 new units in the period 2008 - 2015. 229 Table 1 Different Legal Forms of Business According to the Creation of Net Balance of Newly Established Entrepreneurial Units (by Index of Net Increase/Decrease) Self-Employed Person 2008 2015 2014 2015 EST/CLO 2008 - 2015 A Private entrepreneurs in business under the Trade Act 1.26 1.13 492 524/ 390 025 B Agricultural entrepreneurs -natural persons 0.16 3.06 8 571/ 53 773 C Others 0.92 0.99 29 059/ 31 645 TOTAL 1.12 1.15 530 154/475 443 Source: Own research based on Czech Statistical Office (2015) Figure 2 Number of Newly Established and Closed Down Units between 2008 - 2015 in the Legal Form Private Entrepreneurs in Business under the Trade Act (A) and Annual Growth Rate -60000 Source: Own research according Czech Statistical Office (2015) Next group with significant influence on the entrepreneurial environment in the Czech Republic is group of legal persons. Table 2 Different Legal Forms of Business According to the Creation of Net Balance of Newly Established Entrepreneurial Units (by Index of Net Increase/Decrease) Legal Person 2008 - 2015 2014 - 2015 EST/CLO 2008 - 2015 LLC 4.88 4.88 182 218/37 330 Joint-stock companies 1.85 1.43 7 716/4 179 Limited partnerships 0.99 0.50 150/152 General commercial 0.42 0.20 557/1 336 partnerships Others 1 573/30 TOTAL 4.47 4.45 192 214/43 027 Source: Own research based on Czech Statistical Office (2015) The establishment of such kind of unit is administratively more difficult than in case of SEP, there are also more complicated legal and accounting rules. On the contrary, the most frequent legal form, Limited Liability Company, has no requirement on liability with own property, there can be one or more owners, and since 2014 there is no requirement on the beginning contributed capital. The analysis of LLC units shows higher activity in 2013, there are no significant fluctuations in the period 2008 - 2015, and the average growth rate is 2.17%. The tendency of closed down units is stable too, with average 230 growth rate 7.42%. From Figure 3 is obvious that the amount of LLCs has been growing every year, but with respect to the number of newly established companies is the net balance of newly established companies 488.13% in the period 2008 - 2015. That means that there are almost five new companies on every closed down company. Figure 3 Number of Newly Established and Closed Down Units between 2008 - 2015 in the Legal Form Limited Liability Company (LLC) and Annual Growth Rate 30 000 2008 2009 2010 2011 2012 2013 2014 2015 Source: Own research according Czech Statistical Office (2015) The second part of this article analyzes the establishment and termination of companies according to economic activity. It is assumed that analyzed activity is the same activity which is declared to be the main activity of the entrepreneurial unit. From the Table 3 is obvious that the field of business Electricity, gas, steam and air conditioning supply has been the most active based on the index of net balance of newly established entrepreneurial units. Index 11.76 means that during the period 2008 - 2015 have been created 11.76 new companies on every one closed down in this field. Yearly indexes of net balance were significant in this field especially in the years 2008 (12.5), 2009 (48.6), 2010 (18.3), 2012 (21.0), and 2013 (21.3). In the year 2009 were created 1,069 new companies in this field, while only 22 were closed down. Other intensively developing fields are Real estate activities (index 4.47), Arts, entertainment and recreation (2.93), Other service activities (2.79), Information and communication (2.59), and Water supply; sewerage, waste management and remediation activities (2.06). On the other hand, higher number of companies have been closed down in the fields Administrative and support service activities (0.84), Transportation and storage (0.71), Agriculture, forestry and fishing (0.61), and Activities of extraterritorial organizations and bodies (0.31). Table 3 Different Fields of Business According to the Creation of Net Balance of Newly Established Entrepreneurial Units 2008 - 2015 EST/CLO A Agriculture, forestry and fishing 0.61 40 720/66 695 B Mining and quarrying 1.49 268/180 C Manufacturing 1.29 83 273/64 513 D Electricity, gas, steam and air conditioning supply 11.76 5 220/444 E Water supply; sewerage, waste management and remediation activities 2.06 4 839/2 350 F Construction 1.16 86 331/74 711 G Wholesale and retail trade; repair of motor vehicles and motorcycles 1.31 203 749/155 376 H Transportation and storage 0.71 12 431/17 598 I Accommodation and food service activities 1.09 35 586/32 678 231 J Information and communication 2.59 19 882/7 682 K Financial and insurance activities 1.1 27 373/24 947 L Real estate activities 4.47 72 537/16 213 M Professional, scientific and technical activities 1.77 104 476/58 971 N Administrative and support service activities 0.84 11 494/13 604 O Public administration and defense; compulsory social security 1.25 372/297 P Education 1.87 11 051/5 907 Q Human health and social work activities 1.49 10 158/6 819 R Arts, entertainment and recreation 2.93 18 356/6 262 S Other service activities 2.79 69 667/25 012 T Activities of households as employers; undifferentiated goods- and services-producing activities of households for own use 3/0 U Activities of extraterritorial organizations and bodies 0.31 12/39 TOTAL 1.46 Source: Own research according Czech Statistical Office (2015) 4 Conclusions This article used the data from Czech Statistical Office for creation of survey of development of establishment new entrepreneurial units in the Czech Republic in the period 2008 - 2015. The analysis of business activity can be used for revealing disparities among each legal forms, and also for revealing tendencies in fields of business in the Czech Republic in analyzed period. Since 2014, new Civil Code of Law is in charge in the Czech Republic for all entrepreneurs. This change of legislation brought simplification for establishment of one legal form of business, namely LLC. This article dealt with the relations between newly established companies and closed down companies according to the legal form. For LLCs is the total net balance of newly established entrepreneurial units 4.88. In the years 2014 (5.64) and 2015 (4.34) was the relations between newly established LLCs and closed down LLCs on really high level, therefore it can be assumed that the change of legislation is appreciated by the entrepreneurs. On the contrary, the SEP entrepreneurs have not as high index as LLC. Total index of net newly established units' growth during analyzed period is only 1.12. Moreover, there were years with this index lower than 1 during analyzed period of time. In the year 2009 (0.75) and 2013 (0.59) were the amounts of closed down entrepreneurs higher than the amounts of newly established units. It is obvious that this development has no correlation with the change of legislation. The entrepreneurs have no tendency for closing down their entrepreneurial activity only because of the change of legal form on LLC, where there is no need of beginning contributed capital according to the new legislation, but at the same time there is no need for liability with personal property. This brings the questions for discussion, how are entrepreneurs in the Czech Republic open for entrepreneurship, how they see the entrepreneurial environment, and how are they satisfied with self-employment. The analysis of entrepreneurial activity according to the field of business shows that during 2008 - 2015 has been very significant field Electricity, gas, steam and air conditioning supply, where on every one closed down unit have been created almost 12 new units. However, this tendency has been stopped in last two years. Other important fields are Real estate activities, Arts, entertainment and recreation, Other service activities, and Information and communication. On the other hand, on the decline are fields Activities of extraterritorial organizations and bodies, Agriculture, forestry and fishing, and Transportation and storage. 232 Acknowledgments This article is prepared with the support of the Internal project of Faculty of Informatics and Management at the University of Hradec Králové. References Czech Statistical Office (2015). Births and deaths of businesses. Database of the Statistical Business Register. Not public, internal document. European Commission, (2012). Eurobarometer Survey on Entrepreneurship. Retrieved from: http://ec.europa.eu/enterprise/policies/sme/facts-figures-analysis/eurobarometer/ index_en.htm>. European Commission (2013). Entrepreneurship 2020 Action Plan. Retrieved from: http://ec.europa.eu/enterprise/policies/sme/entrepreneurship-2020/index_en.htm. Gorgievski, M. J. (2010). Entrepreneurial motivation: Independence, money, self-realization and passion for work. In M. Lukes & M. Laguna (Eds.), Entrepreneurship, a Psychological Approach, Prague: Oeconomica Publishers, pp. 55-72. Laguna, M. (2010). Positive Psychology Inspirations for Entrepreneurship Research. In M. Lukes & M. Laguna (Eds.), Entrepreneurship, a Psychological Approach, Prague: Oeconomica Publishers, pp. 73-88. Lukeš, M., Jakl, M., (2012). Entrepreneurschip in the Czech Republic. Praha: Oeconomica. Lukeš, M., Zouhar, J., Jakl, M., Očko, P. (2013). Factors Influencing Entrepreneurial Entry: Early-Stage Entrepreneurs in the Czech Republic. Politická ekonomie, 61(2), pp. 229-247. Rauch, A. & Frese, M. (2007). Born to be an entrepreneur: Revisiting the personality approach to entrepreneurship. In J. R. Baum, M. Frese & R. Baron (Eds.). The psychology of entrepreneurship, London: Lawrence Erlbaum Associates, pp. 41-66. Uhlaner, L., Thurik, R. (2010). Postmaterialism Influencing Total Interpreneurial Activity Across Nations, In: Entrepreneurial and Culture, Berlin: Springer-Verlag Berlin, pp. 301-328. 233 Development and the Current Situation of the Mortgages for the Czech Households Martina Hedvicakova1, Libuše Svobodova2 1 University of Hradec Kralove Faculty of Informatics and Management, Department of Economics Rokitanskeho 62, 500 03 Hradec Kralove III, Czech Republic E-mail: martina.hedvicakova@uhk.cz 2 University of Hradec Kralove Faculty of Informatics and Management, Department of Economics Rokitanskeho 62, 500 03 Hradec Kralove III, Czech Republic E-mail: libuse.svobodova@uhk.cz Abstract: The paper is focused on the specialized financial product, esp. on mortgages that are the most often used products in the field of own housing financing in the Czech Republic. These products provide selected banks on the Czech market. The aim of the paper is to analyze the situation on the financial market focused on the mortgages. The organization of the paper is as follows: firstly a theoretical background with a review of the literature is provided, then research methodology is described, the key part brings results of development of mortgages according to the purpose of use since 2005, development of interest rates and the analysis of the current situation on the mortgage market. Interest rates on mortgage loans have on the Czech market downward trend. Statistical data suggest that people are not afraid to borrow money after the economic crisis and invest more in housing funds. Hypoteční banka was classified as the main player on the market closely followed by Česká spořitelna and Komerční banka. The article is based on primary and secondary sources. A detailed research together with the analysis and critical assessment of accessible materials will enable to identify the main objectives in the field of study. The analysis of the initial state will consequently enable to identify the key factors and knowledge. Keywords: mortgage loan, interest rate, bank, households JEL codes: E43, E44, E52 1 Introduction At the present time, there is stiff competition amongst banks on the financial market based on marketing and information basis. The goal is not only to get new clients but especially to retain the current ones. Traditional banks should beware of new „low-cost" banks on the Czech market and should maintain their sizeable amount of clients. Indebtedness is globally enlarging every year and residents take full advantage of loans. Mortgages and loans from building societies are considered to be the most frequent products. Mortgages mature between 5 and 40 years. Most of the consumers conclude a contract for 20 - 30 years. The loans from building societies usually last up to 10 years. From 1999 to 2013, U.S. mortgage debt doubled before contracting sharply. I estimate mortgage inflows and outflows that shed light on the sources of volatility. During the boom, inflows from real estate investors tripled, far outpacing other segments such as first-time homebuyers. During the bust, a collapse in inflows keyed the debt decline, while an expansion of outflows due to defaults played a more minor role. Inflow declines partly reflect a dramatic falloff in first-time homebuying, especially for low credit score individuals. Bhuttas (2015) analysis helps support the notion that the differential decline by credit score reflects markedly tightened credit supply. Recourse mortgages increase the cost of default but also lower equity and increase payments. The effect on default is nonmonotonic. Loan-to-value (LTV) limits increase equity and lower the default rate, with negligible effects on housing demand. Combining recourse mortgages and LTV limits reduces the default rate while boosting housing 234 demand. Together, they also prevent spikes in default after large declines in aggregate house prices. (Hatchondo et all, 2015) Dia and Menna (2016) find that resource costs explain a large share of bank interest margins, which generates a floor for the interest rate on loans. At the present time of extremely low interest rates and slow economic growth, the government and central bank of each country strive to boost their economies. In March 2016, The European Central Bank (ECB) also decided to boost the economy and increase a low rate of inflation by an unexpectedly strong mixture of policies. The ECB decreased all of its interest rates, intensified the purchase of bonds and it offered a new cycle of low cost loans to banks. The basic rate was unexpectedly decreased from historical minimum of 0.05 percent to 0 but according to analyst it is basically insignificant. Monthly bond purchases are increased by €20 milliard to €80 milliard (2.2 billion Czech Crowns). Along with the purchase of state bonds, the ECB is going to purchase bonds with a high rating issued by non-bank financial institutions. Banks in the Eurozone will be allowed to opt for a new 4-year programme. „A bank that is very active in granting loans to the real economy can borrow more than a bank that concentrates on other activities," ECB President Mario Draghi said. As supposed, the deposit rate was decreased to minus 0.40 per cent from current 0.30 per cent. At the time of negative rates, banks are obliged to pay for deposits parked in the ECB, which should make them lend a higher amount of money. The ECB also decreased the rate on the marginal lending facility which equals 0.25 per cent and offers overnight credit to banks by ECB. (ČTK, 2016) Whether the real interest rates respond in a different manner to macroeconomic news at the zero lower bound (ZLB) as compared to the case away from the ZLB is essential for assessing the effectiveness of government policies and the validity of the policy implications of New Keynesian models at the ZLB. By using an identification strategy based on heterogeneity, Zhang (2016) find that at the ZLB, monetary policy news is less effective in affecting short- and medium-term real rates and its effect dies off faster. The goal of the article is to analyse the situation on the Czech mortgage market with an emphasis on the development of mortgage loans from 2005 to the first quarter of 2016. The overview of the history and use of mortgages in the world is presented together with the operation of national and commercial banks. 2 Methodology and Data According to the Act No. 190/2004 Coll., as subsequently amended (including the Act No. 137/2014 Coll.), and Article 28, "a mortgage loan is a loan whose redemption, including appurtenances, is secured by a right of pledge over real estate, where the claim arising from the loan does not exceed twice the amount of the mortgage value of the mortgaged real estate. A loan is considered to be a mortgage loan from the day when the right of pledge takes legal effect. For the purposes of coverage of mortgage bonds, the receivable arising from a mortgage loan or a part thereof may be first used on the day when the issuer of mortgage bonds learns about the legal effect of the establishment of the right of pledge over the real estate." This paper is based on the analysis of the literature and articles on housing. Important information is available on the official websites of each bank institution, the Czech National Bank, the Czech Statistical Office, as well as of some financially oriented portals like Hypoindex, GolemFinance, Hypoteční banka etc. The obtained data were further sorted, processed in custom tables, clearly charts, and further analysed to provide a basic overview of the relevant problem area. Mortgage loans are a widely studied subject and there is an extensive literature dealing with mortgages on the Czech market, which enables us to gain insight into the selected areas of mortgages from multiple sources which deepen or complement the issue. The 235 twenty-year development of mortgage loans in the Czech Republic is included in order to depict the situation on the Czech market. 3 Results and Discussion 20-year development of mortgages in the Czech Republic The development of mortgages is possible to divide into four stages: • The first stage from 1995 to 1999 - It was in 1995 when mortgages entered in banking industry. State subsidies that helped launch mortgages were granted in this period. Distribution of the product and discovering the risks were gradually developed by banks. • The second stage from 2000 to 2005 - Origin of the mortgage market and competition. Owing to innovation in products and new ways of distribution, mortgages became much more available to clients. • The third stage from 2006 to 2010 - Recession occurred in the autumn of 2008 after the growth of tens per cent and lasted until September 2010. • The fourth stage from 2011 to 2015 - The period between 2011 and 2015 is considered to be recovery from the crisis. According to the statistics of Hypoindex, mortgages are commonly used and the interest rates have reached the lowest recorded minimum of 1.97% p.a. It was launched a trend of clients to conclude mortgages on houses that will be leased in the future. The Czech National Bank (CNB) due to record numbers of mortgages and rising risk for commercial banks issued a recommendation that the length of the maturity of mortgage loans do not exceed 30 years. (Hypoindex.cz, April 2016) The graph below (Fig. No. 1), traces the evolution of mortgage interest rates from the beginning of mortgage loans as a new financial product. Figure 1 The Development of Average Interest Rate in the Czech Republic (in %) 1,5 rnrn^^LOioi^i£)r^r^oooocjicjiOO # / # f # # # f f f / Source: MMR.cz, 2016, own processing Although in the period of 2005 and the first quarter of 2016 there was relatively significant fluctuating progress in number of signed contracts and provided funds for mortgages, there was no dramatic fluctuation in terms of the value of mortgages per one contract. The average value of mortgage per one contract was increased by 476 thousand Czech Crowns, which is by 34% more comparing with the year 2005. Mortgages for real estate purchases influenced the increase in value of mortgages of 42%, increase of 577 thousand Czech Crowns. Mortgages for construction influenced the increase in value of mortgages of 33%, increase of 503 thousand Czech Crowns. The other mortgages had the smallest effect on total growth. The increase in value of mortgages was "only" 27%, which is 348 thousand Czech Crowns while the amount of money was increased by 217% (see figure 4). 239 Figure 4 Development of the Value of Mortgage per 1 Contract in the Period of 2005 and the First Quarter of 2016 (in CZK) ■ Others ■ Construction Purchase 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 l.Q 2016 Source: MMR.cz, 2016, own processing Discussion The topic no. 1 in the mortgage market is a new regulatory rules that directed us from the EU. The Ministry of Finance is preparing a new law on loans for consumers. The New Testament comprehensively covers all types of consumer loans - from classic cash loans, credit cards, purchases of goods on installment to mortgages and other loans. The law on consumer credit did not concern it yet (ie. loans on housing and so-called microloans). If we focus on mortgages, there will significantly change the conditions for early repayment of mortgages, which will be easier to repay. Roles and responsibilities of intermediaries will also increase. The quality of services will improve thanks to the certifications for those financial advisors who will want to offer mortgages in the future. Questions for discussion are whether the possibility of cheaper early repayment of the mortgage loan may be the motive for the banks for offering of more expensive mortgages and to what extent they will respond? What will have higher influence on a possible increase price of mortgage? Regulation and the possibility of early repayment or will be decisive monetary policy of the Czech National Bank and the related value of money? The question arises whether it is appropriate that the EU and the state regulate the credit market? 4 Conclusions Czech mortgage market achieved record results in 2015. The volume of closed mortgage loans exceeded according to the Ministry of Regional Development 184 billion crowns, which is up about 41 billion more than in the previous year. The average amount of mortgage loan reached 1.87 million last year, 9.3% more than in 2014. It is because of the growing prices of residential real estate. HB Index, which tracks the price development of apartments, houses and land since 2010 had showed an annual increase in prices of flats by 6.2 percentage points, family houses 2.4 percentage points and land 5.3 percentage points. (Hypoteční banka, 2016) Interest rates on mortgages are currently under two percent, and even fell in April 2016 to a record value 1.94% (see. Fig. 1). The average interest rate has declined, but the number of closed mortgages in April 2016 fell to 9,042. It is 333 less than in March. As well as have declined volumes. People in April 2016 issued a total of 17.06 billion CZK on mortgage loans. The volume was about 846 million CZK higher in March. (Hypoindex, 2016) Interest rates will remain at low levels in 2016 according to predictions. Further decline would occur, however, growth is more expected. Among the reasons why can remain low interest rates in 2017, it may be noted that the Czech national bank confirmed the 240 commitment to remain in the exchange rate to end 2016 and also left its interest rates unchanged. Czech national bank also had not introduced a negative interest rate as have done the European Central Bank. In further research will be analysed what will have influence on higher and possible increase price of mortgage? New regulation or will be decisive monetary policy of the Czech National Bank and the related value of money? Acknowledgments This paper is written with financial support from Internal grant project from Faculty of Informatics and Management at the University of Hradec Králové. References Act No. 190/2004 Coll., (including the Act No. 137/2014 Coll.), Czech Republic. Bhutta, N. (2015). The ins and outs of mortgage debt during the housing boom and bust. Journal of Monetary Economics, vol. 76, pp. 284-298. Czech national bank (2016). Commentary on MFI interest rates, April 2016, Retrieved from: https://www.cnb.cz/en/statistics/money_and_banking_stat/harm_stat_data/mfi_koment ar.html Czech national bank, ARAD (2016). Retrieved from: http://www.cnb.cz/cnb/STAT.ARADY_PKG.VYSTUP?p_sestuid = 19522&p_uka=6,7&p_stri d=AAD&p_sort=2&p_od=200401&p_do=201604&p_period = l&p_des=50&p_format=4&p _decsep=.&p_lang = EN. ČTK (2016). Překvapení pro trhy. ECB podpoří evropskou ekonomiku nečekaně důraznými kroky. Retrived from: http://zpravy.aktualne.cz/ekonomika/prekvapeni-pro-evropske-trhy-ecb-srazila-hlavni-urokovou-saz/r~20a5d590e6c411e59c4a002590604f2e. Dia, E., Menna, L. (2016). Productivity shocks, capital intensities, and bank interest rates. Journal of Macroeconomics, vol. 48, pp. 155-171. Hatchondo, J.C., Martinez, L, Sanchez, J.M. (2015). Mortgage defaults. Journal of Monetary Economics, vol. 76, pp. 173-190. Hovorka, 1 (2016a). Confirmed. Banks have negotiated the most mortgages in history in the last year. Hospodářské noviny. Retrieved from: http://zpravy.aktualne.cz/finance/potvrzeno-banky-loni-sjednaly-nejvice-hypotek-v-historii/r~4ad89578cff911e5a8d7002590604f2e/. Hovorka, J. (2016b). Hypoteční banka defended in the last year its leading position on the market. Česká spořitelna after rapid growth get close. Hospodářské noviny. Retrieved from: http://www.golemfinance.cz/cz/1035.hn-hypotecni-banka-loni-uhajila-pozici-jednicky-na-trhu-ceska-sporitelna-se-ji-ale-po-rychlem-rustu-tesne-priblizila. Hypoindex (2016). Hypoindex vývoj. Retrieved from: http://www.hypoindex.cz/hypoindex-vyvoj/. Hypoteční banka (2016). Komentář Hypoteční banky k vývoji hypotečního trhu. Retrieved from: https://www.hypotecnibanka.cz/o-bance/pro-media/tiskove-zpravy/obchodni-vysledky-cr/komentar-hypotecni-banky-k-vyvoji-hypotecniho-trhu. Ministry of Finance (2016). Macroeconomic Forecast - April 2016. Retrieved from: http://www.mfcr.cz/en/statistics/macroeconomic-forecast/2016/macroeconomic- forecast-april-2016-24521. MMR.cz (2016). Mortgage loans for the years 2002 to 1Q 2016. Retrieved from: http://www.mmr.cz/getmedia/0be62756-63a5-4bb0-94f3-2f49c3093e9c/Hypotecni- uvery-za-roky-2002-az-lQ-2016,-k-31-3-16.pdf?ext=.pdf. Ostatek, L. (2010). Mortgage after 15 years: How was born the Czech mortgage market? Prague: Hypoindex.cz. Retrieved from: http://www.hypoindex.cz/hypoteky-po-15-letech-jak-se-zrodil-cesky-hypotecni-trh/. 241 The Process of Harmonization of Accounting in the Czech Republic Irena Honková University of Pardubice Faculty of Economics and Administration Studentská 95, 532 10 Pardubice 2, Czech Republic E-mail: irena.honkova@upce.cz Abstract: The article was written in response to the amendment to Decree No. 250/2015 Coll., which came into effect from January 1, 2016. This amendment shows the harmonization of the Czech accounting system with the International Financial Reporting Standards (IFRS), e.g. in reporting of extraordinary income and expenses, of formation expenses, or in valuation of inventory produced internally. Consequently, a survey has been conducted, in which 24 persons having experience with the accounting under the IAS/IFRS commented on how they coped with the differences between the Czech Accounting Standards (CAS) and the IAS/IFRS. It has been found that these companies normally keep the accounts simultaneously in both systems (CAS and IAS/IFRS). However, some companies decided to keep the accounts only in one accounting system, supplemented with the differences between the two systems. It is undisputed that the process of harmonization of accounting systems continues. However, the final unification of our national system with the European one has not happened so far. There is a need to harmonize accounting and tax regulations as well. Key words: International Financial Reporting Standards (IFRS), Czech accounting standards (CAS), harmonization of accounting JEL codes: M48 1 Introduction In Europe, along with economic globalization, there is an ongoing political unification, and in relation to these processes, there is a growing need to harmonize accounting (Dvořáková, 2011). Also in the Czech Republic, the harmonization of accounting has been taking place. One of the recent steps in adjustment of Czech legislation is the amendment to the implementing Decree to Act on Accounting for Entrepreneurs, effective since January 1, 2016. There is a clear tendency in the amendment to harmonize Czech legislation with European law, which should help address the accounting differences between the Czech Accounting Standards (CAS) and International Accounting Standards (IAS/IFRS). The subject of this paper is to describe the current situation of Czech companies which report financial statements under the IAS/IFRS and, at the same time, they have to keep the accounts under the CAS for various reasons, primarily for tax reasons. The objective of the survey is to evaluate how these companies proceed and how they perceive different requirements of national and European accounting standards. The need for harmonization of accounting Currently, there are several coexisting systems of financial reporting accepted in global financial markets. The globalization of capital markets also requires a global harmonization of accounting systems. The use of national accounting systems for preparation of financial statements makes gathering information needed for the purposes of comparison very difficult and costly for investors in capital markets (Bohušova, 2008). The harmonization of accounting means the unification of accounting, valuation and reporting of the same transactions. It is a process that results in a gradual elimination of disparities arising from national processing of accounting transactions that may differ from each other. The main reason is to ensure the comparability of the reported 242 information for needs of their users, since these statements are often the only source of information about the company, its performance and changes in financial position. The global harmonization of accounting takes place based on convergence of two main concepts - the European IAS/IFRS and the American - Generally Accepted Accounting Principles in the United States - US GAAP. These issues have been addressed by Peng and Bewley (2010), Liu (2011), Boyle et al (2006). On the European continent, there is a process of harmonization within the European Union (EU). The European Union includes countries with different social, economic and political environment, and has 25 different tax regimes. This is the reason why the financial reporting systems in the EU vary (Fox, 2013). Here, the harmonization has been perceived as part of the single European business environment in order to facilitate comparability of businesses operating in territories where there are unified accounting standards, to alleviate the situation of companies that are expanding their activities outside their home states and to enable an unified understanding and accounting processing of transactions and financial situation of businesses (Šrámková and Janoušková, 2008). The first efforts to harmonize accounting systems on international level arose from the needs of multinational companies and their increasing pressure. These companies needed to compare their different accounting information (Žárová, 2006). Today, however, there is an increasing need to harmonize accounting in general, not just in large multinational companies. This is so because the accounting regulations should be the same for all accounting entities, as stated by Jílek: It is necessary to set up such accounting system that would not be too painful for companies. Accounting must be clear, the same for all accounting entities, transparent, inexpensive and having clear rules of the game. Then, even an average educated person understands it, and costs of accounting and auditing will not be sky-high. At the same time, it supports the competitiveness of companies worldwide (Jílek and Svobodová, 2013). When creating the Czech accounting standards, it is not necessary to invent anything new, but simply adopt what is common in the world. The best way is to fully apply the International Financial Reporting Standards with all of their interpretations and implementation manuals, without any exception. Then, the development of unified Czech accounting standards would not be difficult at all, as it is claimed sometimes. It is difficult now, for the reason that the financial result represents the starting point for calculating income taxes (lobbying interests) (Jílek and Svobodová, 2013). Therefore, Jílek does not see a problem in the complexities of transformation of the CAS to the IAS/IFRS, but rather in the tax aspect. It must be noted that in addition to the accounting harmonization on a European scale, there are also differences between the accounting and tax approach in the national level. This difference should be gradually eliminated as well. In some countries, the accounting and tax reports are identical, in others are not. For example in Germany, financial statements fully comply with tax requirements. On the contrary, in the United States, Great Britain or in our country, the situation is different. Financial statements differ from tax statements that are submitted to tax offices. It means that tax deductible expenses and income are different from accounting expenses and income (Jílek and Svobodová, 2013). The IAS/IFRS and the US GAAP are similar yet different in some areas. For more see Peng and Bewley (2010), Liu (2011), Boyle et al (2006). Also, the Czech accounting standards have not fully merged into the IAS/IFRS yet. The process of harmonization of the CAS with the IAS/IFRS still must be done, especially in 243 the area of reporting of fixed assets, accounting of reserves and contingent assets, in the concept of expenses and income and in the issue of the deferred tax. The complete harmonization still has not taken place at the national level of accounting and tax regulations, as already stated. Legislative change In the 102 Series issue of the Collection of Laws of October 2, 2015, Decree No. 250/2015 Coll. was declared, effective as of January 1, 2016, amending the former Decree No. 500/2002 Coll. This Decree implemented certain provisions of Act No. 563/1991 Coll., On Accounting, as amended, for entrepreneurs as accounting entities who keep double-entry accounting, as amended. The reason for the amendment of the Decree is primarily to complete the transposition of the Directive 2013/34/EU of the European Parliament and of the Council of June 26, 2013 on the annual financial statements, consolidated financial statements and related reports of certain types of undertakings, amending Directive 2006/43/EC of the European Parliament and of the Council and repealing Council Directives 78/660/EEC (of July 25, 1978) and 83/349/EEC (of July 25, 1983), in relation to Act No. 221/2015 Coll. (of August 12, 2015) amending Act No. 563/1991 Coll., On Accounting, as amended. In the amendment, there is a clear tendency to harmonize Czech legislation with European law, specifically in terms of the International Financial Reporting Standards (IFRS) and the International Accounting Standards (IAS). The amendment includes new financial statements (balance sheet and profit and loss account) both full and summarized including new rearrangement and labeling of entries and new or updated content of entries as well as new attachments in financial statements by categories of accounting entities also introduced by the new Decree. In the profit and loss account, there is an abolition of separate reporting of extraordinary income and expenses, which are not reported in the income statement by the IAS 1. The amendment abolishes reporting of formation expenses under the fixed assets. Therefore, formation expenses are no longer considered a real asset of the accounting entity, as with the IAS/IFRS. The reason is the absence of important criterion for recognizing assets, as this asset can be valued by the benefits provided to the accounting entity. The amendment changes reporting and accounting of a change in inventory produced internally and activation of inventory and fixed assets produced internally. The new legislation in accordance with IAS 2 accounts the change in inventory as well as its activation as the elimination of costs incurred. Interim financial results will no longer be affected by an unrealized production. The amendment also defines to set the own costs in inventory produced internally. The amended provision limits the proportion of the variable and fixed indirect costs, and expressly excludes the sales-related costs, also in accordance with IAS 2. The above indicates that there is an ongoing process of harmonization of Czech accounting standards, their moving towards the IAS/IFRS. 2 Methodology and Data Between September and December 2015, 24 respondents, who address the issue of the IAS/IFRS in the Czech Republic, were interviewed by means of semi-structured interview. Their function in the IAS/IFRS, the sector of their activity and information on the placement of their company shares on the stock exchange are included in the Table 1. Interviews' records including contacts are available. However, in order to ensure the anonymity, these persons were coded for the purpose of publication. 244 Table 1 Information about the Persons Interviewed Code Function_CZ-NACE sector classification_Stock Exchange Al Auditor agriculture, forestry, fisheries in preparation A2 Auditor professional, scientific activities and technical no A3 Auditor professional, scientific activities and technical no A4 Auditor professional, scientific activities and technical no A5 Auditor manufacturing industry no A6 Auditor professional, scientific activities and technical no PI project manager finance and insurance PSE P2 project manager manufacturing industry PSE P3 project manager finance and insurance no Ul accountant manufacturing industry no U2 accountant manufacturing industry no U3 accountant manufacturing industry no U4 accountant manufacturing industry Frankfurt Exchange Stock U5 accountant transport and storage Luxembourg Exchange Stock U6 accountant water supply no U7 accountant finance and insurance no U8 accountant manufacturing industry no U9 accountant production and distribution of electricity, no gas, heat U10 accountant manufacturing industry no Uli accountant finance and insurance no U12 accountant finance and insurance no U13 accountant manufacturing industry no U14 accountant wholesale and retail Luxembourg Exchange Stock U15 accountant production and distribution of electricity, no gas, heat Source: Own elaboration The respondents were asked about the way they deal with different CAS and IAS/IFRS requirements in practice. 3 Results and Discussion The Czech companies which prepare financial statements in accordance with the IAS/IFRS, and also must meet different accounting and tax regulations, can decide for one fundamental method of accounting, while the second system supplements it, or they can keep the accounts simultaneously in both systems. Persons who have experience with the accounting primary according to the CAS stated as follows: Al: "The company keeps the accounts according to our national accounting system. When creating documents, so-called reference keys are added, thanks to which the IFRS reports are generated in specific transactions." 245 A3: "We keep the accounts under the CAS, and we fill out papers for the IAS/IFRS in a certain application of our parent company." A6: "In my experience, most companies in the Czech Republic still keep the accounts based on Czech accounting system primarily for tax reasons. Components and systems for creation of the IFRS statements have been extra added into this system." U3: "Although we are obliged to report under the IAS/IFRS, the basis of everything is to keep the accounts according to Czech accounting standards. This basis has been adjusted with entries required by the IAS/IFRS reporting. According to Czech accounting standards, we have to report not only for purposes of reporting income tax, but also for statistical purposes, for reports to banking entities and, last but not least, for internal purposes, since different premium and other indicators are set according to Czech accounting." Ull: "The company continues to keep the accounts under Czech accounting standards considering this method more effective. We would have to recalculate the result from the IFRS to be applicable to the calculation of tax." U13: "We deal with the accounting issues using conversion bridges to convert the CAS to the IAS/IFRS." U15: "We keep the accounts under the CAS. Adjustments for technical preparations of documents for IAS/IFRS reporting have been done in Excel." Persons whose accounting system is based on the IAS/IFRS stated: U2: "The major standards are the IFRS, or their corporate interpretation. Once a year, for the purpose of preparing the local financial statements / tax returns, the differences from the CAS have been calculated." U12: "In our company, there is no concurrence of both systems. We keep the accounts under the IAS/IFRS and make adjustments outside the accounts for tax purposes." P2: "We keep the accounts under the IAS/IFRS. For the purposes of taxation or statistics, we report the required information outside the accounting system." Most of the respondents, however, decided for the concurrence of both accounting systems. Following persons commented on parallel accounting systems: A2, A4, PI, P3, Ul, U4, U5, U6, U7, U8, U10, and U14. Their comments indicate that some concurrence is necessary due to the need to determine the tax base according to Czech accounting standards. They keep parallel accounting books according to Czech accounting standards. Software begins to emerge on the market that can handle both accounting software at once: A5: "At the request of the main investor of the holding, a client has introduced new accounting software. This software allows keeping the accounts under the CAS as well as under the IAS/IFRS." U9: "The official reporting is under the IAS/IFRS. Within the SAP information system, there is a secured option to display statements under the CAS or the IAS/IFRS." Table 2 represents a complex structure of used accounting systems. Table 2 Frequencies of Used Accounting Systems Accounting system Number of persons Primarily CAS, supplemented with IAS/IFRS 7 Primarily IAS/IFRS, supplemented with CAS 3 Parallel concurrence of both systems 14 Source: Own elaboration The results of the survey have showed that there is no uniformity in how businesses cope with the dual legislation. Most companies preparing financial statements under the 246 IAS/IFRS keep their accounts in two parallel accounting systems. This condition is very counterproductive as it denies the main objective of accounting. The objective is to truthfully and honestly inform about the economic activities of the accounting entity (Act No. 563/1991Coll., On Accounting). In the current state, however, there are two truthful and honest approaches. Therefore, clarity has been reduced and demands on the accounting entity are increasing. U12 user said: "Since there is the need to maintain these adjustments and dual methodology as well as to assess each transaction whether it is important for the tax return, the whole process is more complicated." Businesses often reach for the method of the lesser evil to be able to satisfy both legislative directions. They seek to find liaison bodies which primarily comply with both systems, as U4 user commented: "In all areas possible, we have adopted the IFRS perspective." Even a company, which preferred to reconsider its long-term payables to short-term ones in order to avoid discounting and to report payables pursuant to the CAS, is probably no exception. The result is some kind of travesty, a third system, which will last until the CAS and the IAS/IFRS are fully harmonized. In the context of harmonization, it is also necessary to unify accounting systems with tax systems, see Figure 1, chap. 1.1. U14 respondent said in this context "We would appreciate if the IFRS were legally recognized as a basis for calculating the tax base." Therefore, full harmonization of accounting system means not only the harmonization on the territory level (national & European or worldwide convergence), but also the harmonization of accounting and tax regulations. This harmonization has already taken place in some European countries, e.g. in Germany. In the Czech Republic, however, some differences in accounting and tax base still persist. See more in Jílek and Svobodová (2016). 4 Conclusions Employees who prepare financial statements under the IAS/IFRS welcome the change in the amendment to Decree, because it is a step towards the CAS and the IAS/IFRS harmonization. It is undisputed that the process of harmonization of accounting systems continues. However, the final unification of our national system with the European one has not happened so far. Accounting of Czech businesses preparing financial statements under the IAS/IFRS is currently somewhat counterproductive. Businesses have chosen different methods to cope with the dual legislation. However, it is evident that it involves great demands on time and personnel and technical support and the associated costs. References Bohušova, H. (2008). Harmonizace účetnictví a aplikace IAS/IFRS: vybrané IAS/IFRS v podmínkách českých podniků, lsted. Praha: ASPI. Boyle, G., Clyne, S., Roberts, H. (2006). Valuing Employee Stock Options: Implications for the Implementation of NZ IFRS 2+. Pacific Accounting Review, vol. 18(1), pp. 3-20. Dvořáková, D. (2011). Finanční účetnictví a výkaznictví podle mezinárodních standardů IFRS: aktualizované a rozšířené vydání. 3rd ed. Brno: Computer Press. 247 Fox, A., Hannah, G.. Helliar, C, Veneziani, M. (2013). The costs and benefits of IFRS implementation in the UK and Italy. Journal of Applied Accounting Research, vol. 14(1), pp. 86-101. Hronová, S. (2009). Národní účetnictví: nástroj popisu globální ekonomiky. 1st ed. Prahe: C.K. Beck. Jílek, J., Svobodová, J. (2013). Účetnictví podle mezinárodních standardů účetního výkaznictví2013. 3th ed. Praha: Grada Publishing. Kovanicová, D. (2003). Finanční účetnictví: světový koncept. 4th ed. Praha: Polygon, 2003. Liu, C. (2011). IFRS and US-GAAP comparability before release No. 33-8879. International Journal of Accounting & Information Management, vol. 19(1), pp. 24-33. Peng, S., Bewley, K. (2010). Adaptability to fair value accounting in an emerging economy. Accounting, Auditing & Accountability Journal, vol. 23(8), pp. 982-1011. Directive 2013/34/EU of the European Parliament and of the Council on the annual financial statements, consolidated financial statements and related reports of certain types of undertakings. The European Parliament, 2013. Directive 2006/43/EC of the European Parliament and of the Council, The European Parliament, 2006. Council Directive No. 78/660/EEC, The European Parliament, 1978. Council Directive No. 83/349/EEC, The European Parliament, 1983. Šrámková, A., Janoušková, M. (2008). IAS/IFRS: mezinárodní standardy účetního výkaznictví: praktické aplikace. Praha: Institut Svazu účetních. Decree No. 250/2015 Coll. The Ministry of Finance, 2015. Decree No. 500/2002 Coll. The Ministry of Finance, 2002. Act No. 221/2015 Coll. The Ministry of Finance, 2015. Act No. 563/1991 Coll., On Accounting. The Ministry of Finance, 1991. Žárová, M. (2006). Regulace evropského účetnictví: aktualizované a rozšířené vydání. lsted. Praha: Oeconomica. 248 High-Frequency Trading and Price Volatility in the Paris Euronext Stock Market Juraj Hruška, Oleg Deev Masaryk University Faculty of Economics and Administration, Department of Finance Lipová 41a, 60200 Brno, Czech Republic E-mail: oleg@mail.muni.cz, 206887@mail.muni.cz Abstract: Algorithmic trading has become the crucial part of security trading on world equity markets influencing many of its characteristics. In this paper, we consider the effects of high frequency trading on the short term volatility. The aim of the paper is to analyze the relationship between high frequency trading (HFT) and spot volatility in high frequency as well as low frequency data from the French stock market. We employ GMM, GARCH and Markov switching models to estimate the relationship between changes in stock returns and changes in the activities of high frequency traders. We propose our own methodology to proxy changes in the activity of algorithmic traders. We also address the problem of optimal sampling to avoid possible biases in our empirical findings, since high frequency data contain a disruptive volatility component (market microstructure noise), by incorporating Bundi-Russell (2008) test and test of Lagrangian multipliers. Most actively traded stocks listed on the Paris stock exchange are chosen for the empirical analysis. Sampling tests suggest that optimal frequency should be approximately 60 minutes. Results from models confirm the hypotheses of positive impact of high-frequency trading on market volatility. Keywords: volatility, high frequency trading, general method of moments, Markow switching model, GARCH JEL codes: C24, C55, C58, G12 1 Introduction Trading on the world exchanges have been dominated by computers since the 1980s. Securities exchanges are now fully electronic and floor trading is inevitably coming close to its extinction. Rapid algorithmic trading certainly changed the nature of financial markets with no common consensus whether this change improves market efficiency and liquidity or increases volatility and deteriorates prices in turbulent situations (Kendall 2007). Both academics and practitioners cannot agree on the effects of high-frequency trading. On one hand, high frequency traders (HFT) certainly proclaim that their activities are increasing market liquidity, lowering spreads and reducing transaction costs. On the other hand, there are so called "low frequency traders" (mostly long term institutional investors) who are criticizing HFT traders for scalping their orders, manipulating the market and boosting volatility. Academics are usually inclining to praise benefits of high frequency trading, but there are also some researches confirming the HFT fault in volatility outbreaks. Few minor financial market crashes, such as the 2010 Flash Clash or the case of Knight Capital, are believed to be at least partially caused by algorithmic trading. After such events regulators are calling for responsible behavior of HFT traders and searching for ways and methods to control the HFT market. For example, circuit breaks are now implemented on the markets to shut down trading in case of increased volatility to prevent flash crashes. In this paper, we focus on the effects of high-frequency trading on the volatility of securities' prices, which is usually considered to be increasing with growing activity of HFT traders. The aim of this paper is to test the link between the high-frequency trading and price volatility both in high frequency and low frequency data. Several issues should be addressed before analyzing this relationship with market microstructure noise being the most important and complex one. Market microstructure noise - high-frequency order book information that reflects fluctuations in supply/demand of the analyzed security - 249 complicates the estimation of volatility while making standard estimators unreliable. The effect of the market microstructure noise is certainly negligible in the long run, but should be dealt with in the short run. Connection between the trading activity and volatility was well known even before algorithmic trading took a lion share of overall trading volume. Positive relationship between volatility and volume of trading has been proven by Karpoff (1987). Kyle (1985) documented positive relation between volatility and number of trades and order imbalance before algorithmic trading had been introduced. Easley et al. (1997) confirm positive relationship between trade size and price volatility using competitive models. Jones et al. (1994) show that before high-frequency trading the stock volatility had been affected mostly by the number of trades, while average trade size played no role. The introduction of algorithmic trading changed the impact of the size of average trade on market volatility, which became significant as shown by Chlistalla et al. (2011). Even if significant relationship between volatility and average size of trade indicates and subjectively confirms the influence of high-frequency trading, newer studies suggest that order imbalance, and not a number of trades, initiates an impact of trading volume on volatility. Comparing small and large trades by their effect on volatility, Huang et al. (2003) discover that small trades close to the maximum-guaranteed quoted depth tend to affect the price changes more than big trades. While using realized volatility instead of volatility measured by absolute returns as a measure of market instability, Chan et al. (2000) show that only a number of trades matters and not volume of trading and trade size. Some studies also suggest that high-frequency traders in general and especially market makers tend to reduce market volatility (Kirilenko et al. 2015). This paper differs from existing studies in three important respects. First, since the majority of studies on the relationship between HFT trading and price volatility is conducted on the US data, we focus on one of the biggest European stock markets - the French stock market during time periods that have not been examined in the literature. Second, we propose the methodology to uncover the high-frequency trading from the volume of trading, trading activity and average trade size. Third, we address the problem of appropriate test procedure selection to detect market microstructure noise and find the optimal test sample with several data frequencies considered. 2 Methodology and Data Market volatility as a measure of investment risk is calculated in several ways. Most common measure is the standard deviation of market returns, which is largely dependent on returns of previous observations given the sampled data. The most frequent data available for our analysis is minute data. Hence, the best choice for estimating the current market volatility is the logarithm of the ratio of the highest and lowest prices during the given minute (Aldridge 2013). The best way to measure high-frequency trading activity is by monitoring number of messages send by HFT traders and compare them to overall message traffic. However, such data are not widely available, when some exchanges even do not keep the records about such events or do not distinguish between various types of order submission. Hence, it is necessary to create proxy variables to estimate HFT activity based on its specific characteristics, such as number of small orders and increased number of orders Hendershott (2011). We measure difference in HFT activity as logarithm of reverse relative change of average trade size (in number of shares) multiplied by relative change in number of trades. (1) 250 hftiit = In /W-+(g)J^+1>o^)\ \K+(£)J(^ + 1>(^1+H (2) where is voli/t volume of trading of share /' in time t. It is identified as the sum of volume of market orders (vmijt), volume of limit sell orders (va//t) and volume of limit buy orders (vbi/t). Number of orders of share /' in time t is denoted as niit. It is again given by sum of number of trades (nmijt), number of limit sell orders (A73ijt) and number of limit buy orders (nblft). One extra trade (calculated as the mean of average sizes of trades in last h observations) is added to the ratio of the change in average size of trade (or order). This will assure that function will be defined even in cases of complete market inactivity. Average number of trades (again calculated from last h observations) is added to second ratio. Without this change, relative change in number of trades would be higher for lower absolute changes. If change of aggressive HFT activity needs to be calculated only volume of market orders (vmijt) and number of trades (nmijt) are used. On the other hand, when changes in defensive HFT activity are needed, it would be calculated only from volume of limit orders {vaiit and vbiit) and number of limit orders {naiit and nbiit). After addressing possible problems in the modeling of high-frequency data (excluding market microstructure noise) and choosing the most appropriate way to measure high-frequency volatility and high-frequency trading activity, we can now move to formulate the model and select an estimation procedure to analyze the impact of high-frequency trading on market volatility. We test the relationship in two data frequency designs based on the results of optimal sampling procedures. First, we use higher sampling frequency data, where market microstructure noise is present. Next we run the same models on the data with lower sampling frequency or optimal frequency suggested by LM and Budni-Russel (2008) tests. The basic model to test the relationship of changes in HFT activity and market volatility has the following form: 0{i,t} = ai + P{i,i}HFTiit + (3{ii2}RVi,t + PaMht + P{i,4}vi,t + Vi,t (3) where RVm/t is estimation of realized market volatility calculated from previous 30 one-minute returns of CAC 40 index, AFiit is a dummy variable that indicates observations where no trades occurred and Viit is a volume of trades of stock /' during observation t. Control variables were inspired by ones used in Giot et al. (2010). Error in data with high frequency consists of vijt = uu + eu, where uuis an error term and eijt represents market microstructure noise. For low frequency data, market microstructure noise is considered to be insignificant. We use three different estimation procedures for the analysis. The first, linear estimation employs the generalized method of moments (GMM) method with Newey-West (1994) Bartlett HAC estimator to treat autocorrelation and heteroscedasticity. The second estimation procedure represents Markow switching model techniques with three levels. More regimes brought any improvement to our results. This method helps us gain better explanatory power of the model by estimating coefficients for three different levels of explained and explanatory variables. This levels are random and switching between them is a random process, hence, problems might occur if the coefficients in models will not be consistent. Reduced-form model is used in the second estimation procedure: 0{i,t} = ai + P{i,i}HFTiit + (3{ii2}RVi,t + vut (4) For third estimation method, we use GARCH(1,1) model with intraday adjustments for seasonality. If none of the external regressors are non-significant, we switch to the exponential GARCH(1,1) model. In this case we use the same model as in the first case. 251 We have used two different sampling of data. The same models were applied on both samples. This was done to test different effects of algorithmic trading on spot volatility under the influence of market microstructure noise and without its presence. For the version with influence of MMN we have chosen one minute data. Most traded stocks on the Euronext Paris Stock Exchange are picked based on the following criteria: minimal volume of trading, minimal market capitalization, and minimal number of active observations. Only primary issues are selected. After excluding stocks with incomplete data for the chosen period, 65 stocks fulfill the imposed criteria. This might not be the best number for generalization of our results, but as we are working with proxy variables, the stocks should fulfill our strict criteria, or otherwise, our analyses would not give valid results (many other stocks are less frequently traded). The analyzed period starts at April 15, 2015 and ends at October 19, 2015. Daily observations start at 9:06 a.m. and end at 5:23 p.m. to exclude situations usually containing negative spreads and increased volatility caused by opening and closing auctions. This also solves the problem of intraday returns, which can lead to biased estimates of realized volatility. Days with shortened trading time were excluded. The average summary statistics for all stocks is provided in Table 1. We intend to use two type of data sets. First dataset has one-minute frequency, while the observation frequency for the second dataset is established by optimal sampling tests. These tests were test of Lagrangian Multipliers for presence of MMN (Shin 2015) and Bandi-Russel test (Bandi et al. 2008). The second dataset will not thus contain the market microstructure noise. All data were gathered from Bloomberg. Table 1 Average Summary Statistics for the Studied Shares with One Minute Frequency (from April 15, 2015 to October 19, 2015) Variable Mean Maximum Minimum Stand. Dev Price (PLt) 54.37 58.99 47.5 42584.00 Return (n.t) -0.00000217 0.02068 -0.045676 0.000588 Number of trades (nmi>t) 12.76 247.84 0.51 9.00 Number of sell orders (nai.t) 239.04 1912.65 30103,00 152.57 Number of buy orders (nbi.t) 239.04 1912.65 30103,00 152.57 Volume of trades (vmi>t) 3377.21 63160.82 67.18 2686.74 Volume of sell orders (va^) 4735.12 810611.9 67.18 4773.03 Volume of buy orders (vbi>t) 483591.95 5401812.35 2441.2 387366.29 Spread (Si.t) 0.035095 0.131453 0.017 0.007148 Order imbalance (oii.t) 0.225411 0.391466 0.086998 0.039715 Volatility (ai>t) 0.000543 0.007404 0.00007 0.000315 Difference of HFT activity (hfti.t) 0.001225 1.797952 -1.43853 0.210638 Source: Own elaboration The analyzed period is rather stable with slow decline of nearly all larger stocks traded on European markets. The mean return is negative with standard deviation being close to mean volatility. This confirms that this two estimations of market volatility gives very similar results. The average number of trades per minute is approximately 13. Approximately 240 limit orders on both side of limit order book, which indicates sufficient activity of market makers for our analysis. The average correlations showed the relation of volatility of the stock returns with the control variables (turnover, volume and volatility of the market). The correlation with HFT activities (0.42) is weaker but still significant. 3 Results and Discussion First, we provide the results for linear estimations based on full data. In cases, where we use overall changes in activity in submitting market and limit orders, we find the positive relationship between high-frequency trading and market volatility in all cases, as seen in Table 2. Aggressive trading that uses only market orders and reduces liquidity seems to 252 have smaller effect on market volatility, but nevertheless, is still positive. Defensive market making also seems to produce increased pressure on variability of prices. Majority of cases confirms the hypothesis that increased HFT activity induces volatility. Table 2 Results of GMM Estimations for All Types of HFT Activity HFT Mean Standard Number of positive activity coefficient deviation Max Min coefficients Overall 0.000251 0.000103 0.000498 0.000110 64 Aggressive 0.000068 0.000025 0.000124 0.000023 64 Defensive 0.000131 0.000067 0.000280 0.000033 64 Source: Own elaboration J-tests confirm validity of the results in all cases, and HFT activity coefficients are significant for every stock. These coefficients have been always positive, except for one stock where the effects of HFT were not significant in all 3 cases. Second method used for estimation of tested relation was Markov switching model with 3 regimes (Table 3). We have also conducted estimations with more levels, but they added no new results. Coefficients of determination were slightly higher, but the coefficient for HFT factor was the same in some regimes. In all three regimes and for all stocks, the relationship between HFT activity and market volatility is significantly positive. Table 3 Results of Markov Switching Model Estimations for Aggressive HFT Activity Mean coefficient Standard deviation Max Min Average R/v2 Regime 1 0.000334 0.000235 0.000957 0.000053 0.227678 Regime 2 0.000191 0.000149 0.000649 0.000060 0.208485 Regime 3 0.000221 0.000209 0.000620 0.000001 0.208401 Source: Own elaboration GARCH(1,1) models confirmed the hypothesis that volatility of residuals in our models is changing over time, but rejected any effects of HFT activity on spot volatility. Seasonal adjustments of intraday volatility have been proven to have no effect in improving efficiency of the modeled explanation power. Table 4 Results of GMM Estimations for All Types of HFT Activity for Low Frequency Data HFT activity Mean coefficient Standard Number of positive deviation Max Min coefficients Overall 0.004963 0.001439 0.010109 0.002389 65 Aggressive 0.005240 0.002115 0.013895 0.001265 65 Defensive 0.004901 0.001405 0.009697 0.002264 65 Source: Own elaboration Next we applied the same methodology on data with one hour frequency. We chose this sampling period after the analysis of tests for presence of MMN results. Influence of HFT on volatility was positive and significant in all cases. In longer run without the presence of MMN aggressive HFT trading seems to have rightfully stronger influence on variability of prices (Table 4). This might be due to distribution of market orders in time. These results confirms hypothesis that HFT increases volatility in long run and might decrease volatility in short run. But this is more occasional effect then a rule. Strong influence of aggressive algorithmic trading has been also confirmed by Markow switching model (Table 5). The coefficients of determination suggest that this relationship is indeed really relevant. The results for overall and defensive HFT activity were in 253 general similar to aggressive. The only difference was again that for aggressive was impact of HFT stronger. Table 5 Results of Markov Switching Model Estimations for Low-Frequency Data Mean Standard coefficient deviation Max Min Average R*2 Regime 1 0.006448 0.007156 0.032528 0.000669 0.689287 Regime 2 0.005301 0.006478 0.038625 0.000539 0.704681 Regime 3 0.005957 0.005732 0.027851 0.000561 0.651948 Source: Own elaboration For the samples with one hour frequency GARCH(1,1) model estimations accepted effects of HFT on spot volatility in contrary to previous sampling. This can be explained that the technical realization of trading is behind much of the price disturbance in short run, but is smoothed in long run, where other effects such as HFT activity emerge. Table 6 suggests that in all cases researched link was positive. Conclusion that relationship between HFT and volatility is more intensive for aggressive trading activity was apparent again. Table 6 Results of GARCH(1,1) Estimations for All Types of HFT Activity HFT Mean Standard Number of positive activity coefficient deviation Max Min coefficients Overall 0.004963 0.001439 0.010109 0.002389 65 Aggressive 0.005240 0.002115 0.013895 0.001265 65 Defensive 0.004901 0.001405 0.009697 0.002264 65 Source: Own elaboration Intraday seasonality played no role in this case either. These effects might have been more apparent if we did include first and last minutes of trading that were excluded due to presence of opening and closing auctions. 4 Conclusions Frequent news about manipulative techniques and aggressive trading strategies used by some algorithmic traders may indicate that the high-frequency trading is disruptive for other market participants. In this paper, we analyze the relationship between high-frequency trading activities and market volatility on the French stock market. We provide evidence based on three different estimation procedures that high-frequency trading in fact increases the short-term variability of prices. These effects have also been proven in long term without the presence of market microstructure noise. In our specifications of high frequency trading, the aggressive trading has stronger impact on volatility than the defensive trading, which is with accordance with preliminary hypothesis. Our study supports previous findings and enlarges the geographic evidence of the destabilizing effects of high-frequency trading on market volatility. Further research will show whether these effects are similar on other European markets. Acknowledgments The support of the Masaryk University internal grant MUNI/A/1025/2015 is gratefully acknowledged. References Aldridge I. (2013). High-frequency trading: a practical guide to algorithmic strategies and trading systems. Hoboken, New Jersey: John Wiley & Sons, Inc. Bandi, F. M., Russell, J. R. (2008). Microstructure noise, realized variance, and optimal sampling. The Review of Economic Studies, vol.75(2), pp. 339-369. 254 Chan, K., Fong, W. M. (2000). Trade size, order imbalance, and the volatility-volume relation. Journal of Financial Economics, vol. 57(2), pp. 247-273. Chlistalla, M., Speyer, B., Kaiser, S., Mayer, T. (2011). High-frequency trading. Deutsche Bank Research, vol. 7. Easley, D., Kiefer, N. M., O'Hara, M. (1997). One day in the life of a very common stock. Review of Financial Studies, vol. 10(3), pp. 805-835. Giot, P., Laurent, S., Petitjean, M. (2010). Trading activity, realized volatility and jumps. Journal of Empirical Finance, vol. 17(1), pp. 168-175. Huang, R. D., Masulis, R. W. (2003). Trading activity and stock price volatility: evidence from the London Stock Exchange. Journal of Empirical Finance, vol. 10(3), pp. 249-269. Hendershott, T., Jones Ch. M., Menkveld A. J. 2011. Does Algorithmic Trading Improve Liquidity?. The Journal of Finance, 66(1), pp. 1-33. Jones, C. M., Kaul, G., Lipson, M. L. (1994). Information, trading, and volatility. Journal of Financial Economics, vol. 36(1), pp. 127-154. Karpoff, J. M. (1987). The relation between price changes and trading volume: A survey. Journal of Financial and Quantitative Analysis, vol. 22(01), pp. 109-126. Kendall, K. (2007). Electronic and Algorithmic Trading Technology. 1st ed. London: Academic Press. Kirilenko, A. A., Kyle, A. S., Samadi, M., &Tuzun, T. (2015). The flash crash: The impact of high frequency trading on an electronic market. Available at SSRN 1686004. Kyle, A. S. (1985). Continuous auctions and insider trading. Econometrica: Journal of the Econometric Society, pp. 1315-1335. Newey, W. K., & West, K. D. (1994). Automatic lag selection in covariance matrix estimation. The Review of Economic Studies, vol. 61(4), pp. 631-653. Shin, D. W., & Hwang, E. (2015). A Lagrangian multiplier test for market microstructure noise with applications to sampling interval determination for realized volatilities. Economics Letters, vol. 129, pp. 95-99. 255 The Prediction of Economic Activity Growth by Sovereign Bond Spread in France, Germany and Great Britain Jana Hvozdenska Masaryk University Faculty of Economics and Administration, Department of Finance Lipova 41a, 602 00 Brno, Czech Republic E-mail: 174974@mail.muni.cz Abstract: The steepness of the bond yield curve should be an excellent indicator of a possible future economic activity. A rise in the short rate tends to flatten the yield curve as well as to slow down real growth the near term. The relationship between the spread and future GDP activity was proved already before. One question remains - which spread is the best for the future prediction? Is it the spread between sovereign 10-year bonds and 3-mo nth bonds or 30-year and 1- year or 10-year and 1-year sovereign bonds? This paper aims to analyze which spread is the most suitable for predicting of future economic growth in France, Germany and Great Britain between the years 2000 and 2016. The natural and probably the most popular measure of economic growth is GDP growth, taken quarterly. We have found out that the best predictive spreads in France, Germany and Great Britain are the spreads of 30-year and 1-year and 10-year and 1-year government bond yields. These findings might be beneficial for investors and provide further evidence of the potential usefulness of the yield curve spreads as indicators of the future economic activity. Keywords: bonds, slope, spread, yield curve JEL codes: E43, E44, E47, G01 1 Introduction The financial turmoil during 2007-2009 affected the euro area financial sector in ways that differ considerably across market segments and countries. A consequence was a temporary reduction of market activity within national borders. The impact was felt most strongly in the money markets, and relatively less in bond activities. However, economic growth stopped and still many countries are not able to follow Maastricht Convergence Criteria. On one hand, the integrated financial markets and the common currency may help protect the countries from the negative impacts of a financial crisis, because the countries are part of a large, stable economic unit. On the other hand - financial instability may spread easily from country to country, since barriers to the capital movements have been reduced. Many market observes carefully track the yield curve's shape, which is typically upward sloping and convex. However when the yield curve becomes flat or slopes downward (the spread between sovereign 10-year and 3-month bond is negative) it may signal GDP decrease (recession). The spread of 10-year and 3-month government bond is widely used and it is the most common measurement of the yield spread. The yield curve simply plots the yield of the bond against its time to maturity. The yield curve - specifically the spread between long term and short term interest rates is a valuable forecasting tool. It is simple to use and significantly outperform other financial and macroeconomic indicators in predicting recessions four quarters ahead. This paper builds on a wide range of previous researches, but differs in some ways. Bernard and Gerlach (1998) in their paper showed empirically on eight countries that the slope of the yield curve is a good predictor of the real economic activity. Berk and van Bergeijk (2001) examined 12 euro-area countries over the period of 1970-1998 and found that the term spread contains only limited information about future output growth. Their work is based on the previous theoretical researches of Estrella and Hardouvelis 256 (1991), Estrella and Mishkin (1996). There was proven the evidence that the slope of the yield curve and the future GDP activity are related together. However it is necessary to say that this rule was true until the end of 20th century and it mostly disappeared at the beginning of 21st century and appeared again during the financial crisis (from 2008) and later on (De Pace, 2011; Giacomini and Rossi, 2006; Chinn and Kucko, 2010). Most of the studies are focused on the relationship of the yield curve and GDP activity of the United States of America. All the authors used as a spread, which was analysed in their works, the spread of 10-year and 3-month government bonds. This relationship was proved to be the best in the past (Estrella and Hardouvelis, 1991, Estrella and Mishkin, 1996). This paper aims to analyze which spread is the most suitable for predicting of future economic growth in France, Germany and United Kingdom between the years 2000 and the first quarter of 2016 and if this relationship has changed after the financial crisis. The possible spreads are as follows: 30-year and 1-year, 15-year and lyear, 10-year and 1-year, 5-year and 1-year, 30-year and 3-month, 15-year and 3-month, lOyears and 3-month and finally 5-year and 3-month. Of course there are other possibilities, but it is very hard to get a different data for a chosen time period. 2 Methodology and Data There are many ways of using the yield curve to predict the future real activity. One common method uses inversions (when short term rates are higher than long term rates) as recession indicators. Obtaining predictions from the yield curve requires a lot of preliminary work. There is the principle which needs to be held: keep the process as simple as possible. A yield curve may be flat, up-sloping, down-sloping or humped. The standard solution uses a spread (difference between two rates). The problem is to choose the spread between the right terms. The most used spread is between 10-year and 3-month bonds. The problem is that there are rarely bonds which mature exactly in 10 years (or 3 months). In that case the best solution is to use the yield curve, which shows the yield of each maturity. Creating and calculating of the yield curve is a rather difficult task because there are many ways how to do it and every country uses a different model of construction. The yield curves are constructed by Bloomberg, therefore the data for spreads were gained from Bloomberg. For the spreads 30-year and 1-year, 15-year and 1-year, 10-year and 1-year, 5-year and 1-year, 30-year and 3-month, 15-year and 3-month, 10-year and 3-month and finally 5-year and 3-month government bond rates were chosen. Quarterly data were used for the spreads because the data of the economic activity growth are taken on quarterly basis as well. The data of real GDP growth can be found at Eurostat, OECD statistics or Bloomberg. The data of real GDP obtained and used in this paper are from OECD statistics. The selected countries are France, Germany and United Kingdom. There is no recent previous research in European Union which would prove or reject the hypothesis that the spread between 10-year and 3-month government bonds is the best for predicting of the future economic growth. Model Specification As a measure of real growth four-quarter percent change in real GDP was used (thus the percent change of the quarter against the last year's same quarter was calculated, e.g. the change from 1Q2004 and 1Q2003 real GDP was used). GDP is standard measure of aggregate economic activity and the four-quarter horizon answers the frequently asked question - what happens the next year? The sample period starts from 1Q2000 and ends on 1Q2016. This time range covers the period before financial crisis, period of financial crisis and period after financial crisis. The 257 basic model is designed to predict real GDP growth/decrease four quarters into the future based on the current yield spread (Bonser-Neal and Morley, 1997). This was accomplished by running of a series of regressions using real GDP activity and the different spreads lagged four quarters (the interest rate spread used for 3Q2001 is actually from 3Q2000). The last step is to find out which bond spread is the best for which country and to prove the hypothesis that the spread between 10-year and 3-month is the best one. To generate the GDP predictions with the different spread the regression using the whole sample was run, and later on two divided samples of real GDP and spreads of each selected country (the sample is divided in 4Q2007/1Q2008, because this period preceded financial crisis and should show some changes in prediction of the yield curve spread) were run. Time series data structure and ordinary least squares (OLS) method was used. All calculations were carried out in Gretl software. The coefficients a and 3 were estimated for each country: Real GDPt+4 =oc +(3 * spreadt + et (1) Where: RealGDPt+4is a prediction of the future real GDP in time t+4 quarters spreadtis spread between 10-year and 3-month state bonds in timet £t is a white noise 3 Results and Discussion The tests of normality were carried out. For the evaluation of the normality test is probably the easiest to observe the result from graph of the assumed normal distribution in comparison to the actual distribution of residues and analyse p-values of Chi-square test. We test the hypothesis HO: Residuals are normally distributed, against the hypothesis HI: Residuals are not normally distributed, the significance level of a was chosen as 0,01. If the p-value is greater than a then we cannot reject the HO, therefore the residuals are normally distributed. The test contributed that the data have normal distribution. For the testing of heteroscedasticity we chose the White's test. We test the hypothesis HO: Constant variances of residuals - homoscedasticity, against HI: Heteroscedasticity. The significance level of a was chosen as 0.01. If the p-value is greater than a then we cannot reject HO, therefore it contributes homoscedasticity. Results of Regression - Whole Sample The whole sample of dataset contains the real GDP from 1Q2000 to 1Q2016. A regression of the whole sample was run and we got the results as seen in Table 1. Surprisingly we have got different results of spreads for all countries (France - 5-year and 1-year spread, Germany - 5-year and 3-month spread and Great Britain 15-year and 3-month spread). We can say that models for France and Germany are statistically significant, because the p-values are under 1% (***) or 10% (*), however the R2 are very low. The model for Great Britain cannot be used as predictive due to high p-value. Table 1 Results of All Countries and Whole Sample from OLS Regression 1Q00 - 1Q16 Constant Spread P - value R2 France (5Y-1Y) 0.00706371 0.670837 0.0705 * 0.050980 Germany (5Y-3M) -0.0001300 1.54804 0.0019 *** 0.142821 Great Britain (15Y-3M) 0.0213558 -0.20809 0.2110 0.024717 Source: Author's own calculations 258 The R2 coefficients (coefficients of determination) show us how many percent of the sample can be explained by these models. For example we can say that future real GDP of Germany will be: Real GDPGermany t+4 = -0.00013 + 1.54804* spreadGermany t By this model we can predict future real gross domestic product for Germany four quarters ahead. We can test the hypothesis that the behavior of the spread and gross domestic product has changed during the financial crisis, therefore the sample was divided into two samples in order to prove this hypothesis. Results of Regression - Divided Samples The research continued as follows - the whole sample was divided into two samples. The first one is from 1Q2000 to 4Q2007, the second one is from 1Q2008 to 1Q2016 in order to show if there is any change of behavior and dependency between the variables before or after the financial crisis. Regressions of the first sample and the second sample were run. The results for the time span of 1Q2000 - 4Q2007 (first sample) are possible to see in Table 2, the results for the period of 1Q2008 - 1Q2016 (second sample) are in Table 3. In the first period again we cannot judge which spread is the best because in every country we have got different results (France - 30-year and 3-month, Germany 10-year and 1-year, Great Britain 30-year and 1-year). Table 2 Results of All Countries and Sample from 1Q2000 to 4Q2007 1Q00 - 4Q07 Constant Spread P - value R2 France (30Y-3M) 0.0116757 0.479692 0.0578 * 0.114793 Germany (10Y-1Y) 0.0269465 -0.877121 0.0771 * 0.100484 Great Britain (30Y-1Y) -0.00013005 1.54804 0.0019 *** 0.142821 Source: Author's own calculations We can say that all models are statistically significant, because the p-values are under 1% (***) or 10% (*). R2 are higher than in the time period of whole sample - 1Q2000 -1Q2016, but still it is not anything special. In the second period the best results were gained for spreads mentioned in the Table 3 -30-year and 1-year (France and Great Britain), 10-year and 1-year (Germany). We can say that in this case the best spread is a spread of 30-year and 1-year. All models are statistically significant. R2 are higher than in the previous time span, which is interesting. This change in prediction possibility may be caused by different behavior of financial markets after the financial crisis (after year 2008). The results show that the models have much higher explanatory power after the year 2007. Table 3 Results of All Countries and Sample from 1Q2008 to 1Q2016 1Q08 - 1Q16 Constant Spread P - value R2 France (30Y-1Y) -0.0149132 0.791468 0.0011 *** 0.293714 Germany (10Y-1Y) -0.0220346 2.24746 0.0001 *** 0.384775 Great Britain (30Y-1Y) -0.0140444 0.927291 0.0003 *** 0.349980 Source: Author's own calculations At the end we can summarize the new theoretical finding according to which spread is the best for predicting of the future GDP growth. We proved that in these selected countries the best spread is a spread of 30-year and 1-year government bonds (we have added all results together and this spread showed up three times in total). The second best spread is spread of 10-year and 1-year bonds (totally twice in our calculations). The results showed that dividing of the sample made a difference between pre-crisis and 259 after-crisis period and it showed the different relationship of spreads and the models. The finding that the best spread is spread of 30-year and 1-year eventually 10-year and 1-year is in contradiction with the theoretical background when almost everybody who predicts the future GDP growth uses a spread of 10-year and 3-month government bonds. This was found out on data of United States of America (from 1970 to 2000). We must say that to collect data of 10-year and 3-month government bonds is the easiest possible way, when you want to use them for calculations, because they are all published in Bloomberg database, however to get data for 20-year, 6-month and 1-month yields are almost impossible in demanded time period and a good quality (there are many blind values from 1Q2000 to 1Q2016). 4 Conclusions The 10-year and 3-month spread has substantial predictive power and should provide good forecast of real growth four quarters into the future (this was proved in USA). We showed that after the year 2000 the best predictive spreads in France, Germany and Great Britain are the spreads of 30-year and 1-year and 10-year and 1-year government bond yields. The results presented above confirm that these spreads have a significant predictive power for real GDP growth and the behaviour of the models changed during and after the financial crisis. The results show that the dividing of the sample made a difference in use of the best predictive spread. The simple yield curve growth forecast should not serve as a replacement for the predictions of companies, which deal with predicting of many economic indicators, it however does provide enough information to serve as a useful check on the more sophisticated forecasts. Future research could be extended to a wider examination of the best spreads in more countries around the world and especially in European Union. It would be interesting to see if there is the rule which would prove the hypothesis that the spread of 10-year and 3-month bond yields is the best for predicting future GDP growth in the countries of the European Union. Acknowledgments The support of the Masaryk University internal grant MUNI/A/1025/2015 Risks and Challenges of the Low Interest Rates Environment to Financial Stability and Development is gratefully acknowledged. References Berk, 1, van Bergeijik P. (2001). On the Information Content of the Yield Curve: Lessons for the Eurosystem? Kredit und Kapital, vol. 34, pp. 28-47. Bernard, H. 1, Gerlach, S. (1998). Does the Term Structure Predict Recessions? The International Evidence. CEPR Discussion Papers, vol. 1892. Bloomberg (2016). 30-year, 15-year, 10-year, 5-year, 1-year, 3-month Government Bond Yields. Retrieved from: Terminal Bloomberg. Bonser-Neal, C, Morley, T. R. (1997). Does the Yield Spread Predict Real Economic Activity? A Multicountry Analysis. Economic Review: Federal Reserve Bank of Kansas City, vol. 5(3), pp. 37-53. Chinn, M., Kucko, K. (2010). The predictive power of the yield curve across countries and time. NBER Working Paper Series. De Pace, P. (2011). GDP Growth Predictions through the Yield Spread: Time - Variation and Structural Break. Euro Area Business Cycle Network. Retrieved from: http ://ssrn.com/a bstract= 1401752. 260 Estrella, A., Hardouvelis G. A. (1991). The Term Structure as a Predictor of Real Economic Activity. Journal of Finance, vol. 46(2), pp. 555-576. Estrella, A., Mishkin, F. S. (1996). The Yield Curve as a Predictor of U.S. Recessions. Current Issues in Economics and Finance, vol. 2(7), pp. 1-6. Giacomini, R., Rossi, B. (2005). How stable is the Forecasting Performance of the Yield Curve for Output Growth. Oxford Bulletin of Economics and Statistics, vol. 68(1), pp. 783-795. Mishkin, F. (1990). Yield curve. NBER Working Papers Series, vol. 3550. OECD statistics (2016). Quarterly National Accounts: Quarterly Growth Rates of Real GDP, Growth rate compared to the same quarter of previous year. Retrieved from: http://stats.oecd.org/index.aspx7queryid=350. Szarowska, I. (2013). Fiscal Discipline as a Driver of Sovereign Risk Spread in the European Union Countries. In: 22nd IBIMA Conference on Creating Global Competitive Economies: 2020 Vision Planning & Implementation. Rome: IBIMA, pp. 793-804. 261 Difference in Financial Knowledge of Finance Students in the Czech Republic Barbora Chmelíková1, Martin Svoboda2 1 Masaryk University Faculty of Economics and Administration, Department of Finance Lipová 41a, 60200 Brno, Czech Republic E-mail: barbora.chmelikova@gmail.com 2 Masaryk University Faculty of Economics and Administration, Department of Finance Lipová 41a, 60200 Brno, Czech Republic E-mail: svoboda@econ.muni.cz Abstract: The aim of this article is to analyze the results of the study regarding financial literacy topics. The study was implemented at Masaryk University among the students of the Faculty of Economics and Administration. The paper presents the results of empirical study conducted in two consecutive years 2013-2014 testing the students' knowledge in the areas of personal budgeting, numerical literacy, price literacy, payment methods, how to search for relevant information, breach of a contract and its consequences, indebtedness, right and obligation of consumers in the financial markets. The results of data gathered from the studies over the time enable an insight into the shift of the financial knowledge of the surveyed target group. Keywords: financial literacy, university students, financial education JEL codes: A10, A22, A23, 123 1 Introduction Financial literacy is an integral part of today's skill set essential for the harmonic functioning of an individual in the financial markets. It is not only about literacy anymore, but about financial literacy emphasizing the importance of such a significant role the money plays in human life. Financial literacy research identifies groups which might have low levels of financial literacy (Atkinson & Messy, 2015), and these groups include also youth. Yet age is not the only factor influencing the level of financial literacy. The level of education plays important role as well (Xu & Zia, 2012). Positive correlation between the achieved level of education and the level of financial literacy exists based on the international surveys (Atkinson& Messy, 2012). Financial literacy was tested in three areas - financial knowledge, financial behavior and attitudes. For this reason, paper examines the financial knowledge among university students, in particular, students majoring in the fields of economics and finance. This paper builds on the results of study showing not only the lack of knowledge regarding financial matters of finance students (Chmelíková, 2015a), but also of students of other disciplines (Chmelíková & Svoboda, 2014). Furthermore, finance students did not demonstrated considerable exceptional performance in the study testing their personal money management either (Chmelíková, 2015b). Therefore, this paper presents data collected in years 2013-2014 with university students majoring in finance related fields as a target group to see whether there is any difference in knowledge concerning financial topics in this period. 2 Methodology and Data The survey questionnaire was based on the "Empirical verification of university students' literacy" project following the research conducted at the University of Economics (Hradil et al., 2012) to identify the level of financial literacy of university students which appeared to be higher to the certain extent than the adult's financial literacy (Ministry of Finance, 2010). The Faculty of Economics and Administration joined the project through the Institute for Financial Market and contributed to the formation of the questionnaire and its modification in accordance with the standard international best practices. 262 Questions regarding measuring the level of financial literacy (Lusardi et al., 2010) were added to the questionnaire set as well. This paper concentrates on the shift of the financial knowledge between two years. The established hypothesis is whether there is a positive change in financial knowledge of students studying in finance related major. University students were selected as a target group which was further narrowed to the students of the Faculty of Economics and Administration at Masaryk University. The questionnaire was presented mostly to the students of the course Basic Finance which is generally taught in the first year of study. The analyzed data was collected during the two-year period in the year 2013 and 2014. Respondents differ in each year. Also, the sample in the first year (n = 885) is larger than the second year sample (n = 393). The number of respondents to each question varies within the sample according to the number of already answered questions as well as the complexity of the question. The study was conducted by online survey consisting of 100 questions giving respondents more options on the condition that one option is correct. On the other hand, some questions were open-ended with the necessity to write the response. Minority of questions more than one answer was correct. Table 1 The Topics of the Areas Tested in the Questionnaire Topic of the question's block Number of questions in each block Price and financial literacy 13 Household finance management and debts 8 Investments 4 Loans and interests 7 Insurance 4 Law Literacy 8 Social system and benefits 18 Source: Author's own work based on survey results Table 1 provides the topics of the areas tested in the questionnaire and number of questions in each block which were analyzed in this paper. The questionnaire structure is designed in order to test different areas of financial knowledge such as price and financial literacy including questions about pricing, payment methods in domestic and foreign currencies, macroeconomic situation and inflation. Next block consisted of questions regarding household finance management and debt management testing personal finance. Third block of question tested knowledge about investments, fourth block focused on loans, credit products, and interest rates. Following areas covered was insurance and its basic principles as well as fraudulent behavior. The law literacy block included questions of understanding contract arrangements, fraud, who to contact in case of financial difficulties. Last tested area involved knowledge of Czech social benefit system and its current state. In order to test established hypothesis the data was analyzed using mainly the descriptive statistics method due to the nature of collected data. Scores in each question block were calculated as a mean of the percentage of correct answers weighted by number of respondents. For the purpose of comparison, in each year the same questions were presented to the respondents hence the collected data are comparable and suitable for analysis. The data analysis conducted in this paper follows the previous research and results for the year 2014 are adapted from previous study (Chmelíková & Svoboda, 2015). 3 Results and Discussion First of all, we will look closely on the self-assessment questions in which the students were asked if they think that they have a good knowledge in the field of personal finance. As we can observe from the Figure 1 the confidence of students between the year 2013 and 2014 rather declined. The percentage of confident students decreased, whereas the percentage of not confident students increased. 263 Figure 1 Confidence about Personal Finance Knowledge Do you think that you have a good knowledge in the field of personal finance? (A c 70,0 ■g 60,0 c 50,0 o 40,0 & 30,0 a! 20,0 £ 10,0 o 0,0 4'8 4,1 58,7 ^47,6 29,1« |41,0 3'5 3,1 3'9 4,3 Definitely yes Rather yes Rather no Definitely no I don't know ■ Year 2013 4,8 58,7 29,1 3,5 3,9 ■ Year 2014 4,1 47,6 41,0 3,1 4,3 Source: Author's own work based on survey results In contrast, students' opinion about the financial literacy level of the Czech population increased between the year 2013 and 2014 as it is shown in Figure 2 where we can observe a decreasing shift in negative opinion which might indicate optimistic opinion of the financial literacy level in the Czech Republic. Figure 2 The Level of Financial Literacy of Czech Population according to the Students In you in | 80,0 g 60,0 | 40,0 £ 20,0 ^ 0,0 ir opinion, what level of fiancial literac does the Czech population have? 63,9 ■ 1 0,9 1,4 V 54,8 Good Average Bad ■ Year 2013 0,9 35,2 63,9 ■ Year 2014 1,4 43,8 54,8 Source: Author's own work based on survey results The shift of students' opinion in self-assessment questions might explain also the shift in the financial knowledge scores. There is visible decrease in financial knowledge in each individual block of questions with one exception. The calculated results are illustrated below in Figure 3. Figure 3 Difference in Financial Knowledge of Finance Students Difference in Financial Knowledge 53,7 in L. o u 47,0 45,8 |*6,7 ^5,1 Price and financial literacy Household financial ma nagem ent and... Investme nts Loans and interests Insurance Law literacy lYear 2013 24,6 37,7 34,7 53,7 47,0 45,8 lYear 2014 22,7 32,3 32,6 46,7 45,1 39,5 Source: Author's own work based on survey results 264 Provided graph illustrates the differences in financial knowledge according to the tested areas. Darker columns present the scores in percentage for the year 2013, whereas the lighter-coloured columns represent data for year 2014. The calculated scores are shown in the bottom of the Figure 3 for each year and each block of questions. Presented graph shows decreasing pattern in measured financial knowledge of students between the year 2013 and 2014. The exact difference in scores is provided below in Table 2. The highest drop in scores by 7% is witnessed in the topic of loans and interest, followed by the drop of 6.3% in law literacy and decrease of 5.5% in household finance management and debt management. Minor, but still negative, change is occurred in the tested area concerning investments (-2.1%), insurance (-1.9%), price and financial literacy (-1.9%). Only positive difference in knowledge is observed in the topic of social system and benefits, where there is an increase in score by 0.3%. Table 2 The Change in Financial Knowledge during the Period 2013-2014 Financial Knowledge Topics Difference in % Price and financial literacy -1.9 Household finance management and debts -5.5 * Investments -2.1 Loans and interests -7.0 * Insurance -1.9 Law Literacy -6.3 * Social system and benefits + 0.3 Source: Author's own work based on survey results In Table 2 the actual change in financial knowledge between the year 2014 and 2013 is calculated and the difference in each area is shown in percentage. In all areas, except one, the change is negative when the students in the year 2013 achieved higher scores than the students in the year 2014. Table 2 shows how large the difference is. Difference larger than 5% is marked with asterisk (*). 4 Conclusions The results demonstrated the decrease in financial knowledge of participating students between the year 2013 and 2014 implying the conclusion that the education concerning financial literacy, personal finance, debt management, financial products is needed even among the students studying in finance related fields. This might have further implications since these students are likely to have carrier in the finance sector or advising others on personal finance matters which if this decreasing effect continues, it may have larger negative effect. For this reason, the financial education in the area of financial literacy should be further developed. Acknowledgments Support of Masaryk University within the project MUNI/A/0916/2015 is gratefully acknowledged. The authors want to thank the Citi Foundation for the support. References Atkinson, A., Messy, F. A. (2015). Financial Education for Migrants and their Families, Paris: Organisation for Economic Co-operation and Development. Chmelíková, B. (2015a). Financial Literacy of Students of Finance: An Empirical Study from the Czech Republic. International Journal of Social, Behavioral, Educational, Economic, Business and Industrial Engineering, vol. 9, pp. 4000-4004. Chmelíková, B. (2015b). Personal Money Management of Finance Students. In: Proceedings of 32nd International Business Research Conference. Melbourne: World Business Institute Australia, pp. 11. 265 Chmelíková, B., Svoboda, M. (2015). Comparative Study: The Financial Literacy of Students of Economics and Finance versus Students of Law. In: Proceedings of the 12th International Scientific Conference European Financial Systems 2015. Brno: Masaryk University, pp. 66-72. Hradil, D., Křížek, T. and Dvořák, P. (2012). Empirické ověření gramotnosti student vysokých škol, akademický výzkumný projekt, IGA 27/2012, IG102032, Praha. Lusardi, A., Mitchell, O. and Vilsa C. (2010). Financial Literacy Among the Young. Journal of Consumer Affairs, vol. 44(2), pp 358-380. Messy, F.A., Atkinson, A. (2012). Measuring Financial Literacy: Results of the OECD/International Network on Financial Education (INFE) Pilot Study. Paris: Organisation for Economic Co-operation and Development. Ministry of Finance (2010). Národní strategie finančního vzdělávání. Retrieved from: http://www.mfcr.cz/cps/rde/xbcr/mfcr/Narodni_strategie_Financniho_vzdelavani_MF201 0.pdf. Xu, L., Zia, B. (2012). Financial Literacy around the World: An Overview of the Evidence with Practical Suggestions for the Way Forward, The World Bank. 266 The Use of Financial Advisory in Czech Republic: Self-confidence Věra Jančurová1, Petra Formánková2 1 Masaryk University Faculty of Economics and Administration, Department of Finance Lipová 41a, 602 00 Brno, Czech Republic E-mail: 321156@mail.muni.cz 2 Masaryk University Faculty of Economics and Administration, Department of Finance Lipová 41a, 602 00 Brno, Czech Republic E-mail: 420853@mail.muni.cz Abstract: The study of behavioral finance and decision making, especially in finance, have grown on importance during last decades. There were conducted many studies of decision making at financial markets all around the world. This study was conducted to analyze financial decision making and behavioral characteristics of Czech population. The purpose of this study is to show how people in Czech Republic control their financial affairs; one of the studied areas is if they use the help of financial advisors, especially, if this is influenced by gender, the level of education or even by the self-evaluation of own financial skills and knowledge. There was conducted a field research based on electronic and printed questionnaire that was answered by 302 respondents. Based on the results of the testing it was confirmed that level of education is significant for both self-evaluation and the use of financial advisors. However the hypothesis that higher education level increase the probability of self-evaluation and decreases the need to use the help of financial advisors was not confirmed. One of possible explanation could be the overconfidence of individuals that is strongly shown by the sample; 70,52% of total respondents consider their financial knowledge better than average and only 3,32% of the respondents consider them worse than average. And only 29,6% pf the respondents use for their financial decisions the help of financial advisor. Keywords: behavioral finance, financial advisory, overconfidence, self-evaluation, financial decision JEL codes: G110, G190, G020,120 1 Introduction The growing importance of behavioral finance and financial advisory are obvious at financial markets. Financial advisory is nowadays highly discussed topic. The orientation at financial markets is complicated, often requests the assistance of financial advisors. The range of issues individuals have to solve in the context of finance is significant and influence often the quality of their life. Behavioral finance emphasizes the role of psychology in finance in guiding individual investor's decisions and influencing financial market. Behavioral finance aims to explain behavior of individuals and financial market movements based on psychology-based theory. The general assumption of behavioral finance is that the individual's characteristics and information structure influence significantly financial markets. Despite the fact that human decisions are often influenced by human psychology, even in determining economic or financial decision, classical theory insist on fully rational individuals and efficient market. However although the behavioral finance theory and the research in this field is young, its origins go back to Adam Smith. Smith found out that people decide based on impressions and believer, rather than on rational data analyzes. This article provides an analysis of human decision making in the context of use of the help of financial advisors. The goal of the study is to verify whether there can be found connection between using the assistance of financial advisors and gender, reached level of education or self-evaluation of own skills. 267 • The first hypothesis concentrates on how people handle with their money and how to control this handling. Do the respondents use the help of financial advisor or do they rely on own skills and knowledge? • Second hypothesis concentrates on the influence of gender, level of education and self-evaluation of financial skills on the use of financial advisory. • Third hypothesis tests self-evaluation, itself, and its dependency on gender and level of education. Theoretical Backgrounds For the purpose of this study the background of financial advisory and self-confidence or overconfidence has to be mentioned. For the needs of this study financial advisory is considered to be service offered to people to help them with financial decision, especially insurance, savings and investments. Every person needs to make such decisions disregarding on gender or education level. Financial advisor is a person who provides financial advices and guidance to the clients. Self-evaluation is the evaluation of own skills and knowledge at financial markets. It is closely connected to overconfidence, the tendency to overestimate own predictive abilities and the precision of information individual have. Individuals generally overestimate their skills and knowledge, the perceived level of own skills and knowledge is typically high, higher that actual skills and knowledge. In this framework "one of the greatest services a financial advisor can provide to clients is helping to ensure that in times of market turbulence, reason, discipline and objectivity triumph over emotions such as fear, greed, and market regret." 2 Methodology and Data The data for the purpose of this study were collected in a controlled field research, based on a questionnaire proving inclination to behavioral biases and information about financial advisory used by respondents. The survey was divided into two separate parts, altogether 40 questions. First the respondents were asked personal information; e.g. age, gender, highest level of education reached. Second part contained questions asking respondents questions regarding financial decisions - especially use of financial advisors. Within this part respondents were asked to evaluate their own knowledge and skills in finance. The sample consists of 302 respondents, the age between 18 and 70 years. Sample characteristics: • Gender: 70% women, 30% men • Age: 18-30: 50%; 31-50: 30%; 51-65: 17%, more than 65: 3% • Education: secondary school: 2%, high school with school leaving exam 45%, high school without school leaving exam 11%, higher school 4%, university 38% • Marital status: single/divorced 50%, mate 10%, married 39%, widow/widower 1% 3 Results and Discussion The results of the study show that the sample can be considered strongly biased, at least by the overconfidence. First there was tested if there is a difference in use of financial advisory by gender. The zero hypotheses were not rejected. Table 1 Financial Advisory Usage Based on Gender P-value (t - test) P-value (F-test) 0,246 0,620 Source: Own based on conducted research 268 This table summarizes differences between the group of respondents who use/ do not use financial advisors. The use of financial advisors seems not to be influenced by the gender. Received p-value of t-test is higher than the significance level. Moreover, this test does not comply with one of the requirements - the normal data distribution. High p-value rejects a statistical difference between the groups. Based on received results there are no differences between gender in the use of financial advisors. 30,91% of women answered that they use for their financial decisions the help of financial advisors and 26,96% of men answered the same. However the statistical testing shows that the normality assumption is not fulfilled. Therefore the test cannot be taken into consideration. Further test was taken to study if there can be found any connection between the use of self-evaluation of own financial knowledge and the use of financial advisors. Table 2 Usage of Financial Advisory Based on Own Evaluation of Financial Knowledge P-value (t - test) P-value (F-test) 0,291 0,012 Source: Own based on conducted research High p-value rejects a statistical difference between the groups. Despite the fact that this tests complies with all assumptions, normal data distribution, as well as homoscedasticity are fulfilled, the effect of self-evaluation of own financial knowledge on use of financial advisors is not statistically significant. Reached education level was tested as another criterion for the use of financial advisors. Table 3 Usage of Financial Advisory Based on Education Level P-value (t - test) P-value (F-test) 0,011 0,045 Source: Own based on conducted research A significant gap is spotted in terms of the usage of financial advisory based on reached level of education. The computed p-value rejects the null hypothesis (claiming that the education has no impact on the use of financial advisory). Low p-value confirms a statistical difference between the groups. In the right column, there is p-value of F-test providing us with verification of model correctness. Current situation at financial market, especially high number of products and their complicated structure, built up the hypothesis that people with lower education level use more often the help of financial advisors. Following graph shows that this hypothesis is not correct. Figure 1 Usage of Financial Advisors by Reached Level of Education 50,00% 40,00% 30,00% 20,00% 10,00% 0,00% i III I ■<5^ c^" c l Univeristy l High school without SLE l High school with SLE Source: Own, based on conducted research From the graph it is visible that the lowest level of use of financial advisors has been identified by people whose education level is the lowest, secondary school. However this result can be influenced by the number of respondents from this group. On contrary the 269 help of financial advisors is mostly used by people with highest education level, over 35% of respondents with university education use the help of financial advisors. A significant gap can be found between respondents with high school education depending on the type of high school, with or without school leaving exam. Respondents with school leaving exam tent to use the help of financial advisors more that their colleagues without the school leaving exam. To conclude the study of use of financial advisors by individuals, it has to be stated that from the total sample only 29,6% or respondents use the help of financial advisors. This can be caused by the situation at Czech financial advisors market. The following two tests were taken to analyze how do respondents evaluate their skills and knowledge in general financial issues, as insurance, savings, investments etc. First test examines if there is any connection between evaluation of own financial skills and gender of respondents. Table 4 Own Financial Skill Evaluation (by Gender) P-value (t - test) P-value (F-test) 0,950 0,022 Source: Own based on conducted research High p-value declines a statistical difference between the groups. Despite the fact that this tests complies with all assumptions, normal data distribution, as well as homoscedasticity are fulfilled, the effect of this bias is not statistically significant. This characteristic of respondents is not a statistical significant difference, since p-value exceeds the set significance level. As gender seems to have no impact on evaluation of own skills, the second test was considering the reached level of education's impact on self-evaluation. Table 5 Evaluation of Own Skills Based on Education P-value (t - test) P-value (F-test) 0,046 0,162 Source: Own based on conducted research Low p-value confirms a statistical difference between the groups. In the right column, there is p-value of F-test providing us with verification of model correctness. Based on conducted test education level was confirmed to be significant for evaluation of own skills. As the education level was confirmed to be statistically significant for self-evaluation of financial skills and knowledge further analyses was done to confirm/decline if higher educational level means better self-evaluation. This logically set hypothesis was tested and showed the contrary. Following figure shows that respondents with lower level of education, especially respondents with the lowest level of education, seem to over-evaluate their skills. This result could be caused by overconfidence. Figure 2 Usage of Financial Advisors by Reached Level of Education 150 100 50 0 I HURE STEJNĚ LÉPE VS SS s SS s vyučením maturitou ZS VOS Source: Own, based on conducted research 270 In total 70,52 of total respondents consider their financial knowledge to be better than average and only 3,32% of the respondents consider them to be worse than average. To conclude, it is obvious that financial advisory is not used enough and that there is also need to explain people with lower level of education the benefits of such a service. It could be the goal of future studies to investigate on why the financial advisory in Czech Republic not broadly used is and how to improve that. The comparison with other countries could be also interesting. 4 Conclusions The first hypothesis tested how people handle with their money and how to control this handling. Based on the received data only 29% of respondents use the help of financial advisors, it seems to be obvious, that most of them rely on own skills and abilities. Second hypothesis concentrated on the influence of gender, level of education and self-evaluation of financial skills on the use of financial advisory. Gender and Self-evaluation of own skills were not confirmed to be significant for the use of financial advisor. On contrary reached level of education was confirmed to be significant. However the result is surprising because with higher level of education increases the use of financial advisor. The research shows that people with lowest level of education do not use financial advisors. Third hypothesis tested self-evaluation, itself, and its dependency on gender and level of education. Gender was again not confirmed to be significant, however level of education seems to influence the evaluation of own skills. However, the hypotheses that higher education increase the self-evaluation was rejected. Respondents with lowest education evaluate them better then respondents with higher education level. Acknowledgments Support of Masaryk University within the project MUNIA/0916/2015. "Behavioral and knowledge aspects of trading and evaluation of financial assets" is gratefully acknowledged. References Baker, H. K., Vctor, R. (2014). Investor Behavior: The Psychology of Financial Planning and Investing, lsted. New Yersey: Wiley. Goldberg, K., Nitzsch, R. (2004). Behavioral Finance: Gewinnen mit Kompetenz. 1st ed. Muenchen: FinanzBuch Verlag GmbH. Green, H. (1993). Econometric Analysis. 2nd ed. New York: Macmillan. Howard, T.C. (2014). Behavioral Portfolio Management, lsted. London: Harriman House. Nofsinger, J. R. (2011). The Psychology of Investing. Boston: Pearson Education. Pompian, M. (2006). Behavioral Finance and Wealth Management. 1st ed. New Jersey: John Wiley & Tamp; Sons, Inc. Seppala, A. (2009). Behavioral Finance of Investment Advisors. Master's thesis. Helsinki: Helsinki School of Economics, pp. 1-72. Shefrin, H. (2002). Beyond Greed and Fear: Understanding Behavioral Finance and the Psychology of Investing, lsted. New York: Oxford University Press. 271 Face Nominal Effect on Capital Market Transactions. The Case of Poland Magdalena Jasiniak University of Lodz Faculty of Economics and Sociology, Corporate Finance Department Rewolucji 1905 No 39, Lodz, Poland E-mail: magdalena.jasiniak@uni.lodz.pl Abstract: Psychological features have an important role in evaluation processes. Studies indicate that investment decisions are strongly determined by behavioral factors. As a result investors behave irrationally. One of parameters that influence purchasing decision is price. Perceiving price as high generally discourage the purchase. As a consequence, goods at lower price are more attractive to bought what creates higher demand and influence companies' revenues. Following paper analyze whether the face nominal effect observed at consumer market also occur in stock exchange market. The main aim of this article is to verify whether investors are influenced by face nominal values on stock market exchange. Author state that cheap assets characterized by nominally lower prices are more attractive to buy and bring higher profits (as a consequence of increased demand) in comparison to assets described as expensive. In order to verify the hypothesis, database of 13789 quotations from 1.07.1999 to 30.12.2013 was created. The sample was divided into three groups - cheap, average and expensive stocks. Finally the statistical analysis was conducted among 2924 records including only cheap and expensive units. Statistical analysis confirm that assets defined as "cheap" generate higher profits and lower losses. Keywords: face nominal effect, behavioral finance JEL codes: G02, Gil 1 Introduction The final effect of investment process is determined not only on strictly economic factors but also on psychological aspects. Emotions and risk tolerance affect human behavior and determine the investment decisions. Economic psychology is focused mainly on consumer behavior and explains the processes that are related with purchasing goods and services. These analysis are mostly used in marketing strategies. However, the aspects of behaviorism are also observed on stock exchange market. The theory of rational behavior assumes that consumers choices are made on the basis of rational reasons. In the process of investment, the rational behavior means that investor analyses all possible choices, assigns importance to possible options and choose the best one. In predicting future event people are impartial and objective. This procedure maximize the subjective expected utility. However, due to certain limitations of time and in limited access to information, investors avoid rational methods and use faster and simpler procedures. Additionally, they are influenced by behavioral biases which sometimes cause irrational and contradictory decisions. Perceiving price as high generally discourage the purchase. This phenomenon is commonly observed at consumer market. As a consequence, goods at lower price are more attractive to bought what creates higher demand and influence companies' revenues. The question is whether stock market investors are also fragile on stocks' nominal value. Following paper analyze whether the face nominal effect observed at consumer market also occur in stock exchange market. The main aim of this article is to verify whether investors are influenced by face nominal values on stock market exchange. Author state that cheap assets characterized by nominally lower prices are more attractive to buy and bring higher profits (as a consequence of increased demand) in comparison to assets described as expensive. Paper presents the results of preliminary study focused on the relation between nominal values of stock prices, investors' 272 willingness to purchase and profits achieved. It is assumed that investors are more willing to buy stocks at lower prices. The increase in demand is one of determinants that influence stock price increase and finally rates of return. As a consequence, it is expected that low priced stocks generate higher profits than high priced stocks. 2 Methodology and Data How prices affects consumers' decisions According to micro theories, prices have a strong influence on people's purchase decisions. The theory of demand generally explains the relation between the consumers' demand on particular product at given price and time. The general relation between the price and demand is negative - higher prices cause the decrease of consumers' willingness to products purchase. Consumers who behave rationally generally choose products with lower prices. However, the assumption of consumer rational behavior is neglected. Studies focused on consumers behaviour let to highlight some paradoxes' in price and demand relation. Table 1 Paradoxes in Demand Theory Giffen paradox Related with basic products - despite price growth, consumers are still willing to buy these products._ Veblen paradox Increase of luxury products purchase despite their price increase. Scene effect Purchase of goods despite price value, only because product is bought by others. Snob effect Purchase of goods is the more attractive the less people have them. Income effect of price decline The price decrease of substitute products cause similar results to income increase Bolt effect Maintenance of current consumption despite price increase. Shock and effect domestication After price rapid increase customers decline consumption and after some time return to the previous level._ The phenomenon of shopping anticipation Despite price increase, consumers are stil buying products in case of future price growth._ Speculative paradox Reaction on price increase by buying goods in order to future sold at a profit. Source: K. Mazurek - topaciriska, Zachowa Wydawnictwo Akademii Ekonomicznej im nia nabywcowjako podstawa strategii marketingowej, Oskara Langego we Wroclawiu, Wroclaw 1997, s.61 In the field of economic psychology, there are many examples of limited utility of normative models based on human rational decisions. For example: sudden and impulsive purchase decisions, excessive risk aversion, risk overestimating in less known situations and risk underestimating in better known situation (Zaleskiewicz, 2015). According to H. Simon (1979) people's decision are not optimal but rather satisfying and include psychological context. H.Shefrin and M.Statman (1985) state that investors are motivated by willingness to achieve profits but this aim is balanced between the fear of loss and hope of spectacular profit. People are sensitive to prices, especially when they have to bear higher costs. Perception of high prices generally discourage the purchase (Falkowski, Tyszka, 2006). This is strongly used on consumer market and marketing actions. 273 Price perception is determined by many factors. Anchor theory is one that explains how consumers perceive product price. Numbers, their informational value and influence on human behavior is a subject of many studies. The effect of left - side number is most commonly used in price strategies. This is strongly related with human simplified perception of reality. More concentration is focused on left side values while the right are ignored. As a result, prices set at 9.99 are perceived as 9.0 not as 10.0. This influence the rationality of buying decisions what has been used in marketing strategies for many years. This simple method cause that certain prices are perceived as much lower than they really are. The highest influence of this effect is observed in case of threshold prices. In case of euro currency - f.e - 0.99 (less than 1 eur), 4.99 (less than 5 eur), 99.99 (less than 100 eur). As a result, the price decrease from 2.0 to 1.99 is perceived as higher than from 2.50 to 2.19 despite nominally the difference is higher in second case. Behavioral biases on stock exchange market Literature describes many investors' behaviors that are driven by non - rational factors. The most popular is the theory of perspective described by Tversky and Kahnemann (1979). This theory is an alternative to theory of expected utility and explains that the perception of risk differs depending on previously experienced profits or losses. According to this theory, experienced loss drives investors to more risky behavior. In contrast Johnson and Thaler (1990) explained the risk tolerance by previous experiences in following way: previously experienced loss increase the fear of risk and prior earning increase risk appetite. Samuelson and Zeckhauster (1988) observed that investors are more willing to maintain stock they previously bought, being convinced that they it will ensure them profits and focusing only on these market information that confirm this assumption. Anchoring effect is also described on stock exchange market and is perceived as one of most important that influence stock prices. Investors made decisions on the basis of recent quotations (Kahneman, Tversky, 1979). According to this theory consumers made the decisions on the basis of previously noticed information that influence further interpretation. This effect is strongly used in marketing strategies by positioning product price with the previous price of the same product or similar products. Experienced loss, that was too strongly encoded in human memory might limit future engagement on financial market. This phenomena is called snake bite effect. However, the need of equal opportunities aiming to fast compensation occurs regardless of risk (Nofsinger, 2006). Face nominal effect has been not widely studied on financial markets. Low priced stocks (valued less than 1 PLN) are usually perceived as "junk stocks". Low priced stocks are often related with companies at bad financial conditions. Therefore "junk stocks" are perceived as risky and highly speculative. Low priced stocks are usually the result of splits which are conducted to improve stocks liquidity. They usually attract novice investors. However, there are studies indicating that nominals may affect human decisions in financial transactions as well. According to T. Odean (1999), investor have higher willingness to sell assets with higher prices than those with lower prices. However, this phenomena was strongly related with the disposition effect and the effect of such behavior appeared to be negative. The high stock could bring higher benefits than the low ones. Low price effect is an anomaly which consist in that stocks with low prices bring higher return rates that stocks with high prices. This phenomena was observed by Goodman and Pevy (1986) and Branch and Chang (1990). This phenomena is usually related with stock split. It was illustrate by catering theory (Baker et all. 2009). When investors prefer stocks at low nominal value, manager do stock split and deliver assets at expected 274 prices. This is based on assumption that nominal value of stocks do matter for the investors and as a result the expected rate of return is related with stock price. Studies conducted on Polish Stock Exchange by Zaremba and Zmudziriski (2014) stay in contrast. It that case the low price effect was diverted. It was observed that stocks with high prices generate higher profits that low priced stocks. Studies conducted by Biegahska et al. (2016) on the example of M&A transactions proved that in case of stocks at lower prices the probability of profits is higher than in case of high valued stocks. The question is whether this phenomena might be observed in wider extend. 3 Results and Discussion Face nominal effect on capital market - research results Research was conducted on data collected from Polish Stock Exchange Market. In order to verify the hypothesis, database of 13789 quotations from 1.07.1999 to 30.12.2013 was created. At first, database was divided into two groups - those with positive brokerage recommendation (buy or accumulate) at t moment (date of recommendation issue) and those with negative recommendations (sell or reduce) at moment t. Next, the sample was divided into three groups - cheap, average and expensive stocks. Cheap stocks were defined as those characterized by single - digit prices - below 1 PLN per unit. Expensive stocks were those with four - digit prices - more than 1000 PLN per unit. Neutral stocks were excluded from the analysis. Author verified the rate of return after one year measured by 250 trading sessions. The analysis was conducted separately in the group of positive and negative brokerage recommendations. The results of analysis conducted among the group of positive recommendation is presented below. Table 2 Rate of Return (%) of Low and High Prices - Positive Recommendations Low prices High prices Mean 0.721046 0.350316 Standard error 0.218695 0.067061 Median 0.230882 0.201948 Standard deviation 1.956066 0.632657 Variance 3.826192 0.400255 Range 13.2 2.616643 Minimum -1 -0.91028 Maximum 12.2 1.706362 Sum 57.68366 31.17815 Number of units 80 89 Source: Own elaboration The descriptive statistics indicate that the average rate of return in case of low priced stocks is higher than in case of high priced stocks (respectively 0,72 vs 0,35). In case of low priced stock the rate of return is characterized by higher variance and the range of minimum and maximum rates is definitely higher. The highest return rates amounted 12,2% in case of low priced stocks and only 1,7% in case of high. The statistical significance of means was verified. The zero hypothesis assumed that the difference between average return rates of low and high priced stocks is equal 0. Zero hypothesis: the difference between both means = 0 Sample 1: N=80, mean = 0,721046, std. error = 0,218695 Residual std. error = 0,0244508 275 95% confidence interval for the mean from 0,672378 to 0,769714 Sample 2: N=89, mean = 0,350316, std. error = 0,0670615 Residual std. error = 0,0071085 95% confidence interval for the mean from 0,33619 to 0,364443 The test statistic: t (167) = = (0,721046 - 0,350316)/0,0243572 = 15,2205 Double - sided critical area p = 2,25e-033 One - sided critical area = l,125e-033 Test confirms that means are different. This allows to conclude that on Polish stock exchange market low priced stocks generate higher profits than high priced stocks. The results of analysis conducted among negative recommendations is presented in table 3. Table 3 Rate of Return (%) of Low and High Prices - Negative Recommendations Low prices High prices Mean -0.05847 -0.12629 Standard error 0.031235 0.030257 Median -0.00437 0.016502 Standard deviation 0.615262 0.643273 Variance 0.378548 0.4138 Range 5.164841 4.963525 Minimum -3.94481 -3.85496 Maximum 1.220028 1.108567 Sum -22.6874 -57.0816 Number of units 388 452 Source: Own elaboration As descriptive statistics show, in both cases rate of return is negative. However, in case of low priced stocks the loss is lower than in case of high priced stocks (respectively -0,05% versus -0,13%). The variance of both samples is similar. Despite visible differences, this relation is not statistically significant. Zero hypothesis: the difference between both means = 0 Sample 1: N=358, mean = -0,0623703, std. error = 0,610748 Residual std. error = 0,032279 95% confidence interval for the mean from -0,125851 to 0,00111063 Sample 2: N=452, mean = -0,126287, std. error = 0,643273 Residual std. error = 0,030257 95% confidence interval for the mean from -0,185749 to -0,0668244 The test Statistic: t (808) = (-0,0623703 --0,126287)/0,0445101 = 1,436 Double - sided critical area p = 0,1514 One - sided critical area = 0,0757 The zero hypothesis is positively verified. It means that statistically both means are equal. 276 4 Conclusions Studies conducted on consumer market have already verified that prices affect human decisions. In the processes of evaluation, psychological features have an important role. Studies conducted in the area of behavioural finance proved that investors do not behave rationally and made decisions influenced by psychological factors. However, the question is whether perceiving price is one of those factors that strongly influence investors decision. The face nominal effect is not verified widely on capital market conditions. On the one hand, the low price anomaly indicates that low priced stocks are attractive to buy and generate higher profits. On the other, market analytics communicate that purchase of low priced stocks is more risky and are usually issued by companies in bad financial condition. Presented study has a preliminary character. On the basis of conducted analysis it may be concluded that low priced stocks generate higher profits. It was also observed that losses generated by low priced stocks are lower in comparison to the high stock, but this cannot be statistically verified. It is necessary to verify this phenomena in further analysis. References Baker M., Greenwood, R., Wurgler, J. (2009). Catering through nominal share prices. The Journal of Finance, vol. 64(6), pp. 2559-2590. Biegahska, K., Jasiniak, M., Pastusiak, R., Pluskota, A. (2016). Efekt zakotwiczenia w transakcjach fuzji i przeje_c na przykladzie Polski. Finanse, Rynki Finansowe, Ubezpieczenia, vol. 1(79), pp. 585-593. Branch, B., Chang, K. (1990). Low price stocks and the January effect. Quarterly Journal of Business and Economics, pp. 90-118. Falkowski, A., Tyszka, T. (2006). Psychologia zachowah konsumenckich. Gdansk, GWP. Goodman, D. A., Peavy, J. W. III. (1986). The interaction of firm size and price-earnings ratio on portfolio performance. Financial Analysts Journal, vol. 42(1), pp. 9-12. Hersh, S., Statman, M. (1985). The disposition to sell winners too early and ride losers too long: Theory and evidence. The Journal of finance, vol. 40 (3), pp. 777-790. Kahneman, D., Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrics, vol. 47, pp. 267-291. Mazurek-topaciriska, M. (1997). Zachowania nabywcow jako podstawa strategii marketingowej, Wydawnictwo Akademii Ekonomicznej im. Oskara Langego we Wrodawiu, Wroclaw. Odean, T. (1999). Do investors trade too much?. The American Economic Review, vol. 89 (5), pp. 1279-1298. Samuelson, W., Zeckhauser, R. (1988). Status quo bias in decision making. Journal of risk and uncertainty, vol. 1 (1), pp. 7-59. Simon, H. A. (1979). Rational decision making in business organizations. The American economic review, pp. 493-513. Thaler, R. H., Johnson, E. J. (1990). Gambling with the house money and trying to break even: The effects of prior outcomes on risky choice. Management science, vol. 36 (6), pp. 643-660. Tykocinski, O. E., Israel, R., Pittman, T. S. (2004). Inaction inertia in the stock market. Journal of Aplieed Social Psychology, vol. 34, pp. 1116-1175. Zaleskiewicz, T. (2015). Psychologia ekonomiczna. Warszawa: PWN. Zaremba, A., Zmudziriski, R. (2014). The Low Price Effect on the Polish Market. Financial 22 Internet Quarterly „e-Finanse", vol. 10 (1), pp. 69-85. 277 Compensation for Income Lost -Long-term Effects on the Victim's Personal Finance Anna Je.drzychowska1, Ewa Poprawska2 1 Wroclaw University of Economics Faculty of Management, Computer science and Finance, Department of Insurance ul. Komandorska 118/120, 53-345 Wroclaw, Poland E-mail: ewa.poprawska@ue.wroc.pl 2 Wroclaw University of Economics Faculty of Management, Computer science and Finance, Department of Insurance ul. Komandorska 118/120, 53-345 Wroclaw, Poland E-mail: anna.jedrzychowska@ue.wroc.pl Abstract: The purpose of the article is to estimate the consequences that can arise in the household of the injured person, when they discontinue to pay social security insurance contributions because disability caused by motor accident. If the accident causes inability to work arose from someone's guilt, the offender shall bear the cost to repair the damage. If the damage is long-term, this compensation should be based on the payment of annuity, where there can be used two concepts: lifetime annuity, the value of the first payment at the level of net salary of the injured person, term life annuity paid till retirement age, the value of the first payment at the level of the gross salary of the injured person (including contributions to retirement insurance) and indexed by eg. the rate of inflation. In the second case, it is assumed that the injured person receives greater amount and spend it on pension contributions. The authors will make a comparison, which of these two methods of compensation for lost income is more favourable for victims of different age categories from three European countries: Czech Republic, Germany, Poland. Keywords: personal injuries, annuities, MTPL insurance, technical provisions JEL codes: G22, G28, J17 1 Introduction Theory of personal injuries and types of compensations After the end of the professional activity the person who retires receives a benefit, which is paid from capital which is accumulated in pension systems - voluntary, mandatory and mixed. This capital should be collected from the start of professional activity until the end, so that was a hedge on the further life of the person who finishes professional activity. Unfortunately, during the life there may be an event of accidents that cause interruption of professional activity, and thus the accumulation of capital for retirement (whether compulsory contributions, or associated with private financing decisions). If the accident causes inability to work arose from someone's guilt, the offender shall bear the cost to repair the damage. In addition, if the damage is long-term (eg. permanent incapacity for work), this compensation should be based on the payment of pension, which for many years is the compensation of lost income for the victim. In determining the amount of such annuity there are two concepts: 1. Lifetime annuity, the value of the first payment at the level of net salary of the injured person, indexed for example by the rate of inflation; 2. Term life annuity paid till retirement age, the value of the first payment at the level of the gross salary of the injured person (including contributions to retirement insurance) and indexed by eg. the rate of inflation. In the second case, it is assumed that the injured person receives greater amount and spend it on pension contributions. The authors make a comparison, which of these two methods of compensation for the loss of income, is more favourable for victims of different age categories from three 278 European countries: Czech Republic, Germany, Poland. The main criteria is the present value of annuities calculated using both methods. In the study are used data characterizing the realities of individual economies - life tables, the gross and net salary, and net the replacement rate. The first part a methodology for determining the present value of annuity benefits paid to the victim for loss of income is presented, supplemented then by simulation and comparison of the results for selected countries. The last part of the article summarizes the results and conclusions. In insurance theory there is a distinction between damage to property and personal injury. In damage to property, insurance benefit takes the form of indemnity and its size refers to material losses which occurred. In contrast, personal injury should be covered as part of the indemnity and the part constituting the compensation for pain and suffering (not in all countries provided in legal solutions, in some countries, like Germany not included in civil law, but can be found in jurisdiction). Indemnity should refer to the financial losses incurred by the victim (directly or indirectly injured), such as medical expenses, hospitalization costs, costs of care, rehabilitation costs and loss of incomes (more in Je_drzychowska et al, 2014, pp. 280-287). The determination of the amount of the damage is very important, it is the basis for calculating the amount of insurance compensation and calculating the size of adequate reserves (more in: Je_drzychowska et al, 2011 A, p. 237-262; Je_drzychowska etal 2011 B, p. 52-62). The necessity of covering personal injuries may occur due to personal insurance (health accident, life insurance) and liability insurance for the employer with respect to employee for accidents, medical liability insurance, liability insurance for the product or, the most popular, motor third party liability insurance (more in: Je_drzychowska et al, 2015, pp. 238-245; Je_drzychowska, 2015, p. 230-237). The rules on compensation of personal injuries are included in the civil law of each country. For the analyzed countries are: in Germany -art. 249-253 of Bundesgesetzblatt (BGB), in Czech Republic - Czech Civil Code, art. 2958-2967 and in Poland - Polish Civil Code, art. 444-446. Categories of compensatory benefits for personal injuries exemplified by Polish civil law, are contained in the table 1. Table 1 Types of Compensation for Personal Injuries according to Polish Civil Code Claims of directly injured Claims of indirectly injured A) Pecuniary damage (art. 444) -reimbursement of: • medical expenses, medical visits, rehabilitation, necessary prostheses, medicines, additional nutrition to enhance treatment processes, rehabilitation equipment, • costs of preparing for another profession, • costs of travel for medical appointments and consultations, travel costs of persons close to the victim, • annuity, which will be the equivalent of earnings lost as a result of that accident. B) Non-pecuniary damage (art. 445) - compensation should include all the physical and mental suffering experienced by the victim, and those that will be suffered in the future, granted at once_ Includes reimbursement of (art. 446 and 448): • medical and funeral expenses, • annuity to supplement the incomes of relatives of the lost part of the revenue paid to the household by the deceased person, • appropriate compensation in case of deterioration of the living conditions of a close relative (spouses, children, stepchildren, parents and other persons forming a common household with the direct victims), • compensation for the death of a close relative (since 2008). Source: Polish Civil Code Because annuities are the main area of interest of the authors, the rules for granting annuities will be summarized. The claim for annuity are entitled, therefore to: 279 Table 2 Cases for Which Civil Law Provides the Payment of Compensation Benefits for Personal Injuries in the Form of an Annuity Directly injured party: Indirectly injured party (persons close to the _deceased):_ supplementary annuity to compensate loss of income; in case of decreased chances of success for the future, which may be expressed in the damage, which involves the loss of other benefits to property, the injured due to their individual characteristics (eg. high qualifications, special abilities) could achieve with full functionality of the body; annuity for the increased needs (constantly recurring expenditures on maintaining a permanent treatment, rehabilitation, special nutrition, the need for care by third parties, etc.). in respect of which the deceased had a legal duty to maintain, for other people close to the deceased, to whom the deceased were permanently and voluntarily providing livelihoods, and that is required by the rules of social coexistence. Source: Polish Civil Code 2 Methodology and Data There is no clear interpretation of which calculation methodology should be used when determining the amount of the benefit to the victim. General premise states that the benefit should be the replacement of the income previously received by the injured and those which would have received it if the accident did not happen. Remains open the following issues: • the amount of the benefit, • whether include the contributions for health and social insurance, • duration of payments, • indexation, • form - periodically or at one time. In determining the amount of annuity there are two concepts: 1. Lifetime annuity - in this case, the annuity payments base is the net salary received by the victim in recent times before the accident. Whole life geometrically increasing lifetime annuity due for the first payment equal to 1 unit can be expressed as follows: Cl)-X I"x = ^(1 + 0k~1vk kpx (1) k=0 where: kpx- probability that x-year-old person will survive another x years v = ^-discounting factor, where r is the technical interest rate, i - annuity indexation rate, co - maximum age in life tables. 2. Term life annuity paid till retirement age, the value of the first payment at the level of the gross salary of the injured person (including contributions to retirement insurance). It is assumed that the injured person receives greater amount and spend it on pension contributions. This annuity can be calculated as a sum of term life geometrically increasing annuity due and m year deferred life annuity-due: 280 Term life geometrically increasing ™ annuity due for the first payment/a—, = Y(l + i)k~1vk kpx (2) equal to 1 unit: ' 1 ^—' M k=0 Lifetime geometrically increasing annuity-due, m year deferred for m = vm. . m (3) the first payment equal to 1 unit: 1 where: kpx- probability that x-year-old person will survive another x years v = ITf ~ discounting factor, where r is an interest rate, i - annuity indexation rate, co - maximum age in life tables, n - number of annuity payments, m - time to retirement age ( = retirement age-x). In this case it is assumed that the victim provide the equivalent of pension contributions to the social security system, and then, after reaching retirement age (Table 3) receives a pension from the system. Then in the first annuity, the amount of the payment is equal to the net salary earned before the victim recently before the accident, plus the pension contributions (Table 4). In the second annuity, the amount of annuity payments is the value corresponding to the retirement pension (calculated as the product of the average net replacement rates in the country and the amount of the last annuity payment just by reaching retirement age. Table 3 Retirement Age in Czech Republic, Germany and Poland (in 2016) Country_Retirement age for women_Retirement age for men Age 61 and 8 months gradually rising by four months each year Age 62 and 10 months (six months in 2019) until reaching the gradually rising by two months each retirement age for men, thereafter, by two year with no upper limit months each year with no upper limit_ Czech Republic Germany Age 65 and 5 months (gradually rising to age 67 by one month each year until _2024 and by two months until 2029)_ Poland Age 60 and 11 months Age 65 and 10 months gradually rising by one month in January, gradually rising by one month in May, and September each year until January, May, and September each _reaching age 67 in 2040_year until reaching age 67 in 2020 Source: Social Security Programs Throughout the World: Europe, 2014, https://www.ssa.gov In the second approach, of course the present value of the two annuities represents the total present value of benefits obtained by the beneficiary of annuity (directly or indirectly injured). From the insurance company point of view, the costs connected with this annuity consists only on the first part (term life annuity). The second part is paid from the social security system. The next problem is connected with the amount of annuity payments. Is considered that the victim should receive their current salary, therefore, the basis for calculating the monthly benefit may constitute in case of: • employed person - their current salary, or loss of salary in case of loss of its parts, • person working on the basis of civil law agreements, self-employed, running their own business or agricultural holding - their average monthly income from this activity (eg. the average income from the last year), • persons without income (eg. unemployed, without qualification) - minimum wage, • minors, studying - the average wage in the economy. 281 In case of unemployed people, it seems reasonable to use a minimum wage in the economy for the long term unemployed but for ones with short-term break in employment - the average salary, eg. in the last period of employment. As for minors and learners which have not yet taken employment, it is not possible to determine even the approximate size of the income which they would achieve. It cannot be also assumed that these people would receive income at the lowest level, so it seems reasonable to use the average wage in the economy. In the situation of indirectly injured the share of personal consumption in the salary should reduce the amount of benefit. Table 4 Net and Gross Average Monthly Salaries, Pension Contributions and Net Replacement Rates in Czech Republic, Germany and Poland in 2014 _ . Average monthly _ _ . .. .. Gross replacement Countrv salaries [EUR] Pension Contributions_rates Net Gross % of gross salaries Monthly in EUR Women Men Czech Republic 726 945 28% 264,46 64% 64% Germany 2 315 3 829 19% 723,74 50% 50% Poland 635 844 19,52% 164,68 53% 53% Source: Eurostat and National Statistics Offices, and WHO data In the calculations provided in the article, only one case is analysed - annuity payments are equal to the average wage in the economy. Taking consideration the way of indexing - an increase in benefits, it should be noted that indexation should maintain the real value of pension benefits as well as reflect the future potential growth in incomes. The benefit should therefore be indexed by the inflation rate, and it must be also included that the income the victim would vary, in particular in increase of its real value. It should therefore be included the real wage growth in the economy. The inflation rate was equal to 1% (value close to the EU average). Similarly discounting interest rate was based on the EU average rate of return on 10 year treasury bonds (r=2,5%). The calculations has been provided for annuity payments paid monthly, so as to calculate probability of surviving the probability of survival part of the year, the assumption of Uniform Distribution of Deaths (UDD) at each year were used. 3 Results and discussion The resulting amounts (present values of annuities) are not amounts that the victim spent directly on consumption. In both versions of annuities for direct consumption, the victim receives only the net salary. In the second variant the present value of annuity represents not only to continuously consumed funds in the amount of net wages, but also the pension contributions, which should be provided to the system. Thus, the PV annuities must be understood as the financial resources allocated to the victim for direct consumption as well as on account of his/her participation in the social insurance system. Present value of analysed annuities for different age categories are presented on Figures 1-3. The results indicate that one should take into account the age of the victim when choosing a method of determining the amount of annuity payment, and the duration of its payments. For young people it is preferable to obtain a benefit increased by the pension contributions and after a retirement age to receive benefit from the social security system. For the elderly it is more advantageous to receive an annuity of net wage. The decision to choose a variant of annuity also depends on gender, women slightly later should choose a variant of a lifetime annuity. Due to the lowest replacement rate, the decision about choosing whole life annuity should be taken at the earliest for the citizens of Poland and Germany. For women in Poland whole life annuity is preferred over the age of 39,5, and for men 46,5 (Figure 2). In Germany women over the age of 41 should decide to lifetime annuity, while men over 45,5 (Figure 3). At the latest among the countries analysed, the decision to choose 282 lifetime annuity should take women in the Czech Republic (at age of 49), and men at the age of 52 (Figure 1). Figure 1 Present Value of Annuities for Czech Republic in EUR (Women - Left Chart, Men - Right Chart) Depending on Age 9CC0CC BCCOCC 7CC0CC GCCOCC 5CC0CC ^CCOCC 300 COO 200 coo ico coo \ \ t \ ^_ _ _>_ ^ age 4- 23 40 GO 80 IOC -5= 900 000 accccc 7CCCCC sec coo <^cooc 300 OCC 200 000 100 000 ; ^ 1 ' age I 2 • PVof lifetime annuity ■: so -; — PVof IrfEtime annuity ice = = = — — — PVo-f life annuity till retirement 3g a, than pens ic-n fr-txn the social security system Source: Author's own calculations — — — PVoflifE an nuity til I retire ment agE, than pension from t he sad al sec urity syst em Figure 2 Present Value of Annuities for Poland in EUR (Women - Left Chart, Men - Right Chart) Depending on Age — — — PV of life an nuity til I retirement age, than pens kin from the — — — PV-af life an nuity til I retirement age, than pens kin from social security system the social security syst Em Source: Author's own calculations 283 Figure 3 Present Value of Annuities for Germany in EUR (Women Right Chart) Depending on Age Left Chart, Men ; x;; x —--PV of I If e annuity till mtim me nt agE, than pmsion from —--PV of I if e annuity til I retirement agE, than pension from t he 5ad al sec urity 5y5t Em t he 5ad al sec urity 5y5t Em Source: Author's own calculations Net replacement rates used for calculation are calculated for people retiring in 2014. In the simulation should also be included the predicted replacement rates, the ones in the analysed countries are falling. Table 5 The Age at Which Changes the Profitability of Variant of the Method of Calculation PV of Annuities Simulations for Different Values of Replacement Rates net replacement rates 30% 35% 40% 45% 50% 55% 60% Czech Republic women 37 39,5 40 43 45 46 48 men 43,5 44,5 45,5 47 48 49,5 50,5 Poland women 26 28,5 31,5 34,5 37,5 40,5 43 men 37 39,5 41,5 43,5 45,5 47,5 49,5 Germany women 30,5 33 35,5 38 41 43,5 46 men 37 39 41,5 43,5 45,5 47,5 50 Source: Author's own calculations Table 5 shows how much moves to the moment of change decisions about which of the methods of calculating the present value of annuity, gives the injured party more funds. For example, for the lowest replacement rate in Poland should a woman decide to change the variant of the calculation annuity at age 26, while for the highest of the analyzed replacement rates, changing decision should be made 17 years later. Likewise a large discrepancy (17 years) for men can be seen in Germany. For Czech citizens changing the replacement rate is the least important. 4 Conclusions As indicated by the results of a study on the value of the of annuity received has considerable influence its model. From the obtained results it can be concluded that for younger people (up to about 40 - 50 years of age), the higher present value of annuity are determined based on the 2 variant (annuity term with the pension provided under the social security system). It should be remembered that the moment of a decision about which option is more cost-effective, depends on the size of the replacement rate adopted for calculation. It should be emphasized that currently the courts and insurance 284 companies do not use such a method of calculating annuities for victims, and as a basis for of annuity recognize the net salary of the victim. It should be noted that the two compared solutions carry risks which victims must be aware of. Variant of whole life annuity is strongly related to the situation on the capital market. Thus, the risk associated with it is mainly the risk that the insurance company will remain in this market and will continue to be solvent, to pay annuity to the victim. The second variant also is affected by risks, especially important here is political risk. An example of big changes introduced in 2013 is to raise women's retirement age of 7 years, men of two years (subsequent changes, this time on lowering the retirement age, are announced by the current Polish government). Similar reforms were carried out in other countries due to demographic changes. Also, the demographic situation is the risk that affects the annuity established in the second variant. References Czech Civil Code - Law No. 89/2012 Coll. Civil Code. Retrieved from: http://www.czechlegislation.com/en/89-2012-sb. Eurostat. Retrieved from: http://epp.eurostat.ec.europa.eu. German Civil Code - Civil Code in the version promulgated on 2 January 2002 (Federal Law Gazette [Bundesgesetzblatt] I page 42, 2909; 2003 I page 738), last amended by Article 4 para. 5 of the Act of 1 October 2013 (Federal Law Gazette I page 3719). Retrieved from: http://www.gesetze-im-internet.de/engIisch_bgb/. Je_drzychowska A, Poprawska E. (2011). Szkody osobowe z ubezpieczenia komunikacyjnego OC a gospodarka finansowa ubezpieczycieli. Szkody osobowe kompensowane z ubezpieczenia komunikacyjnego OC. Analiza rynku / Kwiecieň Ilona {red.), 2011, Wydawnictwo Poltext, ISBN 978-83-7561-197-7, pp. 237-262. Je_drzychowska A., (2015). Size of compensation for personal injuries offered by the current coverage plans in Medical Liability Insurance. European Financial Systems 2015. Proceedings of the 12th International Scientific Conference, Masaryk University, ISBN 978-80-210-7962-5, pp. 230-237. Je_drzychowska A., Kwiecieň I., (2015). Classification of work-related accidents as the basis of analysis of employers liability risk and insurance decisions. European Financial Systems 2015. Proceedings of the 12th International Scientific Conference, Masaryk University, ISBN 978-80-210-7962-5, pp. 238-245. Je_drzychowska A., Poprawska E. (2011). Ubezpieczenia komunikacyjne i ich wpfyw na gospodarke. finansowa. zakfadów ubezpieczeň. Prače Naukowe Uniwersytetu Ekonomicznego we Wrociawiu, Uniwersytet Ekonomiczny we Wrodawiu, nr 175/2011, pp. 52-62. Je_drzychowska A., Poprawska E., (2014). Disability Benefits for Victims of Traffic Accidents - Size of Compensations Offered by the Current Amount of Coverage in MTPL Insurance. European Financial Systems 2014. Proceedings of the 11th International Scientific Conference / Deev Oleg, Kajudochodrová Veronika, Krajíček Jan {red.), Masaryk University, ISBN 978-80-210-7153-7, pp. 280-287. Polish Civil Code - Ustawa Kodeks Cywilny, Dz.U. 1964 nr 16 poz. 93, Retrieved from: isap.sejm.gov.pl. Social Security Programs Throughout the World: Europe, 2014. Retrieved from: https://www.ssa.gov. 285 Crop Insurance as the Instrument for Risk Financing in Polish Farms Monika Kaczata Poznaň Univeristy of Economics and Business Department of Insurance Al. Niepodleglosci 10, 61-875 Poznaň, Poland E-mail: m.kaczala@ue.poznan.pl Abstract: Risk management is one of the main priorities in the Common Agricultural Policy (CAP) for 2014-2020. Crop insurance belongs to the three CAP indicated types of instruments for financing risks, along with mutual funds and income stabilization tool. In Poland crop insurance is the only available instrument and despite its obligatoriness and subsidizing the market penetration amounts to 40%. The aim of this paper is to present development of agricultural insurance in quantitative terms and its main contributory factors. The answers were provided based on the qualitative and quantitative analysis. The data come mainly from the official statistics and information provided by The Ministry of Agriculture and Rural Development. The results show that apart from very high costs of crop insurance, resulting from a long-term loss ratio of more than 100% for, a very important determinant affecting crop insurance purchasing decisions is the subjective assessment of prosperity in agriculture as well as the scope and type of state aid (investment-related benefits are stimulating while revolving loans are inhibitive). Keywords: agricultural insurance, crop insurance, state aid, risk management, financing of risk JEL codes: Q14, G22, H84 1 Introduction In Poland, agriculture plays an essential role, both socially and economically. It accounts for roughly 3 % of the GDP, similarly to other highly developed countries and provides employment for approx. 16 % of the population, which is a relatively high level (CSO, 2015a). One of the major functions of numerous state interventions is to foster agriculture. The EU Common Agricultural Policy limits the scope of these interventions while emphasizing the need for risk micro-management in farms. As early as the 1950s, in Polish farms crop and property insurance was obligatory and therefore very common. However, in 1990 crop insurance became voluntary and the terms and conditions of insurance policies deteriorated, parallel to a rise in prices. This resulted in a dramatic decline in the number of insurance contracts. It was only in 2005 that the act Act on Agricultural Crops and Farm Animals Insurance (ACAIA) was passed and has been amended several times since then. This is the reason for a renewed interest in crop insurance. The purpose of this act was, among other things, to reduce the subsidies derived from the governmental and other public financial institutions when it came to natural disasters. Also, it was aimed at disseminating crop insurance, particularly in the context of climate change (Sejm of the Republic of Poland, 2005). Today it is estimated that despite obligatory crop insurance only as few as 40 % of crops are covered by insurance, which is by far below the expected level. Therefore, the aim of this paper is to present development of agricultural insurance in quantitative terms and its main contributory factors. 2 Methodology and Data The answers were provided based on the qualitative and quantitative analysis. A relatively short time for which the subsidised insurance cover has been available prevents complex statistical analysis. The study was conducted with reference to correlations between annual changes in the number and value of the signed contracts, the level of the written premium, the size of insured acreage and the annual changes in 286 the prosperity indicators used for agriculture in Poland, the value of compensation for particular types of risk, the number and value of contracts signed within the scheme of "restoration of agricultural production potential damaged by natural disasters and catastrophic events and introduction of appropriate prevention measures", the number and value of loans allocated for resumption of agricultural production in farms and special sectors of agricultural activity (investment and revolving loans). Detrendisation, in turn, increases credibility of the analysis as it makes it possible to assess whether the correlations are only ostensible (such ostensible correlations may result from trends occurring in time series which are not connected by cause and effect). For conclusions, a 95% confidence level was assumed. The data come from the official statistics and information provided by The Ministry of Agriculture and Rural Development. Tests were performed in Statistica 10PL and in Excel 2013. Crop insurance in Poland on 2006-2015 Subsidised crop and livestock insurance was introduced as a response to the Commission Directive (EC) No 1857/2005 of Dec 15 2006 regarding application of the Treaty articles 87 and 88 concerning state subsidies for SME farms and which changed the directive (EC) No 70/2001. Sales of these policies were launched in the autumn of 2006, and the system has been modified several times since then. Initially, the Act covered only the most popular crops in Poland: cereals, corn, colza, agrimony, potatoes and sugar beets. As of 2007 subsidies started covering a wider range of crops, which currently results in an almost complete cover including fruits and vegetables. Except for the initial period, subsidized insurance covered the following perils: hurricane, flood, excessive rain, hail, lightning, landslide, avalanche, drought, winterkill and spring frost. At first, all these perils were covered collectively as a single package. It was only in April 2007 that it was made possible to split the insured perils. Mid-2008 saw the obligation to insure at least 50 % of the arable land against at least one peril. In April 2007 the scope of subsidized insurance was restricted to farms which did not exceed the area of 300 ha. In the case of bigger farms it was possible to insure the surplus area within commercial insurance or resign from insurance altogether. This restriction was removed in August 2008. Additionally, only SME farms were eligible for the subsidy, according the EU regulations. Consequently, all the large farms owned by the Agricultural Property Agency or belonging to capital groups were excluded from the programme. In 2015 the above restriction was removed and large farms were also admitted into the system (after meeting certain criteria). Initially, the level of the subsidy amounted to 35 or 40 % of the total premium paid to the insurance company. In 2008 it was increased to 50 %, and very recently, in 2016, to 65 % of the premium. The subsidy was only possible if the insurance company-generated tariff rates did not exceed 3.5 or 5 % of the insurance cover but not more than 6 % of the insurance cover. When the subsidised insurance system was being introduced, it was assumed that the minimum of the insured land should amount to 7m hectares of crops. In fact, the penetration level (understood as the quotient of the actually insured area to the assumed area of 7 m hectares) did not exceed 50 % (table 1). With reference to the crops of 2014, only 25% of cereals and corn were insured, 86% of rapeseed, 31% of sugar beets, 10% of ground vegetables and 2% of fruit (Rojewski, 2015). After an initial surge in the number of contracts, which was primarily caused by the introduction of mandatory insurance of at least half of the acreage, now the market has stagnated for approx. two years. The trend also refers to the average amount of insurance, whose former growth resulted both from the increase in the insurance rate and the raising of the maximum amount of insurance. Additionally, popularity of fruit and vegetable insurance (especially against hail) could not be overlooked. During this time, the structure of the risks insured changed as well (table 2). It can be seen most clearly after 2008 when the obligation to insure at least half of the acreage was introduced. To this day, this duty has been usually fulfilled in the form of an insurance contract against hail, which is the cheapest option. The second most popular choice is an insurance contract against three perils, i.e. hail, spring frost and winterkill. 287 Table 1 Subsided Crop Insurance in Poland in the Years 2006-2014 - Basic Data Insured Penetra- Value of all The Average Average area tion contracts number of sum area (in ha) level signed contracts insured insured in (in %) (in Tsd PLN) signed per 1 ha one (in PLN) contract (in ha) 2006 311 740 4.5 N.A. 10 738 N.A. 29.0 2007 575 029 8.2 N.A. 28 412 N.A. 20.2 2008 1 832 036 26.2 N.A. 87 150 N.A. 21.0 2009 2 808 104 40.1 6 490 380 144 080 2 311 19.5 2010 2 845 777 40.7 7 843 806 134 986 2 756 21.1 2011 3 032 634 43.3 10 238 599 138 425 3 376 21.9 2012 2 751 438 39.3 12 087 100 135 707 4 393 20.3 2013 3 398 811 48.6 14 232 425 151 101 4 187 22.5 2014 3 269 871 46.7 13 326 951 142 492 4 076 22.9 Source: Author's own calculation based on (CSO, 2015b; Sejm of the Republic of Poland, 2015; Prime Minister, 2016). Table 2 The Structure of Risks Covered by Subsided Crop Insurance in 2006-2015 (in %) 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 drought 9.1 9.0 4.5 3.0 0.4 0.2 0.1 0.0 0.0 0.0 flood 9T1 9T0 4T2 03 03 0T2 0T1 0T1 0T1 0T0" winterkill 1CL9 TO 2T8 14?7 I5T0 16^8 18^6 187l I7T2 27.3 spring 10.3 113 2L9 223 21J 21J 22^9 203 227l 28^9 frost_ hurricane 9.1 9.2 4.2 0.4 1.4 1.5 1.9 3.1 3.7 1.2 excessive 9Tl 9^2 4T2 0~4 1A 13 T9 37l 3?7 1/2 rain_ hail 15.3 14.1 26.9 57.3 55.5 53.4 50.2 50.1 48.1 39.7 lightning 9Tl 9^0 47l 0~4 1A 13 1A 13 1J 03~ landslide 9Tl 9^0 4Tl OA 1A 13 1A 13 1J 03~ avalanche 9.1 9^0 47l 0^4 1A 13 1A 13 1J 03~ Source: Author's compilation according to data from (Sejm of the Republic of Poland, 2015; Prime Minister, 2016). 3 Results and Discussion Factors affecting the use of subsidized crop insurance The reasons for lower than expected sales of subsidized crop insurance come from both the supply and demand on the market. For insurance companies the cost of crop insurance is very high because of accumulation of risks, complex loss settlement, large number of disputes and the implications of information asymmetry. Between 2006 and 2015 the total value of the paid compensation amounted to PLN 2.161m and the amount of the collected premiums - PLN 2.293m (Rojewski, 2012; Jane, 2016). Due to winterkill, it is estimated that in 2016 the compensations will amount to approx. PLN 600m and the premiums collected - PLN 320m (Jane, 2016). This means that in the long term, subsidised crop insurance is not profitable and in the years 2006-2015 the loss ratio stood well above 100 % three times (in 2008 it was 121% because of drought, in 2011 - 288 134% as a result of winterkill and spring frost, while in 2012 as much as 252% due to winterkill). In 2016 it is estimated for the loss ratio to be approx. 190% (Jane, 2016). As a response to the level of compensation due to drought which was particularly high in 2008 a special purpose subsidy was launched in mid-2008 for insurance companies which offer insurance against drought (subsidized or not). It functions like an excess claims agreement. Despite this, the number of contracts including the risk of drought (and also flood) has fallen down nearly to zero. It is worth mentioning that government reassurance is only offered for the risk of drought, although other types of risk, such as winterkill cause a high level of damage (table 3). Correlation indicators between financial means allocated for insurance premium subsidies and the total number and value of the signed contracts for crop insurance are irrelevant. It happens because within the whole period considered the utilisation level of the planned subsidies never exceeded 90%, and what is more, in the first three years of its functioning it was never higher than 14-16%.Because of the aforementioned difficulties crop insurance is offered only by three domestic insurance companies, i.e. about 10% of the companies which are based in Poland. Table 3 The Structure of Subsidized Crop Insurance Compensation in 2006-2015 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Total Drought 0 1 82 1 1 1 0 0 0 0 8 Flood 7 0 0 1 4 1 0 1 0 0 1 Winterkill 31 3 0 2 32 44 82 9 4 5 38 Spring frost 1 64 2 27 8 38 1 2 50 17 19 Hail and others 61 32 16 68 54 17 17 88 45 78 35 Total 100 100 100 100 100 100 100 100 100 100 100 Source: Author's compilation according to data from (Prime Minister, 2016; the Ministry of Agriculture and Rural Development). From the point of view of a farmer, the available insurance cover is limited in scope while loss settlement and assessment stages (especially total loss and deduction of the so called avoided costs) are questionable. Availability of insurance against some occurrences (drought, flood) or crops (some vegetables and fruit) is limited due to the insurance rate which prevents access to the subsidy. Research on the factors affecting the probability of using crop insurance by individual farmers indicate that the most important factors are: location and size of the farm, rape cultivation, having very good or good type of soil, previous use of crop insurance, historical record of losses connected with flooding or hail, perception of risk connected with losses resulting from winterkill (Kaczate and Wisniewska, 2015b). If the farmer had experienced a loss and more or less often had received compensation (i.e. it was at least partially funded by indemnity) this led to a significant growth in probability of signing a crop insurance contract. Probability of signing an insurance deal is twice as much linked with financing the loss by indemnity than solely with the occurrence of a given phenomenon (e.g. hail or flood) (Kaczala and Wisniewska, 2015b). To corroborate the findings of research conducted on a micro level, one should mention the relationship between the total number and value of crop insurance contracts and the level of compensations in the previous year. The correlation coefficient between annual changes in the number of contracts and the value of compensations for drought in the previous period amounts to 0.83 (p-value 0.02). On the other hand, the annual changes in the value of the signed crop insurance contracts are correlated with annual changes in the level of compensation for all occurrences at the level of 0.87 (p-value=0.056), including the changes within compensations for winterkill at 0.9 (p-value=0.04). The indicated correlations are therefore very strong. The relations between prosperity in agriculture and demand for crop insurance were examined by analyzing the correlation between annual changes in the number and value 289 of the insurance contracts, the level of written premium and the size of the insured acreage and the change in the average general prosperity index (GPI) as well as the synthetic prosperity index for agriculture (SPIA). The general prosperity index is estimated as an arithmetical mean of the monetary revenue change rate and confidence index. The monetary revenue rate is calculated on the basis of answers to questions about achieved and predicted revenue of a given farm from two most recent questionnaires. The share of this rate within the GPI is almost twice as large as the confidence index share. Confidence index, in turn, is calculated on the basis of the farmers' responses to the question about their feelings concerning prospects of further farming operations (more: Grzelak and Seremak-Bulge, 2014). SPIA, in turn, is a quantitative index, which synthetically illustrates the changes in market-related factors of agricultural production. It is calculated as an arithmetical mean of the margin squeeze index and the potential demand index. The SPIA encompasses complex conditions relating to supply and demand as well as price related factors of the agricultural market which are evened out with a 6-month moving average (Grzelak and Seremak-Bulge, 2014). No connections were detected between the value or annual change in SPIA and the number and value of subsidized crop insurance contracts, the amount of collected premium or the acreage insured. However, such connections can be seen in the case of the GPI. The annual change in the number of signed contracts shows a strong correlation with the annual change in the GPI (r=-0.82, p-value=0.012), while the acreage insured is correlated on a medium level (r=-0.68, whereas p-value=0.062 so it is slightly higher than the assumed maximum error level). Hence, it is the subjective, not objective, evaluation of prosperity by farm managers that is strongly correlated with positive or negative decisions concerning the purchase of crop insurance. GPI is created in 66% by partial revenue ratio. Considering the r-Pearson correlation coefficient's plus/minus sign, one can arrive at a conclusion that it is the rise in revenue (both achieved and expected) that is linked to decreasing number of signed contracts. This shows coherence with declarations of farm managers who claim that losses in crops are financed in 70% from their own resources (Kaczala and Wisniewska, 2015b). The element which must be taken into account when considering quantitative changes in crop insurance is the access to other sources of loss financing than compensation or farmers' own financial means. In Poland these mainly amount to bank loans and various forms of financial support from the state (Kaczate and Wisniewska, 2015b). In the case of losses in crops, the access to so called "disaster loans" is one of the many forms of government aid intended for agriculture, followed by discounts payment in instalments or remittances on such liabilities as health or social insurance, or, finally, payments due for the Agricultural Property Agency (e.g. for the leased land). In the period discussed here the support for farmers took on mainly two forms: within the Rural Development Programme for 2007-2013 which was aimed at restoring agricultural production potential damaged by natural disasters and catastrophic events and introduction of appropriate prevention measures (a.k.a. action 126), as well as subsidizing loans for resuming production in farms and special sectors of agricultural production. Action 126 co-financed in 75% from the 2nd pillar of CAP using the means of the European Agricultural Fund for Rural Development (EAFRD) was launched in September 2010 because of intense precipitation (including hail) which led to numerous floods and landslides. 2011 saw an increase in the number of listed occurrences within the discussed action which made farmers eligible for state aid: hurricane, excessive rain, hail, lightning, avalanche, drought, winterkill and spring frost. The scope of these occurrences is identical as the scope of the subsidised crop insurance. Aid was offered in the form of partial reimbursement of the eligible costs incurred for operations exclusively encompassing investments concerning agricultural activity. One of the conditions for obtaining the subsidies was that the level of loss in crops, livestock or fish amounted to 30% of the average annual production. By the end of 2014 there had been seven application recruitments. The number and value of the signed contracts is presented in fig. 1. 290 Figure 1 The Number and Value (in PLN) of Contracts Signed under the Restoration of Agricultural Production Potential Damaged by Natural Disasters as Part of Rural Development Programme for 2007-2013 300 000 000 ■ -2 500 Ji 250 000 000 . I -2 000 _ 200 000 000 150 000 000 ° - c u 100 000 000 1 500 1000 j v 50 000 000 " " 500 2010 2011 2012 2013 2014 I The value of contracts signed The number of contracts signed Source: ARMA, 2014 On the other hand, an interest subsidy to the so called "disaster loans" (loans for restoring production in farms and special sectors of agricultural production) was offered within the state subsidy framework throughout the whole discussed period (table 4). Subsidies encompassed both investment loans and short-term revolving loans for restoring production in farms and special sectors of agricultural production where damage occurred because of drought, hail, excessive rain, winterkill, spring frost, flood, hurricane, lightning, landslide or an avalanche. Additionally, farmers could postpone the repayment of principal instalments, could be offered a grace period in the loan repayment and there was no required contribution from the borrower. In September 2013 a possibility was created for farmers to apply for a disaster loan if the level of losses in the farm or a special sector of agricultural production caused by flood, hurricane, hail or excessive rain did not exceed 30 % of the average annual agricultural yield. In this case an interest subsidy was offered as "de minimis" type of assistance in agriculture and fishery. Table 4 Number and Value of the Loans Offered to Resume Farming Production and Special Sectors of Agricultural Production Including ARMA's Interest Subsidy in 2006 - 2014 Investment loan Working capital loan Number Value (in PLN) Number Value (in PLN) 2006 103 1 870 950 136 164 1 772 336 960 2007 167 3 298 220 84 375 1 164 696 800 2008 162 3 072 260 39 903 1 043 695 170 2009 36 855 240 40 581 792 567 310 2010 29 2 290 430 8 512 244 074 730 2011 42 2 179 660 18 957 677 510 030 2012 60 7 783 390 8 241 456 915 780 2013 32 1 623 590 4 945 266 335 200 2014 16 627 460 5 198 243 373 720 Source: Author's compilation according to data from (ARMA, 2007-2015). There is a very strong correlation between the increase in the value of the contracts signed within the framework of action 126 (restoration of agricultural production potential damaged by natural disasters and catastrophic events and introduction of appropriate prevention measures) and the change in the scope of the acreage insured in 291 the following year (r-Pearson=0.94, p-value=0.017). This correlation is also strong with regard to the change in the number of insurance contracts signed in the following year (r-Pearson=0.86 and p-value=0.060, which is slightly higher than the assumed boundary). This means that the decision about insurance is significantly affected by the fact itself (not the value) of being granted a subsidy for restoration of production potential. It may result from the obligation to insure the restored orchards and perennial plantations within five years of receiving the subsidy. At the same time it must be remembered that there is a possibility to be exempt from this obligation and the resources allocated to action 126 may be and often are granted for investments of a different kind (buildings, plant etc.). As regards "disaster loans", a correlation can be seen between the annual change in the number of granted revolving loans and the change in the number of signed crop insurance contracts as well as the acreage insured in the following period of time. Correlation coefficients amount to -0.34 (p-value=0.005) and -0.57 (p-value=0.005), respectively. The reversed dependency probably is down to competitive relation between crop insurance and revolving disaster loan. The loan is intended for purchase of tangible working assets used for production in order to restore productivity, e.g. qualified seeds, mineral fertilisers, plant protection products, fuel for agricultural purposes etc. The amount of interest subsidies to the revolving loan is limited, among other things, by the level of indemnity paid for insurance contract on the risk of losses caused by adverse weather phenomena. On the other hand, it is interesting that the granting of a revolving loan in the previous period of time, not the loan's value, had a definite effect on the decision to sign an insurance contract and its value. Low importance of the value and number of investment loans probably results from the fact that their number and value in the discussed period of time was rather low. Since 2010 the interest on the revolving loan has depended on whether at least half of the farmer's land has been insured. The level of fulfilment of this obligation is most likely to be low enough to exert little if any influence on stimulating relation between these instruments (there is no correlation between the number and value of granted revolving or investment loans and the number of signed crop insurance contracts as well as the acreage insured in the same year). 4 Conclusions The data show that both the amount and value of crop insurance contracts as well as the size of the insured acreage has been stabilizing. However, the level of the stabilization is much lower than expected. A rapid growth was seen in the year when the premium subsidy was increased and it became possible to insure crops against single risks. Due to the increase in subsidy in 2016 and a broadened subject-matter of subsidized insurance, another upturn can be expected. Nevertheless, additional solutions are indispensable as the loss ratio is over 100%. The structure of the risks insured implies that crops are hardly ever insured against drought and flood. Considering that these phenomena are rather commonplace in Poland, steps taken in order to introduce index-based crop insurance against drought should be evaluated as very positive (Kaczate and Wisniewska, 2015a; Kaczate and Lyskawa, 2012). A very strong relationship has been noticed between the annual change in the number of the signed contracts and the annual change in the general prosperity index (r=-0.82. p-value = 0.012). This is an endogenous indicator and it is based on farm managers' subjective assessments. This means that it is precisely the subjective opinion that is highly correlated with making a positive or negative decision about crop insurance. Taking into account the structure of this factor, one can suppose that the cause of falling numbers of insurance contracts is connected with the growth in income (both earned and predicted). This is consistent with the findings of the research on insurance purchases in farms o nmicro level (Kaczate and Wisniewska, 2015b). Another very strong correlation has been noticed between the annual changes in the acreage insured and state aid for agriculture. The granting of the subsidy (not its value) which is aimed at restoration of the production potential and is investment-related has a huge impact on decisions about buying crop insurance. It may result from the obligation 292 to insure restored orchards and perennial plantations within five years of the subsidy reception. At the same time, one should remember that this is just one of the aims of granting such subsidies. As for the so called "disaster loans", one can clearly see a connection between annual changes in the number of granted revolving loans and the number of signed crop insurance contracts and the acreage insured in the subsequent period of time. A negative correlation suggests competitiveness of the tools discussed here. Since the intended effect is to increase the level of crop insurance penetration, a decline in granting preferential revolving loans should be perceived as positive. References ACAIA. The Act on Agricultural Crops and Farm Animals Insurance of 7 July 2005 (with further amendments) consolidated text of 6 June 2016 (J.L. of 2016 item 792). ARMA (2007-2015). Sprawozdanie z dzialalnosci agencji restrukturyzacji i modernizacji rolnictwa. Department of Analysis and Development. CSO (2015a). Statistical Yearbook of the Republic of Poland. Central Statistical Office, Warsaw: Statistical Publishing Establishment. CSO (2015b). Statistical Yearbook of Agriculture. Central Statistical Office, Warsaw: Statistical Publishing Establishment. FRD. The Act on Fostering Rural Development with the Funds of the European Agricultural Fund for Rural Development of 7 March 2007 (J.L. of 2013 item 173 with further amendments). Grzelak. A., Seremak-Bulge, J. (2014). The Comparison of Selected Methods for Testing Business Outlook in Agriculture in Poland. Problems of Agricultural Economics, vol. 4, pp. 117-130. Jane, A. (2016). Challenges in functioning of crop insurance in Poland in 2006-2016. In: The 10th International Conference "Insurance and the Challenges of the 21st Century", 16-18 May, Rydzyna, Poland. Kaczala. M., Lyskawa, K. (2012). The Concept of Index Policies and their Possible Application in the System of Compulsory Subsidised Crop Insurance in Poland. In: International Conference "Trends in Agricultural Insurance in Europe. Drought Insurance in Poland", 5 November, Warsaw, Poland, pp. 64-78. Kaczala. M., Wisniewska, D. (2015a). Factors Influencing Farmers' Decisions on Drought Index Insurance in Poland. In: The Third World Risk and Insurance Economics Congress (WRIEC), 2-6 August 2015, Munich. Kaczala. M., Wisniewska, D. (2015b). Risks in Farms in Poland and their Financing -Research Findings. Research Papers of Wroclaw University of Economics, vol. 381, pp. 98-114. Prime Minister (2016). Draft Law Amending the Act on Subsidies to Insurance of Agricultural Crops and Farm Animals with Draft Implementing Legislation. Paper No. RM-10-19-16. 25 April 2016. Rojewski. K. (2015). Polish system of agricultural insurance. In: IFC Conference, 9-11 June, Kyiv, Ukraine. Rojewski, K. (2012). The History and Current State of Agricultural Insurance in Poland. In: International Conference "Trends in Agricultural Insurance in Europe. Drought Insurance in Poland", 5 November, Warsaw, Poland, pp. 17-28. Sejm of the Republic of Poland (2015). Justification to Draft Law Amending the Act on Subsidies to Insurance of Agricultural Crops and Farm Animals. Paper No. 3247, 12 March 2015. Sejm of the Republic of Poland (2005). Justification to Draft Law of Act on Subsidies to Insurance of Agricultural Crops and Farm Animals. Paper No. 408, 17 May 2005. 293 The Influence of a Low Interest Rate on Life Insurance Companies Silvie Kafková1, Dagmar Linnertová2 1 Masaryk University Faculty of Economics and Administration, Department of Finance Lipová 41a, 602 00 Brno, Czech Republic E-mail: 175424@mail.muni.cz 2 Masaryk University Faculty of Economics and Administration, Department of Finance Lipová 41a, 602 00 Brno, Czech Republic E-mail: dagmar.linnertova@mail.muni.cz Abstract: A protracted period of low interest rate threatens the stability of the life insurance industry, especially in the countries, where life insurance with comparatively high guaranteed returns disposed in the past represents a major portion of the total portfolio. The aim of this paper is to provide an introduction to evaluation of the effects of the actual low interest rate period on the balance sheet of a representative life insurance company. Firstly the balance sheet of insurance companies is introduced. We show the valuation of the assets and liabilities. Then the Solvency Capital Requirement and risk margin is defined. Finally the solvency situation of insurers is valuated. Keywords: life insurance, low interest rates, guaranteed returns JEL codes: G22, G17, E47 1 Introduction The present low interest rates cause problems to life insurers. They must pay guaranteed rates of return to their clients and keep strong profitability in the long period. The minimum return sets at the beginning of the contract cannot be changed in the course of the contract. The result of this product attribute is the contemporary occurrence of contracts with various minimum returns in the portfolio. Moreover, continuously low interest rates would be harmful for the solvency position of subgroup of insurance companies. According to Holsboer (2000) the duration incompatibility between the asset and the liability side have main impact. When the predominant interest rates are significantly lower than they were in the beginning of the contracts, current value of liabilities becomes more expensive to finance. Berdin and Grundl (2015) investigate the effect of prolonged period of low interest rates on the solvency situation of the average German life insurer. The insurance companies in Germany face not only an expensive stock of products sold with a high minimum guaranteed return, but they also face poor investing results because of extraordinarily low interest rates, especially for German sovereign bonds. The aim of our paper is to evaluate the influence of prolonged period of low interest rates on the solvency situation of life insurance companies. We describe a balance sheet model that includes several features affecting the solvency situation. 2 Methodology and Data We distinguish the book value and market value balance sheet of average life insurance company. The book value enables us to identify the yearly profit of the insurer and the final payout of the cohort maturing at time t. However, Solvency II regulation requires market valuation of asset and liabilities to give market-consistent evaluation of the solvency situation. In the Figure 1 we could see the book value balance sheet and the market value balance sheet of the insurance company at time t. Af and LBtv denote book value of assets and 294 liabilities respectively. Et indicates the equity capital endowment and CBt is a capital buffer. OFt denotes the market value of the own funds, which represents the deviation between the market valuation of assets and liabilities. RMt is risk margin and LBtE is the best estimate of liabilities. Figure 1 Book Value and Market Value Balance Sheet at Time t Book Value Market Value Assets Liabilities Assets Liabilities OF±_ LBtE Source: Berdin and Gründl, 2015 CBt The Assets According to the Data Series System ARAD, the book value of aggregated asset portfolio of insurers operating in Czech Republic amounte 483 733 332 000 CZK in 2015. The investments were valuated at 348 722 616 000 CZK. The main part of investments is comprised by bonds and debentures (almost 80 %). Other major components are stocks (around 7 %), deposits with financial institutions (around 5.5 %), participating interests in affiliates (around 5.3 %) and real estates (around 1.4 %). Due to the lack of available data for life insurers we consider the above mentioned shares of investments for them. We consider that all bonds are bought in face value and the stocks and real estates are bought for their market value at time t. The book value of the fc-th cohort of bonds BKBV at the time of purchase t equals its face value BKFV and its market value BKMV. If we denote T its maturity time we get r kbv _ o k,FV _ o k,MV H \ D(t,T) T — D{t,T) ■ v1' Similarly, the book value of the Z-th cohort of stocks Sl,BVat the time of purchase t is equal to its purchasing costs'^and its market va\ueSlMV. We get Sl,bv = s1,fv = sl,MV_ (2) If during the holding period the market value of asset falls below the book value, it must be later reduce too. Hence we need to know the market value of assets. The market value for the fc-th cohort of bonds is given by Bk,bv = V lB(T-T) lc,(T-T)\ V-t) ^ where ic is the coupon, id is the discount rate applied to the cohort of assets and t is the life time of the bond that has gone by at time t. The market value of the cohort of stocks follows the Geometric Brown motion. We consider the market value equals book value at time t = 0. But gradually the market values are modified according to Geometric Brown motion. The market value of Z-th cohort of stocks is given by si,bv = plT + (1 - *) ■ - for S? > SV (11) (12) i=i Nk indicates the number of bond cohorts and JV* indicates the number of stock cohorts. The Liabilities Current regulatory framework in the Czech Republic permits a maximum technical interest rate for discounting policy reserves. It means that when the insurer makes a choice on the rate to use, no less than this technical interest rate have to be credited to clients' accounts every year. The technical interest rate is determined by Czech national bank. The performance of technical interest rate we can see in the Table 1. Table 1 Technical Interest Rate in the Czech Republic Year 2000 2004 2010 2013 2015 Interest rate 4.0 % 2.4 % 2.5% 1.9 % 1.3 % Source." Finance: Vývoj TUM [online], [cit. 2016-05-10]. Retrieved from: http://www.finance.cz/pojisteni/osoby/zivotni-pojisteni/tum/ Unlike some other countries (e.g. Germany) the allocation of profit shares is not provided by law in the Czech Republic. Therefore the technical interest rate is only the transparent security for clients. We consider that the duration of typical endowment policy is 12 years and the duration of the typical annuity product varies from 17 to 24 years, see Berdin E. and Grtindl H. (2015). It depends on characteristics of the product. Consider that the insurer sell only homogeneous product. As the cohorts of contrasts mature, they are constantly rolled 296 over by new cohorts of contracts. These cohorts hold the maximum guaranteed interest rate in effect at the time of beginning of the contract. Assume the typical product that allows to insurer to decide how much additional return could be divided based on regulatory limitation and financial information obtainable up to determination moment. Development of client's account is given by ^,SV = ^BV . [± + ^iS^ + pi (13) where ll^fv is the book value of account at time t-l, i indicates the tariff generation which is connected with the minimum guaranteed return fixed at the contract beginning, rtl,a is the regulatory minimum rate of return, rtp is a rate of return with additional return and Pt is the annual premium which the client must pay for the entire duration of the contract. The aggregate book value of liabilities can be determine as £ . (14) i = l where Nz is the number of cohorts in the portfolio. Finally, we define the market value of liabilities. We need the best estimate of liabilities that is a discounted minimum final payment which the insurer has to pay at the end of the contract. The present value for each cohort at time t can be expressed as ftflg_Z^-(l + rOr-T (15) (l + £d,(t,r-T)) where T is the maturity time, T-t is the remaining time to maturity of cohort i and id,(t,r-r) is the discount factor used for the last payoff, where (t,T - t) expresses the point in time t and the appropriate T-t maturity. By aggregating all unfinished contrasts at time t, we get the best estimate of the technical reserves (16) Then we get the market value of liabilities L^v=LBtE+RMt. (17) where RMt is the risk margin estimated at time t. RMt is defined as a function of the solvency capital requirement. The Solvency Position The coordination between the asset and the liability side identifies the solvency position of the insurer each year. According to Solvency II regulation it is required the calculation of the Solvency Capital Requirement (SCR). It is given by SCRt :=arg minjp (OFt--°Ft+1 > x] iSCRt (19) RMt =-- -CoC v ' 1 + r/(t,t+i) 297 where _Li5Cfit denotes the prediction of the solvency capital required to cover the whole life (T) of the liability portfolio discounted to the contemporaneity using the risk free term structure. CoC indicates the Cost-of-Capital rate that returns the RMt of the whole portfolio at time t. 3 Results and Discussion According to the Final Stability Report of EIOPA (2015) low yields and following reinvestment risk is the main problem in the insurance sector. Insurers have troubles reinvesting their assets at an acceptable level. Furthermore, monetary policy in Europe can further extend the actual low return environment. In the Figure 2 we can see the evolution of 3 month EURIBOR. At the beginning of the year 2016 it has even reached negative value. Figure 2 3M EURIBOR — — 3rn Eurh«rfi»r rat* - SprtnE »14 - - 3m Ewiborfor nl* Autwnn ZD14 — — 3-r-m Euribor far i-aw * - S^int 2f>lS 3m E urit»r f w nr* - -m*jrrn> - 3iM EvHbc Source: Final Stability Report of EIOPA (2015) The evolution of the government bond's yields can be observed in the Figure 3. Their returns are at very low levels. After the turbulence made by the conditions in Greece in June and July 2015, euro area government bond returns have at that temporarily decreased. The present market situation will continue to push the insurers' profit down and put pressure on balance sheets. Comparably, Euro area corporate yields stay largely at very low levels. As stated above, bonds represent 80 % of the assets of insurance companies in the Czech Republic. However, current evolution of bond's yields is very unfavorable. Due to this situation the stocks of old policies has expensive guarantees for insurance companies. A protracted period of low interest rate could distinctly affect the solvency situation of life insurers, especially the less capitalized companies. Such insurance companies might default if interest rate stays at the current low level. Figure 3 10-year Government Bond Yields (%) -—' r^j m «_r^j t_n Ln Source: Final Stability Report of EIOPA (2015) 298 4 Conclusions In the recent low yield situation it is still more and more difficult to maintain profitability. This is relevant particularly for life insurance companies who have guaranteed returns on their contracts and some of these old contracts guarantee the maximum rate of return between 4% and 5%. I this situation they don't have any possibility to change the terms and conditions of these contracts, e.g. in Belgium, Germany or France. In the Czech Republic the insurance companies use the technical interest rate which is determined by Czech National Bank. The value of this rate is between 4%- 1.3%. Now we work on quantification of the effects of a prolonged period of low interest rate on life insurers. We want to model a different initial capital endowments and then we want to compare the results of the simulations. Acknowledgments The support of the Masaryk University internal grant MUNI/A/1025/2015 Risks and Challenges of the Low Interest Rates Environment to Financial Stability and Development is gratefully acknowledged. References ARAD - Time Serie System - Czech National bank: Insurance undertakings, total. Retrieved from: http://www.cnb.cz/cnb/STAT.ARADY_PKG.STROM_SESTAVY?p_strid = BC B&p_sestu id=&p_la ng = EN. Berdin, E., Grúndl, H. (2015). The Effects of a Low Interest Rate Environment on Life Insurers. The Geneva Papers on Risk and Insurance-Issues and Practice, vol. 40(3), pp. 385-415. EIOPA (2015). Financial Stability Report - December 2015. Retrieved from: https://eiopa.europa.eu/Pages/Financial-stability-and-crisis-prevention/Financial- Stability-Report-December-2015-.aspx. Finance: Vývoj TUM [online]. Retrieved from: http://www.finance.cz/pojisteni/osoby/zivotni-pojisteni/tum/. Holsboer, 1 H. (2000). The impact of low interest rates on insurers. The Geneva Papers on Risk and Insurance - Issues and Practice, vol. 25(1), pp. 38-58. 299 Impact of the REPO Rate on Commercial Rates in the Czech Republic František Kalouda 1 Masaryk Universitysity Faculty of Economics and Administration, Department of Finance Lipová 41a, 602 00 Brno, Czech Republic E-mail: kalouda@econ.muni.cz Abstract: The paper is focused to further description and analysis the new aspects of behavior of Czech banking as a cybernetic system. The aim of the paper is analysis of the managing system relations (regulator - central bank) to managed system (controlled system - commercial banks) as relationships between operational indicator (REPO rate) and regulated indicator (commercial rates). Methodology of the paper is strategically focused to the time series methods (first of all correlation) and trends analysis. The usual analytic-synthetic methods, literary research, description and comparison are used here as well.The main expected result of the paper are working conclusions related to the fundamental linkages (still not explored) between REPO rate and the commercial rates. Keywords: banking system, cybernetics, REPO rate, commercial rate JEL codes: C67, E58, G21, G38 1 Introduction There is still the opinion between the financial theorists that the market interest rate is highly influenced by the central bank and by its discount and REPO rates. According to (Revenda, 1999) „The main aim of discount rate changes ... caused by central bank ... is affection of movement, resp. of other interest rates level in economy and therefore influence on subjects' loans demand.". These central bank tendencies are in market economies quite analogic. This fact can be supported by an example from the USA "... as the instrument of FED serves discount rate - that is interest rate related to loans granted by FED to banks." (Mankiw, 2000). Previous analysis of the behavior of the banking system in the Czech Republic have therefore focused on processes for controlling the commercial rate (market interest rate) by using the discount rate (Kalouda, 2014a). The general conclusion for the central bank is devastating - their series of regulatory measures lead to a situation in which the regulatory tool, i.e. the discount rate, stopped having any effect (Kalouda, 2014a), (Kalouda, 2014b), (Kalouda, 2015). However there is a second option - REPO rate. „ ..... it is expected that the central bank may/can influent interest rates of client loans relatively well influence through the development of REPO rate." (Revenda et al., 1999). This paper is therefore focused to analysis of the Czech National Bank (regulator -control element) and the system of commercial banks (controlled system) as relationships between operational indicator (REPO rate) and regulated indicator (commercial rates). The main aim of the paper is therefore the analysis of REPO rate impact on market interest rate. To reach this aim there will be investigated following problem areas: • fundamental applicability of cybernetic approaches in selected area • linearity of the Czech banking system • regulation accuracy of the Czech banking system. 300 2 Methodology and Data Methodology The REPO rate influence on market interest rate is in this paper conceptualized in general terms as a problem of communication and of management. Therefore it is suitable to apply methodical apparatus of theoretic discipline in these terms that was in the Czech Republic used for these intensions only marginally. This theoretic discipline is "cybernetics ... as a science about general laws of origination, transmission and processing of information in complex systems......" (Kubfk et al., 1982). From these methodical instruments of cybernetics (more exactly of technical cybernetics) there will be further used: • static function (Kubfk etal., 1982), • theory of hysteresis function (Svarc, 2003). The area in which we are applying these methodological tools is usually designated as „economic cybernetics" (Svarc et al., 2011). Model Specification The real-life object which we shall be modelling in this paper is the banking system of the Czech Republic. We shall model the processes of managing the price of capital at a business level (commercial rate) through the use of the REPO rate. The model for this real-life system is the static characteristics, one of the deterministic methods for identifying systems (Fikarand Mikles, 1999). This is a relatively simple model, based on the linearity of the modelled system (Svarc et al., 2011). The relative simplicity of the model used does not prevent it from being used for primary identification, for acquiring the indicative characteristics of the analyzed system (Fikarand Mikles, 1999). Data This paper draws on data published by the Czech National Bank (CNB) at http://www.cnb.cz, to which we link here (to save space). The basic file values of the variable - REPO rate and commercial rate - are monitored (see Figure 1 Correlation commercial rate - REPO rate). Figure 1 Correlation Commercial Rate - REPO Rate (r = 0,957345) 301 3 Results and Discussion Fundamental applicability of cybernetic approaches in the selected area The source Allen (1971) emphasizes in these terms unequivocal opinion: „There is necessary only the formal similarity to anticipate that the methods used in technics will be suitable for economic models too." This condition is met in our case. Nevertheless, the same source mentions an important problem with application of methods that have been successfully proved in technical sphere on the economic sphere - the linearity of the models. „Linear models can be generally suitable for technics where everything can be accurately managed. Be sure that they are not suitable up to the same extent for the economic models." However, there is accepted the possibility of linearization. Linearity of the Czech banking system - static function of commercial banks The static characteristic of the Czech banking system for all available data are presented in Figure 2. From here it is obvious, that this data file is for our purposes not applicable. Its uncertainty is excessive. Figure 2 Static Characteristics (REPO Increasing and Decreasing) - All Available Data so £ y = o, F 7135X+3 *2 = 0,916 ,0902 5 ♦ tit*4 IB- 012345678 REPO rate (%) Source: Own construction of the author by using http://www.cnb.cz Above that, the cybernetic approach requires data in stabilized state. "Static characteristics of control members are mostly expressed by the static function, i.e. the dependence between the output indicator in stabilized state and entry indicator in stabilized state." (Švarc, 2003). That means, according to our expert opinion, the values constant during the time period long at least three months. This view is supported (using the regression) by source Šerý (2010). It is obvious that the system of commercial banks can be with acceptable inaccuracy rate recognized for linear (see Figure 2 and 3). For comparison - dependence quite same in type is presented as a linear one in source Balatě (2004) too. 302 Here we are only analyzing in the sources those reactions of the system that are usually considered, i.e. the reactions to a rise in the REPO rate. Figure 3 Static Characteristics (Increasing REPO) - Stabilized Data Only OJ 25 .2 °u aJ4 s3 y = i, R 3493x+ 1,960 2 = 0,9724 ^ 4 ♦ REPO rate Source: Own construction of the author by using http://www.cnb.cz Regulation accuracy of the Czech banking system (central bank and commercial banks) From Figure 1 it is clear that meaningful data (for REPO increasing and decreasing) appears in the interval 9/2005 - 9/2012. Interaction between central bank and commercial banks leads to negative synergies when there the non-linearity of the hysteresis type occurs (see Figure 4). This state eventuates in conclusion that REPO rate is in fact not able to manage value of commercial rate. In its implications it means that after certain cycles of "increase-decrease" type the REPO rate will lose its ability to regulate commercial rate. 4 Conclusions With consideration of theoretic knowledge and according to stated available data processing there can be on the discussion basis formulated succeeding paper conclusions: Fundamental applicability of cybernetic approaches in selected area Cybernetic approaches are for the settled task (inquiry into economic processes) undoubtedly utilizable. The possible problems with disputable linearities can be in the first approximation solved by linearization of tackled problem. Linearity of the Czech banking system - static function of commercial banks Commercial banks conduct themselves as a linear system in principle. The rate of the current identifiable nonlinearities is so low that the linearization does not bear any major problems. 303 Figure 4 REPO Hysteresis (REPO Increasing and Decreasing) - Stabilized Data Only [ ♦ < t ♦ - ♦ * A 4,07„4i !»--' 0 0,5 1 1,5 2 2,5 3 3,5 4 REPO rate (%) Source: Own construction of the author by using http://www.cnb.cz Regulation accuracy of the Czech banking system (central bank and commercial banks) The nonlinearity of hysteresis type is typical for coexistence of central bank with commercial banks even in this case. It in fact means the loss of applicability of REPO rate as cost of capital (bank loan) management instrument at the level of commercial rate. Present level of REPO rate (0,05 %) confirms this in praxis. The paper results can be surprising in a certain manner. They theoretically confirm limited possibilities of REPO rate as an instrument for regulation of the market interest rate. In this relation there can be clearly seen management potential of the monobank (Kalouda and Svftil, 2009), (de SOTO, 2009) even if there are taken into account all disadvantages resulting from this variant of banking sector organizational order (Revenda, 1999). References Balatě, J. (2004): AUTOMATICKÉ ŘÍZENÍ, Praha: BEN. Fikar, M., Mikleš, J. (1999). Identifikácia systémov. Bratislava: STU Bratislava. Kalouda, F., Svítil, M. (2009). The Commercial Bank such as the Cybernetic System in Conditions of the Global Financial Crisis. In: Management, Economics and Business Development in European Conditions. Brno, FP VUT, pp. II.24-11.31. Kalouda, F. (2014a). Cost of capital management by central bank like the cybernetic model. In: Jedlička, M., ed., HED2014 (Hradecké ekonomické dny), Hradecké ekonomické dny 2014. Hradec Králové, ČR: Gaudeamus, Univerzita Hradec králové, pp. 428-434. Kalouda, F. (2014b). The Impact of Discount Rate on Commercial Rates in the Czech Republic: The Cybernetic Approach. In: Deev, O., Kajurová, V., Krajíček, J., ed., European Financial System 2014. Brno, ČR: Masaryk University, pp. 307-313. 304 Kalouda, F. (2015). The Banking System of the Czech Republic as a Cybernetic System -a Unit Step Response Analysis. In: Deev, O., Kajurová, V., Krajíček, J., ed., European Financial System 2015. Brno, ČR: Masaryk University, pp. 253-261. Kubík, S., Kotek, Z., Strejc, V., Štecha, J. (1982). Teorie automatického řízení I, Lineární a nelineární systémy, Praha: SNTL. Mankiw, N. G. (2000). Zásady ekonomie, Praha: Grada Publishing. Revenda, Z. (1999). Centrální bankovnictví. Praha: MANAGEMENT PRESS. Revenda, Z., Mandel, M., Kodera, J., Musílek, P., Dvořák, P., Brada, 1 (1999). Peněžní ekonomie a bankovnictví, 2nd ed. Praha: Management Press. SOTO de, 1 H. (2009). Peníze, banky a hospodářské krize. Praha: ASPI a Liberální institut. Šerý. M. (2010). Vliv diskontní sazby na úrokové sazby komerčních bank v České republice (bakalářská práce). Brno, Mendelova univerzita v Brně, Provozně ekonomická fakulta. Švarc, I. (2003). Teorie automatického řízení, Brno, FS VUT. Švarc, I., Matoušek, R., Šeda, M., Vítečková, M. (2011). Automatické řízení, 2nd ed. Brno: CERM. 305 The Interaction between Venture Capital and Innovation in Europe Ozcan Karahan Bandirma Onyedi Eylul University Faculty of Economics and Administrative Sciences, Economics Department Yeni Mahalle, No:77, 10200, BaIikesir, Turkey E-mail: okarahan@bandirma.edu.tr Abstract: Venture Capital (VC) has been a significant source of finance for technology-based investments. Thus VC has played a key role in fostering innovative entrepreneurship and technological progress in new economies. Accordingly, it is generally argued that there is a positive causal relationship from venture capital to innovation, which called "venture capita I-first hypothesis". However, some studies in literature indicates an opposite causality that innovation stimulates venture capital, which called "innovation-first hypothesis". The aim of this paper is to test these hypothesis to determine direction of causality between Venture Capital and Innovation in European countries. We use dynamic panel data analysis in order to investigate the direction of causality between innovation and venture capital based on annual data set related to European patent applications and venture capital investment. Empirical results of our study provide a strong evidence in favour of "innovation-first hypothesis" that innovation induces venture capital investment in Europe. This result presents significant implications for innovative entrepreneurship capacity of Europe. It seems that, although policymakers aim to make more financial resources available for innovation, the absence of innovative ideas is the big issue rather than the lack of available fund for innovative entrepreneurships in Europe. Keywords: venture capital, innovation, Europe JEL codes: G24, 031, 032 1 Introduction Innovation as the basic source of competiveness in global economy increases the significance of technology-based firms which are mostly young and dynamic entrepreneurships. Technology-based firms are basic dynamic of the new developed economies since this kind of firms are closely related to technology based investments. Indeed, improvements of new economies based on productivity growth has been dominated by the high market performance of technology-based investment. Most of the academic researchers have determined the role of the technology-based firms and their investment aiming to produce innovation in the development of new economies. Thus, advocating of technology-based investments become a significant aim for the policy-makers to stimulate productivity growth. The basic challenge from this policy framework arises from the matters concerning with the finance of technology-based investments. So that there is big challenge for the financing of technology-based investment in new economy since these kinds of investments have distinctive features like high risk profile and high degree of uncertainty. Besides these kind of investments have long term growth potential derived from research and development activities. Therefore, compared to other types of investment, technology-based firms are constrained by different or additional financing problems in the financial markets. In other words, technology-based investments suffer from the extreme difficulties in order to get enough funds for their projects. That means the conventional patterns of financial intermediation cannot adequately finance the technology-based investment. In conclusion, the perceptions about how finance the new economy dominated by technology-based firms should be totally different from the financial methods in the past. Consequently, the path of the new economy progress is basically depend on the effectively distribution of financial resources towards technology-based investments. In fact, the relationship between finance and economic growth becomes always popular interest in the literature. However, this interest is much more vital to understand and 306 manage the basic dynamics of new economies. Because of the characteristics of technology-based investment, this kind of activities aiming to innovation as basic source of new economies cannot be adequately financed by traditional channels. In other words, technology-based investments suffer from the extreme difficulties in order to get enough funds for their projects. Thus, the allocation of financial resources to technology-based enterprises is very significant for improving of new economy (Karahan, 2007, 50-52). At this stage, it seems that most effective way in order to fill the financing needs of technology based firms is venture capital. Indeed there are a lot of difficulties indicated literature to finance technology based investment. However, venture capital creates a model to finance of innovative activities which cannot be financed by traditional methods in the markets (Ueda, 2010, 304-308). Consequently, venture capital becomes a significant factor of examine the financial sector preconditions for the successful development of New Technology Based Firms in United States. Indeed, popular firms in the new economies mostly financed by Venture Capital like Apple Computer, Cisco System Microsoft, Netscape and Genentech. Indeed, funds organized in the form of venture capital supported to technology-based firms and caused to the establishment of entirely new industries advocating productivity growth in United States (Gompers and Lerner, 2001, 49). That means venture capital as a specific type of finance acts a crucial role enhancing the high-risk investment projects of technology-based companies. From the point of view of policy implications, it seems that creating stronger venture capital industry is crucial to provide adequate finance for technology-based firms which are the important component of new economies (OECD, 1996, 4). The purpose of this paper is to examine the role of venture capital in the development of new economy in Europe. Thus, our study points out the performance of the venture capital investments in Europe. The paper is structured as follows. In the second section, we review the studies focusing on the relationship between Venture Capital and innovative or economic performance. Third section presents data, methodology and empirical results. Final section concludes and makes some policy implications concerning with venture capital. 2 Literature Review Venture capital has a quite well-established business sector separated among the developed world. However, it can be argued that VC investment mostly concentrated geographically across United States. Nearly half of the VC investment have been realized in United States. This does not come as a surprise because VC as a special financial tool raised and developed in the United States. American Research and Development Corporation started as the first VC firm in 1946. Later United States government showed significant effort to develop own VC as part of the Small Business Act of 1958 (Metrick and Yasuda, 2011, 9-10). Accordingly, in the literature, experience of United States of America has been the most significant experience to show the relationship between VC and innovation. Thus, VC has been accepted as an effective financing method of technology-based investment generating most of the significant innovation in United states. Kenney (2011) historically analysed how VC evolved and advocated the national system of innovation in United States. He clearly indicated that most of the innovation activities performing by Information and Communication Technology and Biocenology Industries have been mostly financed by VC in the first development stages. Thus, venture capital has become an important financial intermediary for financing of technology-based investment resulting in innovation. Accordingly, he argued that Venture Capital has been one of the components of the US national system of innovation. Focusing on firm-level study, most comprehensive empirical study on the relationship between VC and innovation in United States has been performed by Kortum and Lerner (2000). They analysed annual data for 530 VC-backed and non-VC-backed firms in twenty manufacturing industries between 1965 and 1992. The findings of their study indicated that increases in VC activity in an industry are associated with significantly higher patenting rates. Hellmann and Puri (2000) also examined the role of venture capital financing in Silicon Valley by using dataset based on a survey of 149 recently 307 established firms. They specially tried to indicate the interrelationship between the type of investor and aspects of the product market behaviour of start-up firms. They found that firms financed by VC pursue and innovative strategy compare to non-VC backed firms pursuing imitator strategy. Thus they made a direct contribution to literature arguing that venture capital financing have an impact on the development path of a start-up company. Dushnitsky and Lenox (2005) focused on the potential innovative benefits to corporate venture capital in entrepreneurial ventures by incumbent firms to explore the relationship between corporate venture capital incumbent firm innovation rates. They analysed a large panel of 2289 public firms over period from 1969 to 1999. The findings indicated that increases in corporate venture capital investments are associated with subsequent increases in firm patenting. Accordingly, they concluded that corporate venture capital programs may be instrumental in harvesting innovation as a vital part of a firm's innovation toolkit. Besides directly focusing on the innovation impact of VC on firms, some studies examined the relationship between VC and different productivity performance of firms. Dushnitsky and Lenox (2006) explored the relationship between corporate venture capital and firm value creation by using the panel data of United States public firms during the period 1900-1999. They presented evidence that corporate venture capital investment is associated with the creation of firm value. Thus, they concluded that VC may be an important financial tool to enhance firm value by providing a valuable window on novel technologies. Peneder (2010) has examined the relationship between VC and firm's innovation performance using Austrian data. He found that VC-financed firms are generally more innovative and growth faster in terms of employment and sales revenue than other firms. Despite putting some reservation on the effectiveness of VC, he concluded that accessing VC remains an important pillar of well performing innovation system. Guo and Jiang (2013) examined the contributions of VC to entrepreneurial firms in China based on a panel dataset of manufacturing firms. Thus they compared to performance of VC-backed and non-VC-backed firms during the period 1998 to 2007. The results of empirical study showed that VC-backed firms outperforms non-VC-baked firms in terms of profitability, sales growth and Research and Development investment. They concluded that VC in China enhances entrepreneurial firms to provide more value-added. Some studies also concerned with the European experience to analyse the relationship between VC and innovative and economic performance of firm. Croce et al. (2013) aimed to indicate the impact of European venture capital-backed firms in high-tech industries. They used data on entrepreneurial firms operating in seven European countries (Belgium, Finland, France, Germany, Italy, Spain and the United Kingdom). Their findings showed that productivity growth is not significantly different between VC and VC-backed firms before the first stage of VC financing. However significant difference between them are found in the first years after the investment. In spite of the limitation, from a policy perspective, they concluded that VC financing is a valid tool for improving the performance of European entrepreneurial firms in high-tech industries. Lastly, Bertoni and Tykvova (2015) explored whether public venture capital spur innovation in young biotech companies in Europe. They used the dataset of all companies in the VICO database between 1984 and 2004. Findings did not show any impact of public VC on invention and innovation of biotech companies in Europe. Thus, they couldn't find the positive effect of government VC established throughout Europe on innovation performance of biotech sector. From the macroeconomics policy point of view, economists need to broaden the scope of examination the relationship between VC and innovation. Thus, studies started to focus on the interaction among VC investment and innovation or economic performance at national level. Romain and Potterie (2004) attempted to determine the economic impact of venture capital based on a panel of 16 OECD countries from 1990 to 2001. Empirical results showed that VC investment plays a more important role compared to business or public R&D in order to stimulate innovation. The findings also indicated that VC funding has a positive effect on the total factor productivity. In addition they indicated that VC intensity also makes it easier to absorb the knowledge stock generated by universities 308 and firms. Tang and Chyi (2008) provided a new explanation of total factor productivity growth of Taiwanese industry based on venture capital. They found that development of the venture capital industry significantly promotes productivity growth between 1985 and 2001 in Taiwan. Their findings also specifically indicated that venture capitalist in Taiwan not only support start-ups financially but also provide them guidance and expertise, which promotes internal knowledge diffusion channel resulting in total factor productivity growth. In the macroeconomic framework, some economists have also analysed the interactions between VC and innovative and economic performance in Europe. Empirical results found from European experience don't indicate the strong casualty relationship from VC investment to innovative or economic performance. Popov and Roosenboom (2012) provide cross-country evidence of the effect of venture capital investment on patented innovation. They used a panel of 21 European countries covering period 1991-2005. They founded that the impact of VC on innovation is quite weak and varies widely across European countries. The relation between VC and innovation is relatively stronger in countries with lower barriers to entrepreneurship and have adopted a tax and regulatory environment friendly to VC. Thus, Popov and Roosenboom recommend "policymakers in Europe to be careful not to see VC as a panacea to spur innovation". Faria and Barbosa (2014) examined the role of VC in promoting innovation by using panel data of 17 European Union countries over the period 2000-2009. They showed that only late-stage VC has an impact on innovation in Europe. That means European venture capitalist are more willing to support innovation only after the least risky initial stage. Thus, their findings indicated that VC in Europe enhanced the commercialization of innovation rather than fostering its creation. Finally, Faria and Barbosa indicated that this result don't show what not to expect from European venture capitalists regarding their role in supporting innovation. Looking at the literature, it seems that performance of VC investments in Europe is not enough to finance of technology-based investment or innovative entrepreneurships. Furthermore, some economists indicated the presence of reverse causality from innovation to VC investment. This reverse causality called "innovation-first hypothesis" while the argument that VC investments stimulate innovation called "VC-first hypothesis". Accordingly, Geronikolaou and Papachristou (2012) tested these hypothesis based on annual data covering the period 1995-2004 for 15 European countries. Empirical findings showed that causality from innovation to VC exists rather than from VC to innovation in Europe, which confirms "innovation-first hypothesis". That means the causality runs from patents to venture capital and not the other way around. Thus, they concluded that effectiveness of venture capital activity in Europe can be attributable to the absence of value creating innovative ideas or entrepreneurship. Similarly, Hirukawa and Ueda (2011) found same results in the United States manufacturing industry. They analysed the causality between VC and innovation symbolized by total factor productivity growth in the manufacturing industry over the period 1968-2001. Their findings also confirmed "innovation-first hypothesis" indicating the causality from total factor productivity growth to VC investment. Thus, they showed that the relationship between VC and innovation is not as simple as generally accepted form arguing the causality from VC to innovation. That means positive relations between VC and innovation cannot be always interpreted as evidence that VC stimulates innovation. 3 Data, Methodology and Empirical Results In this section, we examined empirically the causal relationship between the venture capital referring to the total amount of venture capital finance at seed, start-up and expansion stages and innovation referring to number of patent applications to European Patent Office. We used annual data covering the period 2000-2013 for 12 European Countries Belgium, Denmark, France, Germany, Greece, Ireland, Italy, Luxembourg, Netherlands, Portugal, Spain and, United Kingdom. The number of patent applications 309 are sourced from the European Venture Capital Association (EVCA) while venture capital data are obtained from EUROSTAT Database. In order to investigate the relationship between VC and innovation, dynamic panel data methodology developed by Holtz-Eakin et. al (1988) and Arellano and Bond (1991) will be applied. Accordingly, the model can be shown below: y«t = Po + _?=i Pj Yit-j + _?=i aj Axit_j + Ft + eit (1) (y) and (x) symbolize variables while (F) and (s) represent individual fixed effect and error term, respectively. Besides, (i) refer to panel unit, (t) represents time, and finally (j) shows the lag number. In order to eliminate the unobserved country-specific effects, equation (1) is differenced to derive the model below: Ayit = _?=! Pjyu-j + Z"=i txj Axit_j + Aeit (2) In the estimation process of Equation (2), Generalized Method of Moments (GMM) is performed by using the appropriate lags of the dependent and the independent variables as instruments. By using Wald Test procedure, the null hypothesis that x does not cause y and that y does not cause x are also jointly tested. Accordingly, based on methodology concerning panel causality model indicated equation (2), we specified and estimated two regressions shown in Equation (3) and (4) below. APAu = I%i YjAPAit.j + Zt=1Aj AVCu-j + Auit (3) AVCit = n=i &j AVCu-j + Zt=1Sj APAit.j + Aeit (4) where, i : 1, 2, 12, number of countries, t : 2000, 2001, 2013, time period (year) and j: 2 (lag number). PA represents the number of patent applications while VC symbolizes the total amount of venture capital. Before proceeding to panel data analysis we applied the Augmented Dickey-Fuller (ADF) and the Im, Peseran and Shin (IPS) panel unit root tests. The results of the panel unit root tests for the variables of venture capital (VC) and patent applications (PA) are presented in Table-1. Both the panel unit root tests concluded that the variables in level terms are non-stationary and become stationary only in first-differences. These results let us proceed to panel data analysis. Table 1 Panel Unit Root Test Variables ADF IPS Levels Differences Levels Differences PA 6.153 (0.47) 61.814 (0.00)* 3.121 (0.18) -2.101 (0.01)* VC 8.718 (0.35) 58.347 (0.00)* 9.328 (0.29) -1.918 (0.00)* Note: p-values in parenthesis and (*) indicates significance at the 1 per cent level. The coefficients estimated by Generalized Method of Moments (GMM), the results of Wald Causality Test and diagnostic tests for Equation (3) and Equation (4) are presented in the first and second columns of the Table.2. As seen from the Table.2, we employed the lagged values of the variables as instruments in levels for the first difference equations, for 2 and earlier. Wald X2 test shows that both models have significant individual coefficients. Two basic types of diagnostic tests are also conducted. The Arellano-Bond (AR) test is operated for indication of serial correlation in the residuals. Sargent Test is also applied for indication whether the instruments are correlated with the error term. The AR (1) and AR (2) tests indicate applicability of models as we expect that first order statistic is significant while the second order is insignificant. Sargent test rejects that the instrumental variables are correlated to some set of residuals, which indicates the validity of the set of instruments in all equations. In conclusion, the results of diagnostics tests 310 show that the models are well specified and two lags is appropriate for the panel GMM estimator. Concerning with the relationship between variables, the empirical findings shown in Table-2 that Venture Capital don't have a significant effect on patent applications in Europe. However, it seems that patent applications effect the level of venture capital investment. In other words, causality operates from patent applications as measure of innovation to venture capital. Thus, we find evidence that supports the "Innovation-first Hypothesis" instead of the "Venture Capital-first Hypothesis". European experience does not confirm the generally accepted form of causality from venture capital to innovation. Instead of this, the causality runs from innovation to venture capital. This reverse causality indicates that innovation creates a demand for VC but not VC supplies for innovation in Europe. This causality can operate like the process proposed by Hirukawa and Ueda (2011). Firstly, increasing innovation opportunities stimulate new firm startups to exploit such opportunities. Later, these start-ups demand and thus stimulate venture capital improvements in Europe. In conclusion, the relationship between venture capital and innovation is not as simple as we thought, since innovation are not only a consequence of venture capital but also likely be a cause. This results also show that European countries generally suffer from lack of innovative ideas or entrepreneurships rather than lack of funds to finance new ideas. Table 2 GMM Estimation and Causality Test Tk\/1 /~\ k"\ /~\ k-\ e-\ /~\ k-\ 4- \/_i>-i_k\ 1 /~\ c Dependent Variables independent vanauies APA AVC APAit-i 2,106 (0,00)* 0, 913 (0,00)* APAit-2 1,178 (0,00)* 0, 389 (0,01)* 0, 171 (0,34) 3,767 (0,00)* AVCit-2 0, 083 (0,47) 2,125 (0,01)* Wald x2 28.79 (0.02)** 54.11 (0.00)* AR(1) -2.34 (0.00)* -3.14 (0.00) * AR(2) -0.64 (0.21) -0.29 (0.39) Sargan Test 23.73 (0.467) 21.56 (0.389) Wald Causality Test 1.04 (0,43) (H0 = A t-i = A t-2 = 0) 10,85 ( 0,00)* (H0 = 8t-i = 8 t-2= 0) Note: P-values in parenthesis and (*), (**), indicate significance at the 1 and 5 per cent levels, respectively. 4 Conclusions In order to generate an innovative idea and utilize it as a commercially viable entity, there needs fund enough to finance all of this process. Given the fact that innovative activities are very risky to get the positive results in the end, the money is very scarce in the financial markets to finance this kind of economic activities. This shows why the venture capital is so important as a financial method providing fund for developments of new economies depending on technological progress. Accordingly, "Venture Capital-first Hypothesis" argues that there is strong causality from venture capital to innovation. We examined empirically the impact of the venture capital on innovation in Europe by using dynamic panel data analysis for the period 2000-2013. Empirical results of study provide evidence in favour of "Innovation-first Hypothesis" that the causality runs from 311 innovation to venture capital rather than generally accepted form of causality argued by "Venture Capital-first Hypothesis". This result arising from European experience showed that the relationship between venture capital and innovation is not as simple as generally accepted form of causality argued by Venture Capital-first Hypothesis. Instead of this, in the framework of Innovation-first Hypothesis, the causality may run from innovation to venture capital. From the perspectives of public policy aiming to increase innovation, as Popov and Roosenboom (2012) recommended, our results also indicated that "policymakers in Europe to be careful not to see venture capital as a panacea to spur innovation". In other words, improving venture capital sector will not be the most significant precondition for the successful development of national innovation systems. It seems that European countries generally suffer from lack of innovative ideas or entrepreneurships rather than lack of well-functioning venture capital sector to finance new ideas. Thus, the absence of innovative idea should be big issue rather than the lack of available funds for policymakers aiming to make more financial resources available for innovation in Europe. References Arellano, M., Bond, S. (1991). Some Tests of Specification for Panel Data: Monte Carlo Evidence and an Application to Employment Equations. The Review of Economic Studies, vol. 58 (2), pp. 277-297. Croce, A., Marti, J., Murtinu, S. (2013). The Impact of Venture Capital on the Productivity Growth of European Entrepreneurial Firms: Screening or Value Added Effect. Journal of Business Venturing, vol. 28, pp. 489-510. Bertoni, F., Tykvova, T. (2015). Does governmental venture capital spur invention and innovation? Evidence from young European biotech companies. Research Policy, vol. 44, pp. 925-935. Dushnitsky G., Lenox J., M. (2005). When do incumbents learn from entrepreneurial ventures? Corporate venture capital and investing firm innovation rates. Research Policy, vol. 34 (5), pp. 615-639. Dushnitsky G., Lenox J., M. (2006). When does corporate venture capital investment create firm value? Journal of Business Venturing, vol. 21 (6), pp. 753-772. Faria, P., A., Barbosa, N., (2014) Does Venture Capital Really Foster Innovation? Economics Letters, vol. 122, pp. 129-131. Geronikolaou, G., Papachristou, G. (2012). Venture Capital and Innovation in Europe, Modern Economy, 3, 454-459. Gompers P., Lerner J. (2001). The Venture Capital Revolution. The Journal of Economic Perspectives, vol. 15 (2), pp. 145-168. Guo, D., Jiang K. (2013). Venture Capital Investment and the Performance of Entrepreneurial Firms: Evidence from China. Journal of Corporate Finance, vol. 22, pp. 375-395. Hellmann, T., Puri, M. (2000). The interaction between product market and financing strategy: The role of venture capital. Review of Financial Studies, vol. 13, pp. 959-984. Hirukawa, M., Ueda, M. (2011). Venture Capital and Innovation: Which is First? Pacific Economic Review, vol. 16 (4), pp. 421-465. Holtz-Eakin, D., Newey, W., Rosen, H. (1988). Estimating Vector Autoregressions with Panel Data. Econometrics, vol. 56 (6), pp. 1371-1395. Kenney, M. (2011). How venture capital became a component of the US National System of Innovation. Industrial and Corporate Change, vol. 20 (6), pp. 1677-1723. Metric, A., Yasuda, A. (2011). Venture Capital & the Finance Innovation. Second Edirion, John Wiley & Sons, Inc. United States. 312 OECD (1996). Venture Capital and Innovation, OECD/GD(96)168, Paris. Peneder, M. (2010) The Impact of Venture Capital on Innovation Behavior and Firm Growth. Venture Capital, vol. 12 (2), pp. 83-107. Popov, A., Roosenboom, P. (2012). Venture Capital and Patented Innovation: Evidence from Europe, Economic Policy, July, pp. 447-482. Romain, A., Pottelsberghe, B. (2004) The Economic Impact of Venture Capital. Discussion Paper Series: Studies of the Economic Research Centre, 18, Deutsche Bundesbank. Tang, M-C, Chyi, Y-L. (2008). Legal Environments, Venture Capital and Total Factor Productivity Growth of Taiwanese Industry, Contemporary Economic Policy, vol. 26 (3), pp. 468-481. Ueda, M. (2010). Venture Capital and Innovation, in Douglas Cumming (Eds), John Willey&Sons, Inc. New Jersay, pp. 299-314. 313 Preventing Crises in the Banking Sector and the Role of Internal Audit in Corporate Governance Maria Klimikova1, Martina Muchova2 1 University of Economics in Bratislava The Faculty of National Economy, Department Banking and International Finance Dolnozemska cesta 1, 852 35 Bratislava, Slovakia E-mail: klimikova@centrum.sk 2 University of Economics in Bratislava The Faculty of National Economy, Department Banking and International Finance Dolnozemska cesta 1, 852 35 Bratislava, Slovakia E-mail: flymatus@gmail.com Abstract: Analysis of the causes of the financial crisis led to define different views and to seek factors triggering the crisis. The financial crisis has revealed that there has been no effective system of governance of financial institutions in crisis, and that there are very few measures that should be adopted by banks in the event of a banking crisis. At the same time, the fall of a number of major financial institutions turned his attention to the Corporate Governance. The attention of the professional public has focused on revising of the Corporate Governance's role and thereby increasing the role of Internal Audit in the Corporate Governance. This paper focuses on implementation of the Corporate Governance in financial institutions as well as on impact of the updated rules of Corporate Governance and the role of Internal Audit in Corporate Governance. Keywords: corporate governance, internal audit, financial crisis, management, financial sector JEL codes: G01, G10, G018, G21, M42 1 Introduction The way of doing business has changed unrecognizable since the time when the first corporations were created. It has chained substantially even during the last 20 years. The recent financial crisis has led to a loss of trust in corporate governance and in particular on remuneration practices and the propensity for these to create excessive risk-taking, particularly in the financial sector. System of corporate governance was in its beginnings focused only on maximizing of a profit without taking into account the maintaining of such evolution. The necessity for improvement of management of the company has led to the emergence of different platforms, which have defined new principles for the definition of the relationships between share-holders, management and other participants (employees, clients, others). Principles of corporate governance firstly appeared in 1992 in the Great Britain, therefore, in the environment of an advanced market economy. Later on, in 1999, the OECD issued a set of corporate governance standards and guidelines to assist governments in their efforts to evaluate and improve the legal, institutional and regulatory framework for corporate governance in their countries. The importance of the professional background for efficient requirement of knowledge, assessment and implementation of principle of corporate governance into practice was emphasized by constituting of company with name Central European Corporate Governance Association (CECGA). After updating of Corporate Governance OECD Principles in 2004, CECGA released the Corporate Governance Code for Slovakia to regulate internal company relations and environmental matters, based on the principles of openness, integrity and accountability. The focus of this paper is the challenge to analyze the development of corporate governance in the scientific literature, in research and in economy practice. The question 314 of interest is to how internal audit contributes to the better corporate governance of bank and to the whole financial system. The question of internal audit's quality therefore extends also on the providing of assurance and on the practical application of "Three lines of Defense Model" due to prevention of crises in banking sector. 2 Methodology and Data The paper will deal with the corporate governance as well as with the internal audit in bank and his role in corporate governance. The methods of description, analysis, synthesis and deduction will be applied. The conclusion will include the current problems and trends in the field of corporate governance, especially in the area of internal audit with describing of the new tools to be used. 3 Results and Discussion It is also necessary to identify the main knowledge of corporate governance in the literature. During the last decade, new approaches of the corporate governance, totally confirmed the board's responsibility for ensuring the effectiveness of their organization's internal control framework. These theories stressed the key role that internal audit can play in supporting the board in ensuring adequate oversight of internal controls and in doing so form an integral part of an organization's corporate governance framework. The key role of internal audit is to assist the board and/or its audit committee in discharging its governance responsibilities. Theoretical background of Corporate Governance The essence of the whole issue of corporate governance lies in the separation of ownership and executive decision-making in a public company (Keasey K. et al., 1997). This creates space for the management of the public company, which, in its decisionmaking, and due to information asymmetry and divergent interests, may refrain from the ideals of profit maximization and cost minimization in the public company, and resort to various redistribution strategies (Klimikova and Muchova, 2013). According to the definition "corporate governance includes a group of relations between the company's management, its board of directors, shareholders and other stakeholders"(OECD, 2015). Corporate governance also establishes the structure by means of which the company goals are set and the means to achieve them are determined. Corporate governance should establish appropriate incentives to the board and management for promoting the goals which are in the interests of the company and shareholders, and should facilitate effective supervision or inspection, and thus facilitate more efficient use of resources on the part of the bank"(Klimikova etal., 2012). The concept of corporate governance was defined by several authors: Sir A. Cadbury defined the report of the company as "a system of governance and control" (Cadbury, 1992). The authors Keasey, Thompson and Wright are of the opinion that "from a closer perspective, the message can be the company understood as the formal system of accountability of senior management in relation to the shareholders (Keasey et al., 1997). The authors Cochran and Watrick define the message of the company as "the relationship between shareholders, administrative bodies, top management and other financially interested parties, the lenders, the banks and other entities" (Cochran and Watrick, 1998). B. I. Tricker defines corporate governance as the branched-chain relations of critical stakeholders (Tricker, 1993). F. Okruhlica stated the most appropriate characteristics as the approach to the definition of B. I. Tricker - it is the issue of "property relations and systems, such owners exercise their rights of ownership and control towards the management area of the company. 315 At the same time, corporate governance includes processes, structures and relationships through which authority oversees the activities of its executive's workers (Okruhlica, 2013). Corporate governance is a means for the promotion of economic efficiency, sustainable growth and financial stability. In 1999, the OECD published corporate governance codex, which is created on the principles of openness, accountability and honesty. It has to avoid non-ethical behavior, the submission of false financial statements, not taking responsibility of the board of directors and the supervisory board for their actions. This is in the interest of the company that is adhering to the principles of corporate governance easier it gets to credit and attract investors. In 2002, the OECD issued a document intended for public's comment with the aim of safe and healthy management of the company. This effort should also contribute to the development of a culture of values, professional and ethical behavior on which well-functioning markets depend, see Figure 1. Figure 1 Risk Culture Framework Fig 1 IRM Risk Culture Framework Source: https://www.theirm.org/knowledge-and-resources/thought-leadership/risk-culture.aspx The Principles have been adopted as one of the Financial Stability Board's key standards for sound financial systems, and have been used by the World Bank Group in more than 60 country reviews worldwide. They also serve as the basis for the guidelines on corporate governance of banks issued by the Basel Committee on Banking Supervision. The Corporate Governance Principles were drawn up in 1998 by representatives of the central banks affiliated with BIS and were previously revised in 2006 and 2010. After a public consultation procedure (October 2014 - January 2015), the third revision was published on 8 July 2015. The revised Corporate Governance Principles are part of a broader trend towards an increased focus on the governance of financial institutions. This is one of the pillars of CRR/CRD IV, the European project which as at 1 January 2014 raised the Basel III agreements to the level of legislation. Compliance with the principles of corporate governance is one of the factors which increases the confidence of the investors against the management and contribute to the protection and development of their investments. Corporate governance reflects an acceptable framework of the business of the bank and destines by and the focus and procedures of control mechanisms, especially internal audit. The Role of Internal Audit in Corporate Governance Internal audit as part of the control mechanism in the bank currently understands comprehensively, with an emphasis on advice and consultation. A more differentiated approach to areas with a different severity of risk. Over the years internal audit in 316 banking has migrated from inspection towards adding value and helping organization to develop a better culture. The essence of internal audit is given in its definition, which is adopted by internal auditors in the year 1999, where it is defined: "independent, objective, assurance and consulting services aimed to add value and improve processes within an organization. Internal audit helps the organization to accomplish its aims by bringing a systematic disciplined approach to the system of risk assessment and management, internal management and control system of the organization and corporate governance" (Kasparovska, 2006). Internal audit has a role in assessing and critiquing management's preparedness to be able to react to any changes. Internal audit should be looking at how well-positioned management is to identify that a currency movement or commodity price drop has occurred and the impact of that on the business, but also how to respond to that. For internal audit, it is crucial to assess management's assumptions, especially when they form the basis for strategic planning and investment decisions. This means critiquing whether those assumptions are reasonable, how they are validated and how often they are updated to reassess investments and strategy. Reviewing payroll and accounts payable is one thing, but challenging the board about the planning assumptions in its strategic process is a bit close to home for some senior directors. It is in adding value by assuring the organization's or bank's global strategy as well as its controls and governance where internal audit can really prove its worth. Internal audit and assurance The value of internal auditing can be described by these three very important words: assurance, insight and objectivity. Management and governing bodies can look to their internal auditors to provide assurance on whether policies are being followed, controls are effective, and the organization is operating as management intends. Internal auditors have unique insight on which risks might lead to disaster; how to improve controls, processes, procedures, performance, and risk management; and ways to reduce costs, enhance revenues, and increase profits. Internal auditors view the organization with the strictest sense of objectivity that separates them from - but makes them integral to - the business. Internal audit is a key source of independent assurance. The regulators exerted great influence in the development of the IIA's code for effective internal audit in financial services (July 2013). For the code to be implemented effectively, we need to appreciate the assurance network, so that stakeholders work collaboratively across all lines of defense and optimize collective risk intelligence. The code reinforces the link between risk management and risk assurance, and requires a more conscious approach to seeking and providing assurance across the risk spectrum. This link is consistent with the drive in corporate governance to align risk management and assurance across boundaries, and these expectations are found in the emerging practices of integrated assurance. Whether these help to minimize gaps and duplication in assurance plans or facilitate discussions about risks and the control environment, integrated assurance is encouraging a more conscious approach - and internal auditors have an opportunity to be the guiding light (Muchova and Klimikova, 2016). Three Lines of Defence Model A framework with which the board can understand the role of internal audit in the overall risk management and internal control process of an organization provides a model called "Three Lines of Defence Model", see Figure 2. To ensure the effectiveness of an organization's risk management framework, the board and senior management need to be able to rely on adequate line functions - including monitoring and assurance functions - within the organization. In order to conceptualize these line functions, the ECIIA endorse the use of the "Three lines of Defence" model. It is already widely adopted within the financial industry, but can also be productively utilized in a wide range of sectors. 317 The "Three Lines of Defence Model" organization's internal control levels: structure is a conceptual delineation of an • first line of defense for risks - is the line of business unit; • second level monitoring controls - is independent risk management (compliance, operations risk, etc.); • third line independent assurance - is the independent audit function. The Three Lines of Defence Model is a strategy much beloved by banks as a risk management filter. The "Three Lines of Defence" model for risk management has been accepted as a best practice by regulators and the Basel Committee on Banking Supervision (Risk Culture, 2016). Therefore, it is now "non-optional" for compliance risk management programs in regulated financial institutions. Each line of defence has a monitoring and/or testing responsibility. This is the area where there is often a great deal of overlap and not as much coordination as would be optimal (ValueWalk, 2015). Figure 2 Three Lines of Defense Model Senior Mana 1st line of Defence T_T |2nd line of Defence! Management controls Internal Control Measures Financial controller Security Risk Management Quality Inspection Compliance I 3d line of Defence Internal Audit I______________________________________________________________________J Source: http://riskoversightsolutions.com/wp-content/uploads/2011/03/Risk-Oversight-Solutions-for-comment-Three-Lines-of-Defense-vs-Five-Lines-of-Assurance-Draft-Nov-2015.pdf It also provides a framework with which the board can understand the role of internal audit in the overall risk management and internal control process of an organization. The revised Principles assign a central role to the "Three lines of Defense" model. This is not surprising because it has been the leading model used by supervisors around the world for some time, as the Basel Committee concluded in its 2011 publication "Principles of the Sound Management of Operational Risk". In connection with the revision of these Principles an extensive compliance investigation was carried out in 2014 into the degree of compliance with the "Three lines of Defense" model. The first line of defence owns regulatory quality control of its products, services, and operations. It should have built-in procedures in all of its processes that ensure that regulatory requirements are followed for all of its product lines. The first line should consult with the second - and they should then reach agreement in the interpretation and implementation of all regulations. Disclosures must be provided, deadlines must be met. Compliance, as the second line, has the responsibility to monitor and test periodically for every regulation to determine the level of compliance. Compliance testing is conducted on a risk-based priority schedule, because everything cannot be tested every year without a huge staff. The second line is responsible not only to the bank's board but to the regulatory agencies for compliance risk oversight. The second line is there to provide a check on the advice of the first line. Finally, the independent audit as the third line should have an audit schedule that also tests the level of compliance within the business units as well as the compliance program within the first and second lines. Each line of defence should be independent of the other. So, while the quality control function of the business line should be a daily process, the testing that Compliance and Audit do should be risk-based and conducted on a schedule so that the widest scope is 318 covered in a reasonable time period. Coordinating these three approaches can do a lot to make the overall compliance performance of the institution comprehensive and efficient. One issue that has surfaced is that the three lines have had difficulty coordinating the required responsibilities without overlapping each other and being inefficient (Muchova and Klimikova, 2016). 3 Results and Discussion The worldwide trend clearly shows the increasing demands in corporate governance, risk management, internal control, ethics and compliance. Companies are trying to be more global, and therefore need to manage the processes of corporate governance better, without which it is impossible to compete in global markets. National regulators and supranational institutions promote greater transparency and compliance with best practice of corporate governance. The main emphasis is on the establishment of a fully independent audit committee, reliable risk management and effective internal audit. The role of internal audit in the staff training is also important. A rapidly developing area of internal audit is ethics and compliance. For this area, it will be necessary to adapt the methods and procedures so that is possible to detect and limit the potential threats arising from the failure of ethical behavior at work and in business meetings. The trends in development of corporate governance are: management of social and psychological aspects of behavior of members of the board, and inclusion of soft criteria in the nomination process, leadership based on shared values as the key to building an authentic cooperation and internal cohesion, systematic objective analysis of drivers, outcomes of work behavior throughout the entire company, including members of the governing board. In internal audit, the first such trend is the expanding role of internal audit and a greater scope of audits. Today, the role of internal audit expanded, particularly in: • risk assessment and management, • process optimization and design of cost reduction, • regulatory compliance, • cooperation in managing changes which lead to enhancement of the performance of organizations • evaluation of established instruments for preventing and detecting fraud. A key trend is also continuing evaluation of performance and quality, or evaluation of the effectiveness and efficiency of key processes and functions. Another key point is to create an organizational structure conducive to greater accountability, transparency and better cooperation. The aim is to create such organizational structures that facilitate cooperation and promote the flow of information between the auditors and the audited. The final key trend is the shift towards the audit based on assessment of risk management. Assessment of risk management is becoming a priority because risks have a greater and faster impact on the company than before. Furthermore, the internal auditors are expected to assure that a comprehensively performed risk assessment served as a basis for planning and implementation of the key activities of the organization (Klimikova et al., 2012). 4 Conclusions The recent financial crisis has been a particularly severe wake-up call, because it has adversely affected employment, consumer spending, pensions, the finances of national and local governments worldwide, and the global economy. Weaknesses in corporate governance structures within companies and banks were cited as reasons for excessive risk taking, skewed incentive compensation for senior managers, and the predominance of a board culture that values short-term gains over sustained, long-term performance. 319 However, these crises are manifestations of several structural reasons why corporate governance has become more important for economic development and a more significant policy issue in many countries. More generally, poor corporate governance can affect the functioning of a country's financial markets and the volume of cross border financing. Our main finding can be summarized as follows: A number of recent cases where the governance system has failed have triggered an intensive search for rules, methods, processes and institutions that could prevent similar failures from occurring in the future. A successful corporate governance system should successfully channel aspirations of experts in these different fields within the organization to the benefit of the organization as a whole. Good corporate governance is not just about an impressive architectural edifice of rules or a facade of institutional framework. The critical elements are processes and outcomes. We can conclude that the internal audit represents an important management tool which assists and supports management in order to identify and manage the risks bringing extra value for company activities. From our perspective the future of internal audit is to move entirely to risk-based audit to enhance shareholder value, assess the business improvement opportunities and even the earnings per share in the areas which they are auditing - and indeed to encourage and develop awareness of the role of the audit function across the entirety of the business. Acknowledgments This paper is the output of a scientific project VEGA 1/0124/14 "The Role of Financial Institutions and Capital Market on the Solution of Debt Crisis". References BIS (2010). Basel Committee on Banking Supervision, Principles for enhancing corporate governance (October 2010 - revised July 2015). Retrieved from: www.bis.org/publ/bcbsl76.pdf, A revised version of this document was published in October 2014: http://www.bis.org/publ/bcbs294.htm. BIS (2015). Basel Committee on Banking Supervision, Guidelines: Corporate governance principles for banks (July 2015). Retrieved from: http://www.bis.org/bcbs/publ/d328.pdf. Cadbury, A. (1992). The Financial Aspects of Corporate Governance: Report of the Committee on the Financial Aspects of Corporate Governance. Gee Publishing, London. Retrieved from: http://www.ecgi.org/codes/documents/cadbury.pdf. Cochran, P. L., Wartick, S. L. & Financial Executives Research Foundation, (1988). Corporate Governance: A review of the literature, New York, Morristown. Kašparovská, V. (2006). Řízení obchodních bank: vybrané kapitoly. 1st ed., Praha: C. H. Beck. Keasey, K., Thompson, S., Wright (1997). Corporate Governance: Economic and Financial Issues. Oxford University Press. Klimiková, M., Schwarzová, M., Vovk, M., Tašková, J., Sabolová, A. (2012). Bankový manažment a marketing I. Bratislava: IRIS, pp. 314-334. Klimiková, M., Muchová, M. (2013). Nové výzvy interného auditu v corporate governance. In: Aktuálne otázky finančných trhov: rec. zborník vedeckých statí (el. source). Bratislava: Fin Star, pp. 1-9. Klimiková, M., Muchová, M. (2016). Assurance - a vital tool of guiding the internal audit. In conference: Evropsky fiskální dialog, NEWTON College, a.s. 320 Klimiková, M., Muchová, M. (2016). Building of values from the view of corporate governance In: 15th International Conference of Doctoral Students and Young Scholars Economic, Political and Legal Issues of International Relations 2016. OECD (2015). G20/OECD Principles of Corporate Governance. Retrieved from: http://www.oecd.org/corporate/oecdprinciplesofcorporategovernance.htm. Okruhlica, F. (2013). Vlastnícke práva spoločnosti (Corporate Governance). Bratislava: Iura Edition, vol. 336, pp. 19 - 22. Drawing a line under the 3 lines of Defence in Banking (2014). Retrieved from: https://www.theirm.org/media-centre/published-articles/drawing-a-line-under-the-3-lines-of-defence-in-banking/. Risk Culture (2016). Institute of Risk Management. Retrieved from: https://www.theirm.org/knowledge-and-resources/thought-leadership/risk-culture.aspx. Three Lines of Defense vs Five Lines of Assurance (2015). Retrieved from: http://riskoversightsolutions.com/wp-content/uploads/2011/03/Risk-Oversight-Solutions-for-comment-Three-Lines-of-Defense-vs-Five-Lines-of-Assurance-Draft-Nov-2015.pdf. Tricker, B. I. (1995). International Corporate Governance. Text, Reading and Cases. Singapore: Prentice Hall College Div. ValueWalk (2015). Bank's Risk Governance Framework Needs Three Lines Of Defense: BIS. Retrieved from: http://www.valuewalk.com/2015/07/banks-risk-governance-framework-needs-three-lines-of-defense-bis/. 321 Mobile Technology on the Retail Banking Market Monika Klimontowicz1, Karolina Derwisz2 1 University of Economics in Katowice Faculty of Finance and Insurance, Department of Banking and Financial Markets ul. 1 Maja 50, 40-287 Katowice, Poland E-mail: monika.klimontowicz@ue.katowice.pl 2 University of Economics in Katowice Faculty of Finance and Insurance, Department of Banking and Financial Markets ul. 1 Maja 50, 40-287 Katowice, Poland E-mail: kderwicz@gmail.com Abstract: For the last few decades, the retail banking market has been influenced by many economic, political, legal, technological and social factors. Mobile technology has changed both the information and communication sharing, and customers' market behaviour. The social revolution has already happened and changed customers' expectations concerning banking services. Today there are 2,6 billion smartphone subscriptions globally. It is estimated that by 2020 the number of smartphone users will reach 6,1 billion. Mobile phones haves increasingly become tools that consumers use for banking, payments, budgeting and shopping. Today they enable using a different kind of financial applications offered not only by banks but also by other providers as telecoms or FinTech enterprises. The increasing number of offerings makes the choice much more difficult for customers. The purpose of the paper is to presents global and regional trends relating to mobile technology application on the retail banking market, customers' expectations and selected factors influencing the adoption of this technology. The paper includes desk research of existing data. The authors try to answer the question if the implementation of mobile technology on retail banking market means the future of retail banking or just new distribution channel. Keywords: mobile banking, mobile service suppliers, banking market, consumer expectations, mobile technology adoption JEL codes: G21, G23, E42, F65, D84 1 Introduction Under the influence of economic, political, legal, technological and social factors the world's economies have evolved from the industrial economy into the network economy. The network economy is based on information technology, connectivity, and human knowledge. The ease of information sharing and networking have changed the process of goods and services' creation and trade. Electronic commerce and electronic services have become one of the fastest developing fields of economy. Concurrently new category of finance called electronic finance has been created and implemented. As a result, the economic transactions between countries and their citizens have substantially risen, and financial transactions have grown remarkably. Today's economy is subordinated to money circulation. Money has become the condition and the basis of goods' creation, gathering, division and exchange. Both, the economic and social sphere, rely on the access to cash and cash equivalents. The finance is increasingly embedded in contemporary social, political and economic life and play a crucial role in the domestic and international economies (Epstein, 2005; Montgomerie, 2008; Freeman, 2010; Dolphin, 2012). The development of digital transactions has caused changes in the way of life, consumer behaviour on the markets and companies' business models. All these processes are strengthened by increasing usage of mobile technology. The purpose of the paper is to present the overview of the global and regional trends relating to mobile technology application on the retail banking market. 322 2 Methodology and Data The paper methodology applied includes analysing data gathered by desk research. It starts with the overview of mobile technology application on retail banking market. The data analysis shows the mobile technology usage in different regions and countries with the special focus on banking and changes of customers' needs and expectations. The next section characterises the market structure and presents the variety of contemporary financial services providers. The last section discusses current trends and tries to answer the question if the implementation of mobile technology on retail banking market means the future of retail banking or just new distribution channel. 3. Results and Discussion Mobile technology application on retail banking market The retail banking is usually defined as a banking that provides services for individual consumers (Harasim, 2005; Klimontowicz, 2013). Retail banks usually offer personal accounts, savings accounts, consumer loans, mortgages and payment services including credit transfers, debit transfers, and payment cards. The term is used to distinguish this banking from investment banking, commercial banking, and wholesale banking. Since the 1990s retail banking market has changed remarkably all over the world. Many external factors have influenced retail banks and customers' market behaviour (see Figure 1). As a result of last economic crisis, the number of new acts, directive, and supervisory recommendations has significantly increased. For many authors, the regulatory pressure is the most important factor influencing not only the retail banking market but the whole economy (Marcinkowska et al., 2014; Kasiewicz and Kurklihski, 2012; Kalicki, 2012). For others, the most important are those that influence banks' ability to create sustainable competitive advantage. Creating, delivering and capturing value for customers is mostly influenced by social and technological factors (Sullivan, 2008; Switalski, 2005, Ernst & Young, 2016). Figure 1 External Factors Influencing Retail Banking Market Economic factors Political factors Social factors Technological factors Legal factors PI economic u development •_• changes in LJ economy n market price u fluctuations periods of growth □ and decline (bull and bear market) □ monetary policy □ fiscal policy PI demographic uchanges customer needs □ and expectations' changes □changes in the life style communication □ technologies' development mobile □ technologies' development changes in □ technological support models □ acts □ resolutions □ directive Source: Klimontowicz, M. (2016), Knowledge as a Foundation of Resilience on Polish Banking Market. The Electronic Journal of Knowledge Management, vol. 14(1), pp. 60-74. The development of telecommunication and information technology has impacted banking products and distribution channels. The history of electronic banking (e-banking, online banking, digital banking) started in the 1980s. The online banking referred to the use of a terminal, keyboard, and television or computer monitor to access one's bank account using a landline telephone (Sarreal, 2016). Today the e-banking means conducting a range of financial transactions through the financial institution's website. Contemporary electronic banking services include mobile internet banking technology. Mobile financial services are among the most promising mobile applications in the developing world. Mobile money could become a universal platform that transforms 323 entire economies, as it is adopted across commerce, health care, agriculture and other sectors (Donovan, 2012). As a result, new kind of banking, called mobile banking (m-banking) has appeared. Mobile banking is a service provided by a bank or other financial institution that allows its customers to conduct a range of financial transactions remotely using a mobile device such as mobile phone or tablet and using software, usually called application (app), provided by the financial institution for the purpose. The type of mobile technology used by retail banks determines the range of products available for customers (see Table 1). Table 1 The Range of Banking Services Enabled by Technology Technology A range of banking services SMS ■ information about the balance on the bank account (Short Messaging Service) ■ information about last few operations ■ money transfers for accounts defined in bank's outlet or via internet access WAP ■ information about the balance on the bank account (Wireless Access Protocol) ■ money transfers ■ bank deposits Lite Website ■ information about the balance on the bank account ■ the operations' history browsing ■ money transfers ■ mobile phone prepaying ■ cards' activation ■ cards' cancelling Mobile Applications based on ■ information about the balance on the bank account JAVA, Android, ■ the operations' history browsing ■ money transfers ■ bank deposit ■ loans' repayment ■ credit cards' service ■ ATM and branches' location browser NFC ■ POS payments (Near Field Communication) ■ P2P payments QR Codes ■ online payments ■ POS payments ■ check sending ■ P2P payments RWD ■ information about the balance on the bank account (Responsive Web Design) ■ online payments ■ mobile phone prepaying ■ saving account ■ deposits ■ credit card repayment ■ taking out a loan ■ cards' order Source: Own work based on Wolna, 2015; Zalewska-Bochenko, 2013, Swiecka, 2007; Sla_zak and Borowski, 2007; Jane and Kotlihski, 2004 The analysis of banking services available by mobile banking shows that the contemporary technology enables offering the broad range of banking products and services and full access to banking products available via providers' websites. The technology potential is quite impressive, but it is only one of determinants of mobile banking market success. The further development of mobile banking depends on its ability to fulfil customer needs and expectations and customers willingness to use it. Customers adoption of mobile banking Mobile banking has become the most important touchpoint, with clients using banking applications an average of 30 times a month (Arnfield, 2016). It is estimated that in 324 2015 weekly mobile bankers exceed the number of weekly branch bankers for the first time (Van Dyke, 2015) - see Figure 2. Figure 2 Mobile Bankers vs. Branch Bankers 2010 2011 2012 2013 2014 2015 Mobile Banking Branch Banking Source: Van Dyke, 2015. Mobile banking development is strictly correlated with m-commerce and should be considered as a part of customer value chain. The M-commerce share in e-commerce differs across countries (see Figure 3). Estimates suggest that it will be growing very fast in the nearest future. More than 20 percent of customers are already said to shop on mobiles (EBA, 2014). As the overall e-commerce market near doubles from a value of €755 billion in 2014 to an expected €1,460 billion in 2015, mobile payments will account for a significant share of that growth. Undoubtedly, it is correlated with increasing usage of smartphones. Today there are 2,6 billion smartphone subscriptions globally. It is forecasted that by 2020 the number of smartphone users will reach 6,1 billion. Mobile phones haves increasingly become tools that consumers use for banking, payments, budgeting and shopping (Lunden, 2015). As a result one of the fastest growing mobile banking services is mobile payments. Their share in e-payments is still low, but their value and number are systematically increasing (Worldpay, 2014). Figure 3 M-commerce as a Percentage of E-commerce by Selected Countries Source: Worldpay, 2014. The willingness of using mobile banking depends on customer needs and expectations. Thus, the generation is the next factor influencing mobile banking adoption (see Table 2). 325 Table 2 Characteristics of Financial Needs and Expectations across Generations Generation Characteristics Generation X: ■ prefer personal interactions and communication born between ■ used to ingrained status quo banking 60's to 80's, ■ some find their bank's existing mobile interface difficult to aged 35-55 work with ■ display some willingness to adopt mobile banking but they lack the enthusiasm of next generations ■ security concerns concerning new technology ■ declare the openness to the idea of mobile banking Generation Y: ■ commonly use communication via telephones and the born between 80's internet to 90', ■ active on social networking sites aged 25-35 ■ a half of them would switch banks to mobile payments capability from their primary bank Generation Z: ■ use mostly mobile devices as smartphones, tablets and born after 1995, laptops for communication aged 21-25 ■ expect everything to be digital ■ expect seamless cross-channel customer service Source: Own work based on ATM Marketplace, Crittercism, 2015, Arnfield, 2015, Rozzo, 2016 The representatives of Generation X still prefer in-person branch banking and face-to-face consulting service, but they are open to online and mobile banking. However, they will expect a high degree of personalization in their digital banking experience. The generation Y and Z adopt mobile banking easily. Especially, as a transaction-rich segment, Generation Z should be treated as crucial customer segments for mobile banking (Arnfield, 2015, Arnfield, 2016). Despite the generation, customers have a core set of requirements that are unlikely to change over the next few years. Among them are: simplicity, mobility, free or low costs, security, real-time immediacy, flexibility and choice, preferences specialization and refunds (EBA, 2014). The rapid adoption of technology by consumers is changing their needs and the way they interact with banks (CFSA, 2015). Mobile channels will probably become the "first screen" through which customers interact with a bank. Banks will need to innovate to meet customers' expectations and compete with nonbank institutions. Entering the market by FinTech startups and other vendors creates more options for customers. Banks should offer mobile apps different from every other mobile banking apps so as to keep the customer engaged and provide them an experience similar to that of using the Amazon, Google or Apple apps (CFSA, 2015, Arnfield, 2016). The characteristic of mobile banking market The mobile banking is not just usage of mobile devices and technology. It needs a cash-in, cash-out infrastructure, usually accomplished through a network of cash merchants (or agents), who receive a small commission for turning cash into electronic value (and vice versa). Because the mobile money industry exists at the intersection of finance and telecommunications, it has a diverse set of stakeholders, with players from different fields in competition (Donovan, 2012). The mobile banking market's structure include traditional financial institutions as banks, mobile network operators, and new market players as payment card firms, FinTech companies, and other vendors. Traditional financial institutions usually are quite conservative ones. Their size, organizational structure and culture, the necessity of fulfilling all legal regulations and recommendations and other features make the flexible reaction to fast changing customers' expectations difficult. As banks have been slow to respond these changes, some nonbanks players have grabbed the opportunity to meet rapidly evolving customer needs. The threat of new entrants is looming and expected to accelerate (CFSA, 2015). Nonbanks are targeting the most profitable retail banking market segments now such as payments, personal finance management, lending, investments and core banking. The state of European online alternative finance available is strong. In 2014 the total transaction volume was 2957 million euro. It means an impressive average year growth 326 of 146% comparing with 2013. The United Kingdom is an innovative leader in Europe and dominates the European market with some of the most advanced online platforms and instruments. Its transaction volume constitutes 79% of the European market. Among other markets the most developed are France, Germany and Sweden (Baeck et al., 2014, Wardrop et al., 2015, Agnew, 2015) -see Figure 4. Figure 4 The Alternative Finance Transactions' Value in Selected Countries 160,0 154'° 140,0 120,0 100,0 80,0 60,0 40,0 20,0 0,0 140,0 10 7,0 78,0 6 2, 0 22'° 17 0 ' 12 0 | 8^2 4^0 3,6 2,5 2,5 2,0 1,0 *** 4 & 4* £ # Source: Own work based on Wardrop (2015). New market entrants have already made banks to face a challenge to convince their customers that their products and service is still valuable for them. It causes the necessity to continue creating innovative offerings. It is especially important concerning new generations entering the market. According to Norcross' survey (2016), Millennial are open to using non-traditional organisations for banking services (see Figure 5). They are focused on their needs and choose the providers whose products meet them in the best way even if it means abandoning traditional suppliers. Figure 5 The Millennial' Willingness to Use Non-traditional Financial Services Providers Amazon Google PayPal Apple Walmart Facebook Verizon AT&T Starbucks Twitter Square 37 34 31 26 24 24 24 22 19 16 40 10 15 20 25 Source: Norcros (2016). 30 35 40 45 327 The changes in mobile banking market structure and customer openness for changing banking service provider remain a challenge for retail banks. The solution might be the engagement with emerging FinTech companies and view them rather as potential partners than competitors (BNY Mellon, 2015). 4 Conclusions Undoubtedly, the mobile technology development is one of the most significant drivers of change on retail banking market. It allows creating new distribution channels and new ways of transactions' acceptance. They follow changes in consumers' behaviour including demand for multichannel platforms. As a result, traditional banking services' suppliers start to become just one of the possible option for customers. Today they have to compete with non-bank companies. Among customers' decision factors user-friendly smartphones' applications are crucial. They are changing the way customers utilise banking services. The mobile applications enable using the different kind of financial services and access to full range of product available by bank's website. They allow evolving customer expectations by using innovations concerning both, the products and process of serving them. The attitude to mobile banking and mobile technology adoption differs across generations. Today, as Generation X is the most profitable customer segment banks can still focus on traditional distribution channel but electronic channels as a layer for them will not be enough in the nearest future. Banks need to prioritize and critically address the areas of digitalisation, simplification, agility, insights and data to enhance the ability to provide enriching customer experience. Retail banking market growth opportunities and profit margin for banks are declining due to industry fragmentation. Nonbanks are targeting the most profitable retail banking market segments now such as payments, personal finance management, lending, investments and core banking. Some of them will surely become crucial market players and strong competitors for traditional financial institutions. Concerning the generation Z openness for switching financial services providers, it should be treated as a threat. Meeting all these challenges might require a co-opetition with the FinTech companies. Trends presented in the paper show that mobile technology will probably be the future of banking. Its further development needs implementing innovations. Concurrently, banks' sustainable development require meeting these innovations with sufficient demand from customers. A host of supporting businesses, such as agents and liquidity management firms, are also necessary. Finally, all of this must happen in an environment with appropriate government regulations, as well as adequate safeguards for consumer protection. References Agnew, H. European market for online alternative finance surges. Retrieved from: www.ft.com/cms/s/0/5c61bfca-baad-lle4-8447-00144feab7de.html#axzz3XPDpXY10. Arnfield, R. (2015). The Customer's Journey: Transforming the Branch Network. ATM Marketplace. Networld Media Group, pp. 2-7. Arnfield, R. (2016). Transforming the bank customer experience for the digital revolution. ATM Marketplace. Networld Media Group, pp. 3-9. Baeck, P., Collins, L, Zhang, B. (2014). Understanding Alternative Finance. The UK Alternative Finance Industry Report 2014. Nesta, London 2014, pp. 21-22. BNY Mellon (2015). Innovations in Payments: The Future is Fintech. The Bank of New York Mellon Corporation, pp. 13-17. CFSA (2015). Top 10 Trends in Banking in 2016. Capgemini Financial Services Analysis, pp. 4-5. 328 Dolphin, T. (2012). Don't bank on it. The financialisation of UK economy. Report of Institute for Public Policy Research. London. Donovan, K. (2012). Mobile Money for Financial Inclusion. In: Maximizing Mobile. Information and Communication for Development, pp. 61-74. EBA (2014). Opinion Paper on Next Generation Alternative Retail Payment (e-AP) User Requirements. Euro Banking Association, pp. 4-9. Epstein, G. (2005). Introduction: Financialization and the world economy. In: Epstein, G., ed., Financialization and the world economy. Cheltenham: Edward Elgar Publishing, pp. 3-16. Ernst & Young (2016). Building the bank of 2030 and beyond The themes that will shape it, pp. 21-32. Retrieved from: www.ey.com. Freeman, R. (2010). It's financialisation! International Labour Review, vol. 149(2), pp. 163-183. Harasim, J. (2004). Strategie marketingowe w osiajganiu przewagi konkurencyjnej w bankowosci detalicznej (Marketing strategy in gaining competitive advantage in retail banking). Wydawnictwo Akademii Ekonomicznej w Katowicach, Katowice, pp. 22-24. Jane, A., Kotliňski, G. (2004). Nowe technologie we wspóiczesnym banku. Wydawnictwo Akademii Ekonomicznej w Poznaniu, Poznaň, pp. 335. Kaisewicz, S., Kurkliňski, L. (2012). Szok regulacyjny a konkurencyjnošč i rozwój sektora bankowego, Warszawski Instytut Bankowošci, Warszawa. Kalicki, K. (2012). Wptyw Bazylei III/CRD4 na sytuacje. sektora bankowego. Retrieved from: www.alterum.pl/pdf/Dr_hab_Krzysztof_Kalicki.pdf. Klimontowicz, M. (2013). Aktywa niematerialne jako žródio przewagi konkurencyjnej banku (Intangible assets as source of bank's competitive advantage). Wydawnictwo CeDeWu, Warszawa, pp. 137. Klimontowicz, M. (2016). Knowledge as a Foundation of Resilience on Polish Banking Market. The Electronic Journal of Knowledge Management, vol. 14(1), pp. 60-74. Lunden, I. (2015). 6.IB Smartphone Users Globally By 2020, Overtaking Basic Fixed Phone Subscribers. Retrieved from: http://techcrunch.com/ 2015/06/02/6-1 b-smartphone-users-globally-by-2020-overta king-basic-fixed-phone-subscriptions/. Marcinkowska, M., Wdowiňski, P., Flejterski, S., Bukowski, S., Zygierewicz, M. (2014). Wptyw regulacji sektora bankowego na wzrost gospodarczy - wnioski dla Polski, Materiály i Studia NBP nr 305. Warszawa, pp. 35-52. Montgomerie, J. (2008). Bridging the critical divide: global finance, financialisation and contemporary capitalism. Contemporary Politics, vol. 14(3), pp. 233-252. Norcross (2016). Millennials: Financial Insights. Georgia-based Synergistics Research. Rizzo, T. (2016). Extending the Mobile Enterprise. Building Effective & Profitable M-Commerce Apps. Blue Hill Research and Apteligent, pp. 2-9. Sarreal, R. (2016). History of Online Banking: How Internet Banking Became Mainstream. Retrieved from: http://www.gobankingrates.com/banking/history-online-banking/. Šla_zak, E., Borowski, K. (2007). Bankowošč elektroniczna. In: Zaleska, M., ed., Wspólczesna bankowošč. Difin, Warszawa, pp. 247-249. Sullivan, M. B. (2009). Post Crisis, Innovation Will Rule. ABA Banking Journal, pp. 30-32. Šwiecka, B. (2007). Bankowošč elektroniczna. CeDeWu: Warszawa, pp.57-59. Šwitalski, W. (2005). Innowacje i konkurencyjnošč. Wydawnictwo Uniwersytetu Warszawskiego, Warszawa, pp. 78-81. 329 Van Dyke, D. (2015). Mobile Banking, Smartphone, and Tablet Forecast, Javelin Strategy & Research. Retrieved from: www.javelinstrategy.com/node/22711#sthash.8dgLC QUX.dpuf. Wardrop, R., Zhang, B., Rau, R., Gray, M. (2015). Moving Mainstream. The European Alternative Finance Benchmarking Report. University of Cambridge, London, pp. 15-16. Wolna, J. (2015). Rozwöj systemöw platnosci mobilnych w Polsce, Studia ekonomiczne, Zeszyty naukowe239 Uniwersytetu Ekonomicznego w Katowicach. Wydawnictwo Uniwersytetu Ekonomicznego w Katowicach, Katowice, pp. 169. Worldpay (2014). Your Global Guide to Alternative Payments, London. Zalewska-Bochenko, A. (2013). Bankowosc telefoniczna i jej wplyw na bankowosc elektroniczna.. In: Pyka, I., Cichy, 1, ed., Innowacje w bankowosci i finansach. Studia ekonomiczne, Zeszyty Naukowe Wydziatowe 173. Wydawnictwo Uniwersytetu Ekonomiczny w Katowicach, Katowice, pp. 301. 330 The Role of Accounting Policy in Management of Polish Hospitals Magdalena Kludacz-Alessandri Warsaw University of Technology The College of Economics and Social Sciences tukasiewicza 17, 09-400 Plock, Poland E-mail: m.kludacz@pw.plock.pl Abstract: The Polish accounting Act is not able to create an ideal accounting model for all organizations which carry out diversified activities in various economic conditions. Therefore, managers have the right to choose their own legally accepted accounting principles, in order to present a true and fair view of their financial results in financial statements. However, the freedom to choose accounting principles may also lead to use such tools of accounting policy that have a direct impact on the manner of recognizing revenues and costs and allow the creation of financial performance and other elements of financial statements. The aim of the article is to describe the essence of the accounting policy and analyze the use of selected instruments of material accounting policies in management of Polish hospitals that have impact on financial results and other components included in financial statements. The article presents the results of an empirical survey conducted in selected Polish hospitals in the years 2012 - 2013. The respondents gave information regarding various instruments of accounting policy used in their hospitals and evaluated their use in creating financial performance. The study showed that accounting policy is not the essential tool in the management of the hospitals in Poland. From one point it is good because it means that accounting policy isn't the tool of any manipulative practices in hospitals. On the other hand the managers don't use their possibilities and rights to choose the accounting methods to achieve the most desirable financial results. They don't try to optimize the financial statements of their hospitals using the instruments of accounting policy which don't violate the balance sheet law. Keywords: accounting policy, hospitals, accounting rules, balance sheet policy JEL codes: M41, M48, 118 1 Introduction The Polish balance sheet law is not able to create an ideal accounting model for all entities which carry out diversified activities in different economic conditions. Therefore, management staff of each entity have the right to choose their own legally accepted accounting principles in order to present a reliable picture of their financial condition in financial statements. On the other hand they have to be careful, not to use this freedom to choose such accounting principles that lead to creating an unreliable image and presenting a false picture of the condition of entity (Ste_pieri, 2012). Sometimes it happens that that accounting policy is the hub of many manipulative practices. According to the Polish Accounting Act also hospitals have a free hand in choosing the method of valuating and accounting for assets and ongoing economic operations. Those methods are a constituent part of hospital's accounting policy. If a hospital is given a free hand to choose its accounting methods, it may adjust them in such a way as to achieve the most desirable financial results. Such individually chosen methods are considered correct if they operate within the boundaries set by the provisions of the law. According to the Polish Accounting Act the choice of valuation methods of assets and liabilities depend on their conformity with the balance sheet law, it means whether they are provided for in this Act. According to IAS 1, accounting policy is defined as a set of rules and methods for estimating conventional values, and procedures regarding the presentation of financial statements (IAS1 -Presentation Of Financial Statements). The principles adopted by hospital as part of its accounting policy depend on the following factors: 331 1) objectives set by the entity's owner, 2) conformity with the current legal interpretation of the balance sheet law and the adopted standards of accounting (Michalczyk, 2012). The accounting policy comprises general principles, methods and features of accounting designed for the organization of accounting and financial accountability of the economic entity. Main aspects of accounting policy are organizational aspects, technical aspects and methodical aspects (Adomaitiene, 2004). The accounting policy is an important element of the entity, and its use is possible thanks to the fact that the legislation does not regulate all possible issues and allow to choose and use permitted by law various alternatives in accounting system. In this context, the accounting policy is often referred to as a balance sheet policy. There is no agreement among authors, whether these terms can be identified and used interchangeably. In Polish accounting literature the terms "balance sheet policy" and "accounting policy" are commonly used. The problem is that various authors have various opinions regarding their meaning and their relations with each other. According to many opinions accounting policy and balance sheet policy represent the same meaning and scope (Grabowski, 2010). In other opinions balance sheet policy is only a part of accounting policy because it does not cover actions taken to achieve the objectives of the entity and its owners and it doesn't influence the recipients of financial statements. However the professional literature is dominated by the opinions that the two meanings can not represent the same issues because the accounting policy is a broader concept. For instance the issues of accounting policy include not only issues related to the creation process of reporting information, but also the technical elements of accounting, which does not affect the image of the financial statements (Kaczmarek, 2008). The discussion over the differences between accounting policy and balance policy is also reflected in the geographical definitions. Accounting policy refers to the accounting systems of the Anglo-Saxon tradition, while balance sheet policy is more closely related to German accounting. However the two terms most often are seen as synonymous in everyday accounting practice For the purposes of this study, it was assumed that the term of accounting policy will be understood not only as a set of rules, methods, solutions, principles and procedures of accounting adopted by the hospital in the framework of the law, but will also include all activities, arrangements and procedures related to the creation of financial information to achieve the objectives of the entity and its owners. The article presents the part of results of an empirical study relating to the instruments of accounting policies used in Polish hospitals. The author conducted a survey regarding cost accounting used in hospitals (Kludacz, 2015). One part of this survey included the questions regarding accounting policy. The purpose of this part was to examine the extent to which Polish hospitals use instruments of accounting policy. The empirical study was used to verify the hypothesis that Polish hospitals do not apply any instruments of accounting policy , or apply them in a limited scope. 2 Methodology and Data The study involved managers of hospitals who responded to questions regarding the use of selected instruments of material accounting policies in shaping the information content of financial statements for the purpose of achieving the objectives of the entity. The survey was conducted in 2012-2013. In the survey group were mainly public hospitals, however, two questionnaires were sent back by two non-public hospitals. The request for participation in the study was addressed to 100 Polish hospitals from the ranking list "Safe Hospital", which are characterized by high quality of performed health services. The questionnaire was sent by e-mail together with a cover letter. The completed questionnaires were collected mainly by e-mail and via the website with electronic version of questionnaire. Replies were received from 30 Polish hospitals. The 332 research materials were interpreted and analysed using descriptive statistics. Most of the replies came from the directors, deputy directors or chief accountants. The first part of the questionnaire regarded the characteristics of the hospital. Respondents answered the questions on the hospital status, the type of the founding body, the level of employment, the number of hospital beds and wards, education and experience of a hospital manager. As previously mentioned, the hospitals that participated in this study were mainly public hospitals - their representatives has sent back 28 questionnaires. Most of respondents has represented provincial hospitals (about 53%), whose founding body is the marshal office or the provincial government. The remaining group consisted of the hospitals controlled by the counties and city councils. The non-public hospitals were conducted in the form of a civil partnership. The largest share of surveyed hospitals (40%) is located in medium-sized towns with a population between 51,000 - 150,000. Other hospitals are located in the towns with a population: under 50,000 - 27%, between 151,000 -300, 0000 - 20%, above 301,000 - 13%. The majority of surveyed hospitals employed more than 50 physicians (28 units and half of them employed more than 151 physicians), more than 150 nurses (22 units, and 12 hospitals employed more than 300 nurses) and more than 50 non-medical employees (28 units and half of them employed 50-150 people). The two smallest hospitals taking part in the survey employed less than 50 doctors, nurses and non-medical staff. All hospitals differed from each other mainly in terms of size. The largest group of surveyed hospitals were large units, with more than 19 wards (47%) and 500 beds (53%). The two smallest units involved in the study, had less than 100 beds and less than 6 wards. Directors of most of the hospitals (67%) have an university degree in economics. On the other hand, 20% of hospitals is managed by directors of medical education. The average period of experience as director of the hospital was just over 11 years. The degree of computerization of economical part of a hospital was well or very well assessed by all respondents and 87% of them gave the same assessment to the degree of computerization of medical part. On the other hand, the degree of software integration regarding the medical and administrative part of a hospital was not sufficient according to almost half of respondents. It is worth noting that the lack of integration of the administrative and medical software can generate problems with data access. It can be concluded that the average hospital included in the survey is represented by large public hospital, that has over 500 beds, and over 19 wards, is located in medium-sized town and is managed by the director of economic education with considerable experience. 3 Results and Discussion The aim of second part of the questionnaire was to analyze the use of selected instruments of accounting policies in Polish hospitals that have impact on financial results and other components included in financial statements. The aim of the instruments of accounting policy which follows also the definition of balance sheet policy is to achieve a certain level of information in the financial statement and obtain expected assessment of this information by the recipients of financial statement and other reports, as well as to induce them to perform the desired behavior. For instance the values presented in the expenses reports of hospital can be used in negotiations with a payer of health care 333 services (Cygahska, Gierusz, 2007). Such instruments typically involve operations affecting: 1) the financial result - maximising or minimising the reported financial performance, 2) the goodwill of a hospital- adjusting it to the strategic objectives in management process (Michalczyk, 2012), 3) financial liquidity and other financial ratios used in the assessment of the entity. In the private entities they affect also the tax result. For instance the aim of accounting policies can be to maximize the financial result. In such situation the entity will select such instruments of accounting policies that concern the rules that lead to lower costs in the income statements (Gurau, 2014). The examples of instruments of accounting policy that aim to minimize costs are as follows: choice of straight-line instead of degressive depreciation method to not increase the costs in the first period of asset use; decisions regarding the expenses related to renew fixed assets; if they are above the limits of repair and maintenance costs, they will be recognized as a modernization costs and activated in fixed assets; purchase of items which have an input value less than the limit established by legislation (3500 PLN in Poland.); in such case they won't be recognized as a component of fixed assets and they will be recognized as expenses; choice of output costing inventory method first in - first out (FIFO), which can ensure lower costs; it is worth noting that material costs are one of very important indicators determining optimal solution for maximising profit (Walczak, 2014). use of the right to create prepayments. The basis for defining the instruments of the accounting policy is the right to choose between various accounting solutions. There are many instruments of accounting policy influencing on financial reports of hospitals that are used to achieve the objectives of entity. Most frequently they are divided into: material instruments; time instruments; formal instruments (Weber, Kufel, 1993). The second part of the questionnaire was designed to find out which instruments of material accounting policy are used by Polish hospitals and in which scope. Analysis was conducted in two areas, namely taking into account the distribution of instruments of material accounting policy on (Kamihski, 2001): shifting the economic operations in time e.g. purchases of materials, goods, capital goods, services, or sale of services. using the right to choose the method of valuation of assets and liabilities, revenues and expenses, gains and losses. The first kind of instruments may include activities that shift business operations, such as the choice of the moment of the planned purchases of tangible fixed assets, goods and materials, postponement of projects entailing the current costs (e.g. external services), undertaking of the projects based on off-balance sheet financing (e.g. use of rented or leased assets). Most of these solutions can be economically justified (as they may, for instance, help to shifting in time the tax liabilities). The answers of respondents regarding the selected instruments of material accounting policies shifting the economic operations in time are presented in Table 1. 334 Table 1 Instruments of Accounting Policies Shifting the Economic Operations in Time The scale of assessment Instruments 1 2 3 4 5 shifting in time maintenance and repairs of fixed assets in relation to the balance sheet date 18 4 8 - - making modernization of fixed assets instead of maintenance and repairs 14 6 2 6 2 shifting in time the purchase of low-value items that are charged as expenses 12 6 6 4 2 shifting in time the sales of services in relation to the balance sheet date 17 8 4 - 1 shifting in time the procurement of materials and goods 12 5 9 2 2 * the scale of asessment: 1 (definitely not applicable) - 5 (definitely applicable) Source: Own work The results show that Polish hospitals do not use properly selected instruments of accounting policy shifting the economic operations in time. In most cases (according to 40 - 60% of respondents), the assessment of the use of these instruments to shape the information contained in the financial statements , in order to achieve specific objectives, was insufficient. The most popular are instruments postponing current costs to future periods, such as: making modernization of fixed assets instead of maintenance and repairs - according to 27% of directors the use of this instrument is good or very good; shifting in time the purchase of low-value items that are charged as expenses -according to 20% of directors the use of this instrument is good or very good. The second kind of instruments may include activities that use the right to choose the methods of assessing individual items of the financial statements, such as assets and liabilities, revenues and expenses, gains and losses. Such possibility exists because various kind of assets might be valued at the various costs e.g. at cost of purchase, the cost of acquisition, the net selling price, or the cost of manufacture. The choice of valuation rules depends on whether the valuation is made at the balance sheet date or before it (Czubakowska, 2009). The rights to choose occur when a certain valuation problem may be solved in at least two different ways. Each hospital specifies the selection of selected principles and methods in its accounting policy. They also have to choose the methods regarding: 1) calculation of depreciation value (Michalczyk, 2007), 2) creating, calculating and releasing provisions, 3) calculating prepayments and accruals. They have the right to choose the date of cost settlement over time, the right to choose the manner in which to account for the provisions, reserves, interim settlements, amortization and depreciation write-offs and revaluation amounts (Kamihski, 2012). The answers of respondents regarding the selected instruments of material accounting policies using the freedom of choice with respect to valuation methods, are presented in Table 2. 335 Table 2 Instruments of Accounting Policies Using the Choose Rights of Valuation Methods Instruments The scale of assessment 1 2 3 4 5 the right to choose the methods of assessing individual items of the financial 12 4 12 2 statements (e.g. intangible fixed assets) the right to choose the depreciation method, including the right to choose a one-time 14 4 6 4 2 depreciation low value fixed assets The right to choose the type of inventory records and valuation methods of materials 14 6 6 4 expenses Using the right to create write-offs of assets 16 -) 8 A (fixed assets, inventories, receivables) A. H Using the right to create accruals and f; 8 f; 4 prepayments and accrued revenues * the scale of asessment: 1 (definitely not applicable) - 5 (definitely applicable) Source: Own work Again, the results show that most of Polish hospitals do not use instruments of accounting policy using the right to choose the method of valuation of assets and liabilities, revenues and expenses. According to 40-53% of respondents, the assessment of the degree of use of such instruments is insufficient. Unfortunately, only few hospitals are interested in these instruments of material accounting policies, that have the greatest impact on the image of the entity, so the methods of valuation and depreciation that directly affect the value of assets and the financial results presented in the financial statements. The most popular instrument is the right to create prepayments (activate costs) and accruals and deferred revenues. According to 33% of directors, the use of this instrument is good or very good. Prepayments are created when a hospital pays for services in advance but has not yet received them. This might be something like paying for contracted services which has not been performed yet. They are treated as an asset because they are effectively owed to the hospital. Accruals are created when a revenue or expense has not been recorded at the end of the accounting period. Accruals are normally the result of revenue being earned or an expense being incurred before any cash is received or paid (Maxwell, 2007). 4 Conclusions The main goal of this study was to analyze the use of selected instruments of material accounting policies in Polish hospitals that have impact on financial results and other components included in financial statements. This article presented also an analysis of the meaning and scope the accounting policies and provided explanation of existing opinions regarding differences between "accounting policy" and "balance sheet policy". The aim of using the instruments of accounting policy is to facilitate the attainment of its economic objectives and influence the recipients of financial statements so as to encourage behaviour conforming to the entity's expectations (Sawicki, 1998). The study confirmed the hypothesis according to which the hospitals operating in Poland do not apply any instruments of accounting policy to support management processes, or apply them in a limited scope. The information contained in financial statements of various hospitals largely derives from their accounting policy, but the shape of this policy is not determined by the hospital's objectives. The results indicate that most of the analysed hospitals apply insufficient instruments of material accounting policy that have the greatest impact on the image of the hospital. In this context it can be concluded that the ability of these hospitals to shape the 336 information content of financial statements for the purpose of achieving the hospital objectives is considerably limited. Most of hospitals use the same accounting policy for long period of time, it means the same methods of measurement and presentation of financial statement information. The directors are not interested at all to change their accounting methods at the start of the next accounting period and to use other instruments of accounting policy. Most of them accept the solutions that can be considered as the most popular, least complicated and comply with the law. Whatever the impact that the instruments of accounting policy have on the performance of hospital presented in financial statement, the necessity of assurance of continuity and consistency in the application of accounting policies has to be taken into consideration. This is necessary for the financial statements to provide true and fair view of financial position, result and changes in financial position (Baran, Stanisz, 2005). References Adomaitiené, G. (2004). Methodological Aspects of Accounting in Lithuanian Companies. Acta Universitatis Lodziensis. Folia Oeconomica, vol. 173, pp. 5-13. Baran, W., Stanisz, B. (2005). Wewne_trzne mechanizmy ksztaftowania wiarygodnošci finansowej. In: Sprawozdawczošč i rewizja finansowa w procesie poprawy bezpieczenstwa obrotu gospodarczego. Centrum Promocji i Rozwoju Akadémii Ekonomicznej w Krakowie, Kraków, pp. 12-18. Czubakowska, K. (2009). Wybór metód wyceny zapasów w polityce bilansowej. Zeszyty Teoretyczne Rachunkowošci, vol. 49 (105), pp. 31-40. Cygaňska, M., Gierusz, J. (2007). Ewolucja zásad kontraktowania usfug zdrowotnych w Polsce na tie wybranych krajów. Prace i Materiály Wydziaiu Zarzadzania Uniwersytetu Gdaňskiego, vol. 4, pp. 15-27. Goyal, S. K. (2007). An integrated inventory model for a single supplier-single customer problem. International Journal of Production Research, vol. 15(7), pp. 107-101. Grabowski, R. (2010). Relacje pomie_dzy polityka. bilansowa. a polityka. rachunkowošci w šwietle rôznych modeli rachunkowošci. Wspóiczesna Ekonómia, vol. 4(4), pp. 175-185. Gurau, M. (2014). Three Types of Accounting Policies Reflected in Financial Statements. Case Study for Romania. Global Economic Observer, vol. 2(1), pp. 209. IAS 1. Presentation of financial statements. IASB, 2014. Kaczmarek, M. (2008). Polityka bilansowa jako narzedzie rachunkowošci. Wydawnictwo Naukowe Uniwersytetu Szczeciňskiego, Szczecin. Kaminski, R. (2012). Investor Relations as a Determinant of the Company's Accounting Policy. China-USA Business Review, vol. 11(5), pp. 686-696. Kamiňski, R. (2001). Polityka bilansowa w ksztaitowaniu wartošci ksiegowej przedsiebiorstwa. Wydawnictwo Naukowe UAM, Poznaň. Kludacz, M. (2015). Rachunek kosztów i jego wykorzystanie w zarzadzaniu szpitalem. Research Papers of the Wroclaw University of Economics/Prace Naukowe Uniwersytetu Ekonomicznego we Wroclawiu, vol. 389, pp. 60-171. Maxwell, 1 C. (2007). The 21 irrefutable laws of leadership: Follow them and people will follow you. Thomas Nelson Inc. Michalczyk, L. (2007). Rola umorzenia aktywów trwafych w powstaniu róžnic w wyniku finansowym. Zeszyty Naukowe TD UJ. Séria: Ekonómia i Zarzadzanie, vol. 1. Michalczyk, L. (2012). The Role of Accounting Engineering in Shaping the Balance Policy of a Company. e-Finanse, vol. 8(4), pp. 44-52. 337 Polish Accounting Act - Ustawa z dn. 29 wrzesnia 1994 r. o rachunkowosci (Dz.U. 2013 poz. 330). Sawicki, K. (1998). Wybrane zagadnienia polityki bilansowej. Zeszyty Teoretyczne Rady Naukowej SKwP, t. 49, Warszawa. Ste_pieri, K. (2012). Polityka rachunkowosci jako instrument kreowania wizerunku przedsie_biorstwa w okresie kryzysu gospodarczego. Zeszyty Naukowe/Polskie Towarzystwo Ekonomiczne, (13), pp. 289-298. Walczak, R. (2014). Production management analysis using Monte Carlo method. Journal of International Scientific Publications: Economy & Business , pp. 1062-1069. Weber, 1, Kufel, M. (1993). Wprowadzenie do rachunkowosci spölek: Bilansowanie majatku i kapitalöw: Poland: Bielsko-Biala, Park. 338 Revenue Efficiency in European Banking Kristina Kočišova1 technical University of Košice Faculty of Economics, Department of Banking and Investments Nemcovej 32, 04001 Košice, Slovakia E- ma i I: Kri sti na. Koci sova @tu ke.sk Abstract: This paper analyses revenue efficiency of banking sectors in the European Union countries over the period 2008-2014. The Data Envelopment Analysis (DEA) method is applied. The results show decrease of the average revenue efficiency in the whole sample. The results of DEA analysis, by country, indicate that the efficiency ranged from 52.47% in case of Romania to 100% in case of Germany, United Kingdom, Netherlands, Sweden, Malta, and Luxembourg. The revenue efficiency was also analyzed in three groups of banking sectors, classified according to the volume of total assets. Large banking sectors seem to be most efficient, where the average revenue efficiency during the whole analyzed period was 97.11%. On the second place, there were banking sectors within the medium-sized group (85.3%) and the least efficient were banking sectors in the small sized group (71.6%). In the last part of our analysis there were determined four main European "regions" and the average revenue efficiency was analyzed within them. According to the results of analysis, levels of efficiency in case of Northern (89.4%) and Western European (89.5%) banking sectors were higher than the average in the whole sample (79.8%); on the other hand, the average revenue efficiencies in case of Southern (76.7%) and Eastern European (63.5%) banking sectors were under the total average. Keywords: banking sector, revenue efficiency, DEA JEL codes: G21, C14, C6 1 Introduction The efficiency of banks and banking systems is one of the most important issues in the financial market, as the efficiency of banks ultimately affect the effectiveness of a whole monetary system. In modern society, there exists a number of approaches how to define efficiency. Our definition is based on the study of Farrell (1957), who stated that the efficiency of a firm consists of two components: technical efficiency and allocative efficiency. Technical efficiency reflects the ability of a firm to obtain maximal output from a given set of inputs. On the other hand, allocative efficiency reflects the ability of a firm to use the inputs in optimal proportions, given their prices and the production technology. These two types of efficiency are then combined into an overall economic efficiency, which can be examined from the perspective of input or output based models. Then, we can talk about overall cost efficiency (input perspective) or overall revenue efficiency (output perspective). Farrell's (1957) paper led to the development of many approaches of measuring the input and output efficiency. Greatest importance was assigned to a Stochastic Frontier Approach (SFA), created by Aigner, Lovell and Schmidt (1977); and Data Envelopment Analysis (DEA) developed by Charnes, Cooper and Rhodes(1978). According Bader et al. (2008) commercial banks make profits from the spread between the interest rate received from borrowers and interest rate paid to depositors. Profit efficiency than indicates how well a bank is predicted to perform in terms of profit in relation to other banks in the same period for producing the same set of outputs. We can also define cost efficiency and revenue efficiency. Cost efficiency gives a measure of how close a bank's cost is to what a best-practice bank's cost would be for producing the same bundle of output under the same conditions. Revenue efficiency indicates how well a bank is predicted to perform in terms of revenue relative to other banks in the same period for producing the same set of outputs. 339 This paper deals with DEA method and describes its application in measuring revenue efficiency. The structure of the paper is as follows. The review of relevant literature is described in section 2. Used methodology is discussed in detail in section 3. Section 4 contains the practical application of DEA method for measuring revenue efficiency in European banking sector during years 2008-2014 using the R software. Finally, the paper ends with some concluding remarks. Literature Review Data envelopment analysis (DEA) is a non-parametric mathematical (linear) programming approach to frontier estimation. The basic DEA model developed by Charnes, Cooper and Rhodes (1978) was based on the assumption of a constant return to scale. This basic model has been modified by Banker, Charnes, and Cooper (1984) and based on the assumption of a variable return to scale. Both these DEA models have been created in both forms - the input and output oriented. Sherman and Gold (1985) applied DEA to banking as the first. They used DEA analysis to evaluate technical efficiency of 14 saving bank branches. As the result of the analysis, they not only measured the level of efficiency but also defined how to eliminate inefficiency by adjusting input and output of inefficient bank branches. Motivated by the DEA results, management indicated that the service outputs and the resources used to provide these would be further evaluated as distinct from the liquidity issues. Most studies have focused on the input side, estimating cost efficiency. Only few studies have examined the output side evaluating revenue and profit efficiency, but they preferred to use SFA methodology. Bos and Kolari (2005) employed SFA to measure cost and profit efficiency of large European and U.S. banks between the 1995 and 1999. They found out that the large U.S. banks had higher average profit efficiency than European banks on average. They applied intermediation approach. As the bank outputs were defined loans, investments and off-balance sheet activities. As the input variables were used number of employees, interest expenses and operating expenses. They conclude that their empirical results tend to support the notion that potential profit efficiency gains are possible in cross-Atlantic bank mergers between European and U.S. banks. Rouissi (2011) investigated the efficiency levels of commercial domestic versus foreign banks in France during the period 2000-2007 by comparing the use of the stochastic cost and profit frontier analysis. He found out that foreign banks exhibit higher cost and profit efficiency than domestic banks. Analysis of the determinants of banking efficiency in France suggested that revenue efficiency was more important than cost efficiency for domestic banks. In the case of authors from the Slovakia as well as from the Czech Republic, efficiency of financial institutions was examined for example in works of Stavárek (2006), Pančurová and Lyócsa (2013), Řepková (2013), Boďa and Zimková (2015), and Zimková (2015). Pančurová and Lyócsa (2013) measured bank cost and revenue efficiencies using DEA. They estimated efficiencies and their determinants for a sample of 11 Central and Eastern European Countries over the 2005-2008 period. They adopted the intermediation approach and assumed that banks produce two outputs: total loans and other earning assets. The prices of those outputs were represented by the ratios of interest received on loans to total performing loans and noninterest income to other earning assets, respectively. Total deposits and total costs represented the two inputs. The prices of those inputs were total interest expenses to total deposits and total costs to total assets, respectively. They found out no dramatic changes in the average cost and revenue efficiencies during the analyzed period, although cost efficiency declined slightly and revenue efficiency increased. The average cost efficiency was higher for the Baltic countries and the Czech Republic. Lower values were observed for Romania and Hungary. 340 Repková (2013) estimated the cost and profit efficiency of the Czech commercial banks in the period 2001-2010 using SFA. The average cost efficiency ranged the value 78-91% and the average profit efficiency ranged 64-99%. The highest average cost efficiency achieved the group of the medium-sized banks following by the group of small banks and the highest average profit efficiency achieved the group of small banks. The largest banks were the lowest efficient in the case of the cost and profit efficiency. The reason for the inefficiency of the Czech banks was mainly an excess of client deposits in the balance sheet of banks and improperly chosen size (range of operation) of individual banks (especially the largest one). Bod'a and Zimková (2015) used three approaches: the services-oriented approach, intermediation approach and the profit oriented approach to investigated efficiency of the Slovak banking industry over the years 2000-2011.They used DEA models to measure technical efficiency of eleven commercial banks in three sub-periods: 2000-2003, 2004-2008 and 2009-2011. In each of these periods, banks were pooled together in one data frame. 2 Methodology and data In this paper, we discuss some extensions of basic DEA models. If price data are available then it is possible to measure allocative, technical efficiency as well as overall cost, revenue, or profit efficiency. To calculate these main types of efficiency, a set of linear programs should be solved. The output-oriented DEA model under the assumption of variable return to scale can be used for calculation of output-oriented technical efficiency and revenue efficiency. Output-oriented model under the assumption of variable return to scale can be written in the following form (Coelli et al., 2005): Max 0 j = \,2,...,n Where 4>q is output-oriented technical efficiency (TEq) of the Decision Making Unit (DMUq) in the output-oriented DEA model, yrq is produced amounts of Ith output (r = l,2,...,s) for DMUq, xiq is consumed amounts of itn input (/' = l,2,...,m) for DMUq, is produced amounts of r*h output (r = l,2,...,s) for DMUj (J = 1, 2,...,n), xy is consumed amounts of itn input (/' = l,2,...,m) for DMUj (j = 1, 2,...,n), Kj is weight assigned to the DMUj (j = l,2,...,n). To calculate revenue efficiency the following revenue maximisation DEA problem is necessary to solve (Coelli etal., 2005): s Max prqyrq (2) s.t. Yux^i-X* i = l,2,...,m 341 n r = 1,2,...,s n Xs > 0 j = 1,2,..., n Where prq is a vector of output prices of DMUq and y rq is the revenue maximising vector of output quantities for DMUq, given the output prices prq and the input levels xiq. The overall revenue efficiency (REq) is defined as the ratio of observed revenue to maximum revenue for the DMUq (Coelli etal., 2005): The overall revenue efficiency can be expressed as a product of technical and allocative efficiency measures. Therefore, the allocative efficiency of the DMUq can be calculated as the ratio of revenue efficiency (REq) to output-oriented technical efficiency (TEq) of the DMUq. These three measures (technical, allocative and overall revenue efficiency) can take values ranging from zero to one, where a value of one in case of TE, AE and RE indicates full efficiency. If the production unit is fully technically efficient (TEq=l) and displays allocative efficiency (AEq=l); it is also overall revenue efficient (REq=l). This production unit achieve the maximum possible outputs at given inputs, while the proportion of outputs will guarantee the maximum possible revenues. If the production unit is technically efficient (TEq=l) but doesn't demonstrate allocative efficiency (AEq