Masarykova Universita Brno – Ekonomicko-správní fakulta
Habilitační práce
Ing. Irena Jindřichovská, CSc.
Transformace českého finančního a
kapitálového trhu na přelomu tisíciletí
Studie finančního prostředí
České republiky
Brno březen 2012
Obsah
Část 1: Shrnutí publikovaných prací a jejich nejdůležitější vědecké poznatky
1.1 Komentář k předkládaným pracím 5
1.2 Nejdůležitější vědecké poznatky článku „The Relationship between Accounting
Numbers and Returns: Some Empirical Evidence from the Emerging Market
of the Czech Republic” 8
1.3 Nejdůležitější vědecké poznatky článku „Accounting for Good News and Accounting
for Bad News: Some Empirical Evidence from the Czech Republic” 10
1.4 Nejdůležitější vědecké poznatky článku „Implementing IFRS: A Case Study
of the Czech Republic” 12
1.5 Nejdůležitější vědecké poznatky článku „Response of Regulatory Bodies to Financial
Crises: Role of Auditors and International Comparison” 14
1.6 Abstrakt 15
Část 2: „The Relationship between Accounting Numbers and
Returns: Some Empirical Evidence from the Emerging Market of the
Czech Republic” 16
1 Introduction 18
2 Background 19
3 Review of the Literature on Emerging and Developed Markets 21
4 Methodology and Model Specification 25
5 Description of the Data 28
6 Regression Results and Implications 31
7 Conclusions and Implication for Future Research 35
8 Notes 37
9 References 39
10 Appendix 42
11 Tables 1–7 44
Část 3: „Accounting for Good News and Accounting for Bad News:
Some Empirical Evidence from the Czech Republic” 51
1 Introduction and Motivation 53
2 Czech Accounting and its Capital Market Context 55
3 Literature Review 57
3.1 The Influence of Losses on Earnings Response Coefficients 57
3.2 Timeliness and Conservatism 58
4 Methodology 60
5 Sample Data 62
6 Empirical Findings 64
6.1 The Impact of Losses 64
3
6.2 Asymmetric Timeliness 65
7 Conclusion 67
8 Notes 69
9 References 72
10 Figures 78
11 Tables 1–5 80
Část 4: „Implementing IFRS: A Case Study of the Czech Republic” 86
1 Introduction 88
2 Literature Review 88
3 Approach to Analysis 93
4 The Stock Exchange and Overall Enterprise Funding 95
5 Implementation of IFRS into Local Legislation 96
6 Findings 98
7 Conclusions 113
8 Appendix 116
9 References 117
10 Notes 122
Část 5: „Response of Regulatory Bodies to Financial Crises:
Role of Auditors and International Comparison” 124
1 Introduction 125
2 Investicni a Postovni Banka (IPB) 125
2.1 Circular Ownership 126
3 Reaction of Regulatory Bodies (CNB and the Chamber of Auditors) 127
4 The Enron Scandal 128
5 The US Regulatory Response 129
6 Monitoring and Sanctions in the Czech Republic 129
7 Conclusion and Directions for Further Research 130
8 References 131
9 Shrnutí a závěr 133
4
Část 1: Komentář k publikovaným pracím
1.1. Shrnutí a ucelený komentář k předkládaným pracem
Soubor níže uvedených prací tvoří celkem čtyři články: tři články byly publikovány v recenzovaných
zahraničních časopisech a jeden článek byl vydán v recenzovaném zahraničním časopise a následně
v monografii zahraničního nakladatelství Routledge. Poslední recenzovaný příspěvek vyšel
v monografii, jež byla jedním z výstupů výzkumného záměru Fakulty sociálních věd Univerzity
Karlovy v roce 2004.
1 Jindrichovska, I. 2001. The Relationship between Accounting Numbers and Returns: Some
Empirical Evidence from the Emerging Market of the Czech Republic. European Accounting
Review 10(1), pp. 107–131 [Impakt faktor 1,47].
2 Jindrichovska, I., McLeay, S. 2005.Accounting for Good News and Accounting for Bad
News: Some Empirical Evidence from the Czech Republic. European Accounting Review
14(2), pp. 1–21 [Impakt faktor 1,471] (podíl autora 90 %).
3 Sucher, P., Jindrichovska I. 2004. Implementing IFRS: A Case Study of the Czech Republic.
Accounting in Europe. 1, pp. 109–141. (Podíl autora 66 %)
4 Jindrichovska, I. 2004. Response of Regulatory Bodies to Financial Crises: Role of Auditors
and International Comparison. In Kotábová, V., Prážová, I., Schneider, O. (eds.) Rozvoj české
společnosti v Evropské unii. 2 díl. Prague: MatfyzPress, s. 193–198.
V devadesátých letech 20. století docházelo k postupné transformaci české ekonomiky z plánované
ekonomiky na ekonomiku tržní. Finanční sektor hrál v této transformaci významnou roli. Finanční
sektor je úzce navázán na kapitálový trh, který následně reflektuje a ovlivňuje na finanční situaci
firem. Předkládaná habilitační práce se v prvních dvou článcích zabývá vazbou českého kapitálového
trhu na trh společností. Následující dvě práce se věnují institucionálním zázemím trhů a jejich regulaci
a analyzuje, jak vnímají české firmy změny účetních standardů po jejich harmonizaci v roce 2005.
Studie pojednávají o institucionálních úpravách účetnictví a o měnících se pravomocech finančních
autorit v ČR na přelomu tisíciletí.
První dvě studie obsažené v habilitační práci se zabývají vztahem účetních výsledků a akciových
výnosů na českém kapitálovém trhu. Zkoumají dvě podoby tohoto vztahu: způsob, jakým jsou účetní
data reflektována kapitálovým trhem, resp. analýzou toho, jak trh předvídá účetní výsledky firem, a
dále dopadem „dobrých zpráv“ – tržních výnosů a „špatných zpráv“ – tržních ztrát na účetnictví
firem. Tyto články se zabývají těsností vztahu mezi tržními a účetními výnosy a jeho bližšími
charakteristikami – „včasností“ a konzervatizmem v podmínkách České republiky na přelomu
5
tisíciletí. O tom, že toto téma je stále velice aktuální svědčí mimo jiné množství ohlasů na tyto práce i
v nedávné době. K výzkumu byl v obou případech použit rozsáhlý vzorek českých společností, které
byly na přelomu tisíciletí obchodovány na českém kapitálovém trhu – Pražské burze cenných papírů.
Práce statisticky testuje tzv. „earnings response coefficient“ – koeficient reakce ceny akcie na změnu
účetních výnosů. Na českém trhu byl identifikován statistický významný vztah, potvrzující souvislost
tržních výnosů a účetních výsledků firem. Práce byla zajímavým doplňkem studií v daném oboru,
protože používala stejnou metodiku jako ve vyspělých kapitálových trzích na datech z čerstvě se
transformující ekonomiky. Výsledky byly publikovány v časopise European Accounting Review
v roce 2001.
V následné studii z roku 2005 se testy soustředily na těsnost tohoto vztahu a na to, zda je možné
pozorovat odlišnou reakci kapitálového trhu v případě účetní ztráty než v případě účetního zisku.
Smyslem testu bylo zjistit, zda můžeme pozorovat odlišnou, resp. odlišně silnou reakci ceny na
oznámení účetní ztráty než v případě očekávaného zisku. Studie doplňovala široké portfolio prací
napsaných na podobné téma v jiných evropských zemích a rovněž v zemích s transformující se
ekonomikou a byla rovněž publikována v časopise European Accounting Review v roce 2005.
Spoluautorem práce byl prof. S. McLeay z University of Bangor, UK.
Obě výše uvedené práce jsou rovněž citovány ve Web of Science, stejně jako přípravné studie a
working papery, které jim předcházely.
Zájem o vztah mezi účetními výnosy a výsledky na kapitálových trzích vedl k podrobnějšímu studiu
účetních standardů a jejich změn a k tomu, jak jsou tyto změny vnímány vykazujícími společnostmi a
samotnými kapitálovými trhy. Na počátku nového tisíciletí docházelo na celosvětové úrovni ke
sjednocování účetních standardů. Hlavní motivací tohoto postupu byl výklad účetních dat
kapitálovými trhy. Na světových burzách bylo obchodováno stále více společností ze zahraničních
trhů a bylo třeba zpřístupnit informace o investičních příležitostech zahraničním investorům. Vedla se
rovněž diskuse, jestli se firmy mají přiklonit spíše k evropským standardům nebo ke standardům
americkým. Navazující dva články promítají obraz této problematiky do České republiky a
pojednávají o institucionálních úpravách účetnictví, které byly zaváděny do roku 2005, a
o pravomocech finančních autorit na konci minulého a na počátku nového tisíciletí.
V květnu roku 2004 přistoupilo deset nových zemí k Evropské unii (jednalo se o Českou republiku,
Estonsko, Kypr, Lotyšsko, Litvu, Maďarsko, Maltu, Polsko, Slovinsko a Slovensko). V okamžiku
přistoupení se přistoupilo také na mnoho nových evropských regulací, které vyvolaly dramatické
změny v oblasti práva, místních pravidel a způsobů prosazování těchto pravidel. Jednou
z významných změn je změna finančního výkaznictví u společností, které jsou simultánně
obchodované na zahraničních burzách. Tyto společnosti jsou od roku 2005 povinny připravovat
6
finanční výkazy podle mezinárodních pravidel finančního výkaznictví (International Financial
Reporting Standards – IFRS).
V tomto období totiž probíhala harmonizace účetních standardů v celé Evropě, a proto se jako vhodné
téma k zamyšlení nabízela studie o zavádění účetních mezinárodních standardů v ČR.
Třetí článek zkoumá úpravy účetnictví v ČR po vstupu do Evropské unie v roce 2005 a zabývá se
požadavky vymahatelnosti a adekvátností úprav finančních a účetních standardů v kontextu právě
probíhajících účetních a regulačních změn v České republice. Tento výzkum, materializovaný
v publikaci se spoluautorkou P. Sucher z Royal Holloway s názvem „Implementing IFRS: A Case
Study of the Czech Republic“ byl v roce 2004 publikován v mezinárodním recenzovaném časopise
s názvem Accounting in Europe. Tento výzkum měl kvalitativní charakter. Použitá metoda se opírá o
semi-strukturované interview, pomocí kterého jsou zkoumány odlišnosti a charakteristické rysy
zavádění účetních standardů na vzorku českých firem. Výzkum byl doplněn o širší pohled: (1) jak
vnímají zavádění mezinárodních standardů průmyslové firmy, (2) jak jej vnímají reprezentanti
finančního trhu, tj. banky, (3) jak jej vnímá český kapitálový trh reprezentovaný brokery a (4) jaký má
tato změna dopad na účetní profesi, tj. profesionální účetní firmy. Výsledkem této průřezové studie
byla charakteristika zavádění jednotných účetních standardů v České republice.
Počátek nového tisíciletí byl rovněž obdobím velkých finančních a účetních skandálů po celém světě.
Finanční skandály prorůstaly z průmyslových firem do firem účetních a finančních Poslední studie se
zabývá porovnáním reakcí regulačních institucí v ČR a USA na velké finanční skandály. Článek
porovnává reakce institucionálního zázemí v obou zemích na velké skandály a formy regulace, jež
byly realizovány. Srovnání probíhá pomocí porovnání dvou případových studií podniků Enron v
USA a české Investiční a Poštovní Banky.
7
1.2. Nejdůležitější vědecké poznatky článku „The Relationship between Accounting Numbers and
Returns: Some Empirical Evidence from the Emerging Market of the Czech Republic“ [Vztah mezi
účetními výsledky a tržními výnosy: empirické důkazy z transformujícího se trhu v České republice]
Irena Jindřichovská
Generally, stock prices reflect future expectations of earnings, whereas accounting data reflect past
performance. This paper attempts to discover the relationship between accounting data and market
price returns of the companies listed on the Prague Stock Exchange (PSE). The Prague Stock
Exchange was established in 1993 and provides an opportunity to make a comparison between a
newly established market and the findings of studies of established markets.
There has been a wealth of publications and accounting research studies on developed markets.
Generally, accounting attributes are thought to be relevant because they tend to be contemporaneously
statistically associated with stock prices. Some studies have suggested, and empirically tested, that
stock prices lead earnings (e.g. Collins, Kothari, Rayburn, 1987; Kothari, 1992, Kothar, Sloan, 1992;
Kothari, Zimmerman, 1995). This study tests the existence of such a relationship in the Czech capital
market, relying partially on the methodology proposed by Kothari and Sloan (1992) and Kothari
(1992).
This paper investigates, whether there is a statistically significant permanent relationship between
returns and accounting data on the Czech market. The study was conducted using accounting earnings
and stock prices during the period of 1993-1998. The empirical evidence here suggests that a similar
relation exists on the emerging Czech market. The relation is statistically significant for measurement
windows of one year and longer. The increase in the mean response coefficient, reported later in this
study, suggests that one-leading-year returns are as important as contemporaneous returns in terms of
their sensitivity to annual earnings changes. However, one cannot infer with a degree of confidence
that the Czech capital market views earnings changes to be largely permanent, which would be
consistent with the time series properties of annual earnings.
Část 1.2 Shrnutí v češtině
V obecné rovině odrážejí ceny akcií očekávání budoucích podnikových výnosů. Účetní data na
druhou stranu odrážejí minulé výsledky. Tento článek se zabývá vztahem mezi účetními výsledky a
výnosem cen akcií českých společností obchodovaných na Burze cenných papírů v Praze (PSE).
Pražská burza byla znovu otevřena v roce 1993 a nabízí se tedy příležitost srovnat výsledky mladých
trhů a trhů již rozvinutých.
O rozvinutých trzích byla již na toto téma napsána celá řada výzkumných studií. Účetní
charakteristiky jsou z hlediska časového významně statisticky spjaty s cenami akcií, a proto jsou pro
tvorbu ceny relevantní. V předchozích studiích především na americkém a britském trhu byla
8
empiricky testována hypotéza, že ceny akcií předbíhají účetní výsledky (např. Collins, Kothari and
Rayburn, 1987; Kothari 1992, Kothari and Sloan, 1992; Kothari and Zimmerman, 1995). Studie
testuje existenci podobného vztahu na českém kapitálovém trhu a částečně se opírá o metodologii
autorů Kothariho a Sloana (1992) a Kothariho (1992).
Tento článek zkoumá, jestli existuje v českých podmínkách statisticky významný a trvalý vztah mezi
výnosy kapitálového trhu a účetními daty. Ve studii byl použit vzorek účetních výnosů a cen akcií
z období let 1993 až 1998. Empirické výsledky ukazují, že na transformujícím se trhu existují
podobné vztahy jako na trzích vyspělých. Vztah je statisticky významný při intervalu měření jednoho
roku a delším. Nárůst střední hodnoty koeficientu reakce výnosů (mean response coefficient), který
popisujeme v této studii, dokresluje, že jeden rok staré (loňské) výnosy jsou co do citlivosti změn
ročních účetních výnosů stejně významné jako současné výnosy. Nelze však s velkým stupněm
spolehlivosti říci, že český kapitálový trh považuje změny účetních výnosů za trvalé, což by bylo
v souladu s vlastnostmi časových řad ročních výnosů.
Na efektivních a vyspělých kapitálových trzích odrážejí ceny akcií změny očekávání budoucích
hotovostních toků. Ve srovnání s tím mají účetní data jen omezenou schopnost tato očekávání
zachytit. Autoři Kothari a Sloan (1992) uvádějí, že podstatou tohoto problému je objektivita,
verifikovatelnost a další účetní konvence, na kterých jsou založené obecně přijaté účetní principy
(GAAP). Tyto principy vlastně omezují schopnost účetních výnosů odrážet tržní revize změny
očekávání budoucích cash flow. Změny cen akcií odrážejí mnohem širší informační množinu, tzn., že
koeficienty reakce ceny na účetní výnosy budou podhodnocené.
V českých podmínkách byl tento vztah rovněž testován a výzkum potvrdil existenci statisticky
významného vztahu mezi cenami účetních výnosů za pomoci několika modelových specifikací.
Statistická významnost vztahu není příliš silná, přestože je porovnatelná s výsledky v jiných zemích.
Důvodem je nejspíše nízká likvidita na českém kapitálovém trhu a jeho nevyspělost.
9
1.3. Nejdůležitější vědecké poznatky článku „Accounting for Good News and Accounting for Bad
News: Some Empirical Evidence from the Czech Republic“ [Účetnictví Dobrých zpráv a účetnictví
Špatných zpráv: empirické důkazy z České republiky]
Irena Jindřichovská (90 %), Stuart McLeay (10 %)
This paper is motivated by the links that continue to be forged between security pricing and
accounting, building on recent findings that firms tend to be asymmetrically conservative in the
timeliness of earnings recognition. The evidence is that firms in the European Union tend to recognise
unrealised losses more quickly in their earnings than unrealised gains (Giner and Rees, 2001; Raonic,
McLeay and Asimakopoulos, 2004), and there is evidence of even greater accounting conservatism in
the US (Basu, 1997; Ball, Kothari and Robin, 2000; Givoly and Hayn, 2001). This article investigates
whether the Czech market exhibits conformity with the behaviour that has been documented
elsewhere by examining the earnings/returns relationship, focusing to begin with on the impact of
losses on earnings response coefficients and then considering the asymmetric timeliness of income
recognition in the Czech market.
The findings indicate that the Czech market is similar to more developed markets, at least in one
respect: There is statistically significant evidence of different market effects of profits and losses, in
that profits are more persistent than losses. However, contrary to the findings in more developed
markets, there is no statistically significant evidence of earnings conservatism in the Czech market.
These results are most probably due to the continuing influence of restrictive tax regulations that
mitigate any tendency towards conservatism, as well as the transitional nature of the economy. A
further reason is likely to be that the regulatory environment in the Czech Republic is close to the kind
of stakeholder corporatism that is described by Ball et al. (2000), who show that conservatism tends to
be less pronounced in such regimes where there are fewer managerial incentives to bias current
earnings. In conclusion, if changes in market prices signal good news and bad news about future risky
outcomes, there is no evidence of asymmetry in the Czech market in accounting for such risks.
Část 1.3 Shrnutí v češtině
Intenzivní snaha poslední doby o vytvoření standardních kapitálových trhů a zavedení správného
finančního výkaznictví v tranzitivních ekonomikách by měla vést k vytvoření lepšího informačního
prostředí pro investory a umožnit jim získat lepší přehled o podnikovém hospodaření, řízení a
výsledcích. Zodpovědnost podnikového managementu vůči investorům je tak posilována (Sucher,
Jindrichovska, 2004).
Vztah mezi podnikovými výsledky a výnosy kapitálového trhu byl zkoumán na mnoha tranzitivních
trzích např. v Polsku (Jermakowicz, Gornik –Tomaszewski, 1998) a v České republice (Jindrichovska,
10
2001), a do jisté míry také v Maďarsku (Chun, 2000). Tento výzkum se soustředil na
konzervativismus s použitím různých specifikací modelu testujícího vztah mezi účetními a tržními
výnosy, narozdíl od konzervatismu v účetnictví a různé dopady účetních ztrát a zisků na výsledky cen
akcií na kapitálovém trhu.
Tento článek se zabývá vztahem mezi vývojem cen akcií a účetními výnosy společností a opírá se o
nedávná zjištění, že firmy jsou asymetricky konzervativní co do včasnosti vykazování účetních
výnosů. Tento konzervatizmus vyplývá ze zjištění, že firmy v Evropské unii zveřejňují nerealizované
(očekávané) ztráty „rychleji“ než nerealizované (očekávané) zisky (viz např. Giner, Rees, 2001;
Raonic, McLeay, Asimakopoulos, 2004). Studie ze Spojených států poukazují na ještě větší účetní
konzervatizmus (Basu, 1997; Ball, Kothari, Robin, 2000; Givoly, Hayn, 2001). Tento článek zkoumá,
jestli český trh vykazuje stejné chování, jaké bylo dokumentováno na jiných trzích. Studie zkoumá
vztah mezi účetními zisky a cenovými výnosy. Nejprve se soustředí na dopad účetních ztrát na
koeficient reakce výnosů (earnings response coefficients) a poté se soustředí na časovou asymetrii
patrnou ve zveřejňování výnosů na českém trhu.
Zjištění ukazují, že český trh je v jedné věci podobný vyspělým kapitálovým trhům: existuje
statisticky významný důkaz rozdílného tržního dopadu u zisků a ztrát, kde zisky jsou považovány za
trvalejší než ztráty. Na rozdíl od zjištění z oblasti vyspělých kapitálových trhů není na českém trhu
statisticky významný důkaz o konzervatizmu výnosů. Tyto výsledky jsou tedy nejspíše důsledkem
pokračujícího vlivu restriktivních daňových předpisů, které snižují tendenci ke konzervatizmu, jakož i
jiných ekonomických charakteristik transformující se ekonomiky. Dalším důvodem je pravděpodobně
to, že regulační prostředí v České republice je blíže vlastnickému korporatismu zájmových skupin
přítomných v podniku než zájmům akcionářů. Tento rozdíl. jak popsal Ball et al. (2000) ukázal, že
konzervatizmus je významnější v režimech, kde je menší motivace manažerů ke zkreslování účetních
výsledků běžného období. Jestliže změna tržní ceny signalizuje „dobrou zprávu” resp. „špatnou
zprávu” o budoucích nejistých výsledcích, v České republice zatím důkaz takovéto asymetrie
neexistuje a konzervatizmus není na tomto trhu statisticky významný.
Informačně efektivní trh znamená, že všechny relevantní informace se zobrazí v cenách okamžitě, a to
včetně očekávání budoucích zisků a ztrát plynoucích z rizikového podnikání firem. Konzervativizmus
v účetnictví plyne z jeho regulace a požadavku, aby se očekávané ztráty již zahrnuly do výsledků,
zatímco očekávané zisky jsou zahrnuty teprve poté, co jsou realizovány. Podle Basu (1997) je
výsledkem to, že špatné zprávy (ztráty) se odrazí v účetnictví rychleji (výrazněji) než očekávané
účetní zisky. Nepříznivá informace se zobrazí na kapitálovém trhu úplně, zatímco příznivá informace
jen částečně. Proto bylo zjištěno, že dopad očekávaných účetních ztrát je na kapitálovém trhu
výraznější než u účetních zisků (Hayn, 1995).
11
1.4 Nejdůležitější vědecké poznatky článku „Implementing IFRS: A Case Study of the Czech
Republic“ [Zavádění Mezinárodních standardů finančního výkaznictví: Případová studie České
republiky]
Patricia Sucher (33 %), Irena Jindřichovská (66 %)
This empirical paper presents a study of the implementation process for international financial
reporting standards (IFRS) in one of the accession countries, the Czech Republic. Based upon a
review of the legislation, institutional framework and context, and drawing upon recent interviews
with Czech companies required to prepare IFRS accounts, auditors and institutional players in the
Czech Republic, the paper highlights some of the key issues that are arising with the move to the
implementation of IFRS for listed group companies and other enterprises in the Czech Republic.
The paper considers the issues that arise when implementing new accounting regulations, some of
which are not new and have been well covered in the literature, but others of which are particular to
the implementation of IFRS. The method of implementation, the scope of IFRS, particular issues with
local accounting practice and IFRS, the issue of enforcement of compliance with IFRS and its
relationship with audit, the link between IFRS and taxation and the provision of education and
training are all considered. There is also a review of the state of preparedness of local group listed
entities with respect to the implementation of IFRS.
There are many potentially rich areas for accounting research where the work could also inform about
the practice of IFRS accounting. The paper provides a contribution by highlighting how one country
has moved to implement the requirement for group listed enterprises in order to prepare IFRS
accounts and the issues that then arise for legislators, preparers and users.
Část 1.4 Shrnutí v češtině
Tato empirická studie zkoumá proces zavádění mezinárodních standardů finančního výkaznictví
(international financial reporting standards - IFRS) v České republice. Česká republika v období
tohoto výzkumu právě přistupuje k EU. Práce se opírá o studium legislativy, institucionálního rámce a
aktuálního kontextu a dále o rozhovory se zástupci českých společností, které již mají povinnost
připravovat podnikové účty podle IFRS, s auditory a se zástupci účetních institucí v České republice.
Studie vysvětluje a analyzuje klíčové problémy, které vyvstávají při přechodu na IFRS u skupiny
společností kotovaných na burze nebo na mimoburzovním trhu.
Článek rozebírá otázky vznikající při zavádění nové účetní regulace. Některým z nich se již věnovala
dřívější literatura. Zavádění mezinárodních standardů IFRS má však i svá specifika, a to metodu
zavádění, způsob a rozsah standardů IFRS, konkrétní otázky spojené s tradicí české účetní praxe,
otázky vymahatelnosti sankcí při nedodržování standardů IFRS a dále vztah k auditu, vazbu mezi
12
IFRS a daněmi a systém vzdělávání a kvalifikace profesionálních účetních. Zkoumá také stav
připravenosti českých kotovaných společností na zavádění IFRS.
V tomto výzkumu účetnictví je potenciálně mnoho oblastí, které by mohly informovat a ovlivňovat
účetní praxi ve vztahu k IFRS. Mezi přínosy článku patří, že ukazuje, jak se jedna země vypořádává
se zaváděním IFRS u skupiny kotovaných společností. Článek otevírá také další otázky v oblasti
právní, v oblasti práce regulačních institucí a konečných uživatelů těchto účetních standardů.
V květnu 2004 se k Evropské unii připojilo 10 nových států. Jednu z velkých změn, která stojí před
těmito státy, je zavedení mezinárodních standardů (IFRS) u podniků, které jsou obchodovány na
několika burzách. Během posledních 14 let tyto země zavedly do svých právních systémů mnoho
změn (Bailey, 1995; Bailey, Alexander, 2001; Schroeder, 1999; Sucher Zelenka, 1998).
Studie uvádí přehled relevantní literatury z oblasti mezinárodní harmonizace účetnictví a zkoumá,
jaký dopad může mít zavedení IFRS na v transitivní ekonomiky. V článku je podle jednotlivých
relevantních témat analyzován legislativní a institucionální přístup k zavádění IFRS v České
republice. Analýza je podpořena a rozhovory interview s regulátory, auditory a dalšími
institucionálními aktéry. Ve studii je ukázáno v jaké fázi se momentálně nachází příprava regulací
finančního výkaznictví podle mezinárodních pravidel IFRS a ukazuje stav připravenosti k zavedení
IFRS ve vybraném vzorku podniků. Ukázány jsou rovněž problémy a prospekty očekávaného
přechodu k implementaci mezinárodních standardů ve společnostech obchodovaných na zahraničních
burzách a na české burze.
13
1.5 Nejdůležitější vědecké poznatky článku „Response of Regulatory Bodies to Financial Crises:
Role of Auditors and International Comparison“ [Odpověď regulačních orgánů na finanční skandály:
role auditora a mezinárodní srovnání]
Irena Jindřichovská
In the light of recent crises capital markets are re-evaluating value relevance of financial information
for price formation on the stock market. In this paper we investigate this issue through a comparison
of recent developments in the US market as compared to the Czech market.
The paper identifies specific and significant contrasts between the US and Czech economies and
draws conclusions on regulatory structures in developing as compared to advanced economies.
In this paper we seek to utilise some of the work of Ball et al. (2003) to review the interaction between
accounting standards, preparer incentives and economic income in the Czech Republic and the USA.
In particular we focus on the regulatory and market reactions to particular financial crises: Enron in
the USA and Investicni a Postovni Banka in the Czech Republic. (Jindrichovska et al, 2004)
Část 1.5 Shrnutí v češtině
Ve světle finančních krizí poslední doby přehodnocují kapitálové trhy relevantnost finančních
informací pro tvorbu cen akcií. V tomto článku se uvedenou otázkou zabýváme prostřednictvím
porovnání nedávného vývoje ve Spojených státech a v České republice.
Článek identifikuje specifické a významné rozdíly ve způsobu regulace finančního trhu ve Spojených
státech a v České republice a porovnává regulační orgány v rozvinutých a rozvíjejících se
ekonomikách .
Studie vychází z prací Balla et al. (2003) a zabývá se interakcemi mezi účetními standardy, motivací
regulačních orgánů a ekonomickým příjmem na akcii v České republice a v USA. Zaměřujeme se
konkrétně na reakce regulačních orgánů na určité finanční krize: případ Enron ve Spojených státech a
na případ Investiční a Poštovní Banky v ČR.
Reakce regulačních orgánů a reakce trhu v České republice (krize IPB) kontrastuje ostře s reakcemi na
krizi v USA (bankrot společnosti Enron). V České republice nedošlo ke změně pravidel finančního
reportingu nebo pravidel auditu a k zostření jejich vymahatelnosti, mimo změn, které byly vyvolány
přístupem České republiky k Evropské unii. Auditorská firma v českém případě nezkolabovala a
nebyla zrušena. Naopak dále se rozrostla a absorbovala lokální pobočku firmy Artur Andersen, která
zkolabovala po kolapsu firmy Enron v USA.
Podle anglosaské tradice bychom mohli říci, že neprůhlednost je velké negativum, ale na druhou
stranu je třeba připustit, že očekávání kvalitního finančního reportingu a jeho ověření a vynucení
14
auditem nebylo v České Republice nikdy tak velké, jako je tomu v USA. Investoři měli k dispozici
jiný způsob jak se dostat k informacím, které pro své investiční rozhodování potřebovali. Proto
můžeme také říci, že pohyby cen na finančním trhu nebyly nezbytně motivovány změnami
v ekonomických výnosech firem. To následně implikuje, že český finanční trh není efektivní podle
hypotézy polosilné efektivnosti kapitálového trhu.
To může být důsledkem konkrétního institucionálního zázemí v České republice, které odráží určité
spletité cirkulární vztahy viz Ball et al. (2003). Jestliže jsou v ekonomice propletené ekonomické
a vlastnické vztahy mezi společnostmi není zpravidla možno standardně vynutit dodržování standardů
finančního výkaznictví (Sucher, Kosmala-MacKullich, 2004) potom musí investoři použít jiné
způsoby, jak se k potřebným informacím v jednotlivých podnicích dostat. Odpověď na finanční krizi
kolem IPB je toho dobrým příkladem. V tisku se objevilo mnoho článků, ale na rozdíl od případu
Enron nenastaly žádné okamžité regulatorní kroky – zřejmě z důvodů popsaných výše. Je
pravděpodobné, že odpověď byla spíše neformální a orientovaná na podnikové insidery.
Tento příklad ukazuje konflikt mezi tradičním, ale v České republice stale novým přístupem a AngloSaským
přístupem, který se snaží zejména zvýšit hodnotu podniku pro akcionáře a novým
kontinentálním přístupem se širším záběrem koncentrovaným na stakeholdery. Chování banky bylo v
souladu s existujícími pravidly a právními předpisy, ale bylo také v souladu s dobrou správou
společnosti?
Abstrakt:
Tato habilitační práce se zabývá charakterem finančního a kapitálového trhu v ČR v období
transformace na přelomu tisíciletí. Zabývá se vztahem trhu k účetnictví, současným stupněm
harmonizace účetnictví a vymahatelností sankcí v porovnání s vyspělým kapitálovým trhem v USA.
V prvních dvou článcích je použita metoda kvantitativní analýzy dat z českého kapitálového trhu
a účetních dat. V dalších dvou článcích se metodologie opírá o semi-strukturované interview
a o průzkum norem a regulací trhu dokreslený případovou studií.
Podíl autorů je u prvního a čtvrtého článku IJ = 100 %, u druhého článku IJ = 90 % a SML = 10 %
a u třetího článku PS = 33 % a IJ = 66 %.
Habilitační práce reflektuje vývoj vědeckého poznání a přispívá k poznání finančního prostředí
v České republice v období transformace na přelomu tisíciletí. V první části je největší důraz kladen
na charakteristiku českého finančního odvětví, zejména studium vztahu kapitálových zisků a účetních
výnosů podniků. Další část se zabývá připraveností českých podniků k implementaci mezinárodních
standardů IFRS a vymahatelností sankcí v souvislosti s finančními skandály na českém finančním
trhu.
15
Část 2
The Relationship between Accounting Numbers and Returns: Some Empirical
Evidence from the Emerging Market of the Czech Republic
[Vztah mezi účetními výsledky a tržními výnosy: empirické důkazy z
transformujícího se trhu v České republice]
Irena Jindrichovska
European Accounting Review 10(1), 2001, 107-131.
16
The Relationship between Accounting Numbers and Returns: Some Empirical
Evidence from the Emerging Market of the Czech Republic
Irena Jindrichovska
University of Wales, Aberystwyth
Generally, stock prices reflect future expectations of earnings, whereas accounting data
reflect past performance. This paper attempts to discover the relationship between accounting data
and market price returns of the companies listed on the Prague Stock Exchange (PSE). The Prague
Stock Exchange was established in 1993 and provides an opportunity to make a comparison
between a newly established market and the findings of studies of established markets.
There has been a wealth of publications and accounting research studies on developed
markets. Generally, accounting attributes are thought to be relevant because they tend to be
contemporaneously statistically associated with stock prices. Some studies have suggested, and
empirically tested, that stock prices lead earnings (eg Collins, Kothari and Rayburn, 1987; Kothari
1992, Kothari and Sloan, 1992; Kothari and Zimmerman, 1995). This study tests the existence of
such a relationship in the Czech capital market, relying partially on the methodology proposed by
Kothari and Sloan (1992) and Kothari (1992).
This paper investigates, whether there is a statistically significant permanent relationship
between returns and accounting data on the Czech market. The study was conducted using
accounting earnings and stock prices during the period of 1993-1998. The empirical evidence here
suggests that a similar relation exists on the emerging Czech market. The relation is statistically
significant for measurement windows of one year and longer. The increase in the mean response
coefficient, reported later in this study, suggests that one-leading-year returns are as important as
contemporaneous returns in terms of their sensitivity to annual earnings changes. However, one
cannot infer with a degree of confidence that the Czech capital market views earnings changes to
be largely permanent, which would be consistent with the time series properties of annual earnings.
Key words: Earnings response coefficients, emerging markets, Czech capital market, pooled
regression
17
1. Introduction
There exist a number of studies on privatisation and corporate governance in the
1990s in the Czech Republic. However, there has been little previous research, which
describes and analyses the relation between accounting data and market prices on the
Central European emerging markets in general. This paper attempts to fill this gap by
looking at the association, both contemporaneous and lagged, between accounting data
and security returns. It is the first study on the Czech capital market describing the
relationship between earnings and market returns.
The results of this paper tend to corroborate existing empirical evidence, albeit not
strongly, that the bias in earnings response coefficient estimates is reduced as the
earnings and return measurement window is increased. The relation was tested using
models relating levels of earnings and prices deflated by beginning prices. This
specification was tested also in its cumulative form – summing earnings across the whole
estimation period. Furthermore, two alternative model specifications were tested: a model
exploring earnings changes and a model using earnings as a deflator. To allow for the
possible influence of firm specific characteristics, estimation of pooled data was performed
throughout the analysis with (1) common intercepts, (2) fixed effects, and (3) random
effects. Out of the two alternative model specifications, the second model (using earnings
as a deflator) was not statistically significant.
In an efficient and mature market, price changes reflect revision of the capital
market‟s expectation of future cash flows. In comparison, accounting earnings have only a
limited ability in this respect. Kothari and Sloan (1992) state that the primary reason for
this is that objectivity, verifiability and other conventions that underlie GAAP limit the ability
of accounting earnings to contemporaneously reflect the market‟s revision of future net
cash flow expectations. The change in prices is a response to a much wider set of
information, so that the response coefficient may be downward biased.
18
The rest of the paper is organised as follows: section 1 is the introduction, section
2 provides background to conditions on the Czech capital market, and section 3 reviews
the relevant literature on both emerging and developed capital markets. The methodology
and model specifications are discussed in section 4, and section 5 provides a description of
the data. Results and implications are discussed in section 6. The paper‟s findings are
summarised in section 7.
2. Background
Characteristics of the Czech capital market
The Czech Voucher Privatisation started in 1992 when privatisation „vouchers‟ were
distributed to the public.1
At the same time, company shares were administratively created,
based on the book information of privatised enterprises. The distribution algorithm
assigning shares to privatisation vouchers was executed by the so called „RM-System‟,
which is a computer system matching supply and demand of these administratively-created
shares.2
After distribution the shares began to be traded on the RM-System, which now
resembles an over-the-counter market. Early in 1993, approximately six months after the
establishment of the RM-System, the Prague Stock Exchange (PSE) began trading again
after a break spanning over half a century. Its major purpose at that time was to provide a
market for shares which had been distributed to the public through the voucher scheme.
Most shares were, and still are, traded on both the RM-System and the Prague Stock
Exchange. Even though the matching systems are a little different in each case, they
provide basically the same services and compete on the same market. The major
difference is that shares that are listed on the main market of the Prague Stock Exchange
have to release accounting information in accordance with PSE rules and conditions
(Patton and Zelenka, 1997).
Initially, the capital market and related infrastructure were not relied upon to provide
19
reliable market values, and even less were they considered a source of funds for industry.
Generally, the market was supposed to be a matching system with no human contribution
crossing the bids and offers and fulfilling no other functions. Unfortunately, mistakes and
omissions occurred, including the lack of strict and enforceable rules and regulations, and
this has complicated the situation. Additionally, a lack of transparency has caused the exit
of many foreign investors in the last two or three years.
The Prague Stock Exchange is a relatively illiquid marketplace compared to
markets in the U.S. and in Western Europe. One reason for low liquidity is that there are
too many shares in very small companies which would never have been listed under
normal circumstances, ie if no voucher privatisation had taken place (Filacek et al., 1998;
Sommer, 1996, 1997a and 1997b). During the privatisation process, 1670 companies were
listed on the stock exchange. However, since late 1996, when a substantial administrative
reform of the capital market was initiated, about 80% of these companies have been delisted,
due to poor liquidity and poor reporting standards.
Recently, the need for a strong regulatory and supervisory environment was
recognised, and securities regulation has been strengthened, including the introduction of
more effective laws and regulations for investment funds and trusts. A Securities
Commission was also established in early 1998. However, according to the IMF (1999),
market oversight needs to be tightened further, and the Securities Commission should be
de-linked from the Ministry of Finance and given powers to promulgate regulations of its
own. In addition, these efforts need to be supplemented by the upgrading of accounting
and auditing standards to international norms, including the requirement of mark-to-market
accounting, so that accounting methodologies do not provide distorted incentives, allow
circumvention of regulations or help avoid full disclosure.3
Recent studies of the Czech
capital market have focused on both market efficiency (eg Filacek et al., 1998) and the
institutional environment (eg Svejnar, 1995; and Mejstrik, 1997).
20
3. Review of the literature on emerging and developed markets
The literature provides several explanations of how capital markets view earnings
changes and magnitudes of the estimated earnings response coefficients. This short
survey of the relevant literature starts with several recent articles dealing with the emerging
market of the Czech Republic.
The association between accounting data and market prices in the Czech setting
has been analysed by Sommer (1996), who tested for an association between the stock‟s
riskiness as represented by beta and various financial ratios. He found that the relationship
between the earnings-to-price ratio and the beta coefficient was significant at the one per
cent level. His sample consisted of twenty companies with the largest capitalisation and
covered the period September 1994 to December 1995. Beta coefficients were estimated
as the slope of the characteristic line between each share and the representative PX-50
market index.
The weak-form efficiency of the Czech capital market for the period 1995-97 was
tested by Filacek et al. (1998) using the PX-50 market index. They came to the conclusion
that in this period the Czech market did not satisfy the weak form of the efficiency
hypothesis.
Sommer (1997a, 1997b) has presented the results of an empirical study measuring
risk on the Czech capital market from September 1994 to the end of 1995 by standard
econometric methods. His results show that there is quite a strong relationship between
risk and liquidity of the stock of a company taken in isolation. He also shows a weak
relationship between company beta and industry beta. Interestingly, there seems to be no
relationship between a company‟s fundamentals and its beta coefficient. The study‟s
findings lead to some interesting conclusions about the degree of efficiency of the Czech
emerging capital market and its possible future development.
A number of important papers analyse the association between market returns and
21
company earnings in developed markets. The studies, which are discussed below, focus
on several different aspects of the nature of the relationship between returns (prices) and
earnings.
a) Length of the response window
One approach in assessing the relation between accounting income and stock
return is to regress a measure of return on accounting income. A key decision concerns the
period of returns and income which is to be examined. The commonly estimated univariate
returns / earnings regression models generally have the form
UR a bUX eit it it   ,
where URit is unexpected return, UXit represents unexpected earnings, and eit is a random
disturbance term assumed to be normally distributed with a zero mean and variance 2
.
Research on the contemporaneous association of returns and earnings (earnings
response coefficients, ERC) was initiated by Ball and Brown (1968), who examined the
relation between accounting income and price data. They concluded that accounting
income is a major source of price relative information, but that the market anticipates it
mainly in the period prior to its release. ‘The value of information conveyed by the income
number at the time of its release constitutes on average only 20 per cent of the value of all
information coming to the market during that month.‟ (Ball and Brown, 1968:176)
It has been noted by Easton et al. (1992) that the explanatory power of models
employing a short interval (short measurement window) is rather low as measured by R2
.
In their study they extended the interval over which earnings are determined. They
hypothesise that measurement errors in (aggregated) earnings are thus reduced. The
empirical findings support this hypothesis. Similar results showing that the association
between earnings and returns is stronger as the aggregation interval is lengthened, were
presented by Kothari and Sloan (1992) and Dechow (1994). Studies by Collins, Kothari and
Rayburn (1987) and Kothari and Zimmerman (1995) suggest that earnings lag returns in up
22
to three years.
Donnelly and Walker (1995) investigate the extent to which share prices anticipate
future earnings changes by estimating earnings response coefficients (ERCs) on a sample
of 179 UK companies during the period 1972-1990. The authors found that the extent to
which prices anticipate earnings in the UK was less than that reported by Kothari and
Sloan for US companies. The reason may be differences in the informational environment
or differences between UK and US GAAP.
b) Intertemporal and cross-sectional variations of ERCs and different model
specifications.
A number of studies analyse the specification of regression models of earnings
response coefficients. Kothari and Zimmerman (1995) compare the price, differenced-price
and return models. Based on (1) the extent to which the estimated slopes and intercepts
approximate their predicted values and (2) the mis-specification (heteroscedasticity)
indicated by White‟s (1980) statistics, they suggest that price models are less biased;
return models, however, have fewer econometric problems. Similarly, Christie (1987)
concludes that, while return and price models are economically equivalent, return models
are econometrically less problematic. Estimates of earnings response coefficients are more
biased when the return model is used as against the price and differenced-price models
(eg Kothari and Zimmerman, 1995; Ohlson, 1991; Easton and Harris, 1991).
A number of studies concentrate on cross-sectional and intertemporal variations of
earnings response coefficients: Kormendi and Lipe (1987); Collins and Kothari (1989);
Easton and Zmijewski (1989). In their study, Collins and Kothari (1989) relax the
assumption of their earnings response coefficient should be treated as cross-sectional and
a temporal constant, and study the causes of variation. The cross-sectional variation is
related to earnings persistence and the firm‟s systematic risk and size – proxies for a broad
information environment. The sensitivity of the earnings response coefficients was also
23
tested by reversed regression models, estimating the returns response coefficients
(RRCs). The authors, however, advise a cautious approach to inferring the earnings
process from the inverse of the estimated RRC. The aspect of earnings persistence was
recognised and addressed in several other studies, eg Basu (1997), Donnelly and Walker
(1995), Kormendi and Lipe (1987), Kothari and Sloan (1992).
In his paper Kothari (1992) focused on three ERC models 1) a model specified in
levels, 2) a model specified in changes and 3) a model using the previous year's earnings
as the deflator. The main theoretical conclusions from the point of view of their explanatory
power are that, because prices reflect information about future earnings:
1) “compared with the change specification, the levels specification yields higher
explanatory power and generates a less biased earnings response coefficient
estimate.
2) the levels specification yields a biased earnings response coefficient when prices
contain information about more than one period ahead earnings changes.
3) if an accurate proxy for the market's unexpected earnings is used, the earnings
response coefficient estimate is unbiased and the explanatory power is greater than
when using the levels and change specifications.
4) using beginning-of-the-year price as the deflator compared to the previous year's
earnings yields a less biased coefficient estimate.
5) the explanatory power of a typically estimated price-earnings regression is
expected to be low, perhaps only about 15-20 per cent.” (Kothari, 1992:175)
The literature on earnings response coefficients and the informational content of
stock prices and returns has undergone major development in the last thirty years. Recent
studies are frequently concerned with information asymmetry, timeliness and conservatism.
Although most of the studies were performed on sets of U.S. and U.K. data, similar studies
of other markets exist in which the results of papers written on U.S. or U.K. counterparts
are compared (eg Booth et al., 1997, and Martikainen et al., 1997).
24
4. Methodology and model specifications
a) Length of the response window
This paper investigates whether returns measured over one to four leading periods
contain information about a change in annual earnings. Concentrating on the length of
response window, it should be determined whether enlarging the response window
increases the coefficient while decreasing its variability. This should be especially true in
the case of cumulative estimates, where earnings are summed over the period. The
following part of this paper is concerned with different model specifications of the earningsprice
relation in order to test the strengths of this relation. Three different models are tested
and the results are compared with model 1.
The classical assumption is that a stock's value depends on the stream of cash flow
the company is expected to generate on the investor’s behalf. In the spirit of this analysis,
total stock return (dividend and capital gain) over the period reflects the market’s revision of
its expectations about future earnings. In an efficient market, prices will align themselves
with developments as they take place, which means to say that they move in concert with
the current market expectations of future corporate earnings. It is chiefly the revision of
these expectations, which may be triggered by unanticipated elements, changes in the
underlying economy, earnings surprise or other economy-wide or firm-specific factors that
are reflected in stock returns, or in the terminology of this paper, price relatives.
Kothari and Sloan (1992) surmise that stock returns over a period reflect the
market’s revision of its expectations of future earnings. Accounting earnings over the same
period cannot reflect such revised expectations: returns therefore anticipate earnings
changes. Accordingly, Kothari and Sloan (1992) postulate, and support by empirical
evidence, that returns are expected to lead earnings changes. In other words, price
changes are expected to have a predictive power in relation to future earnings changes.
This, however, does not imply that price changes represent a perfect forecast of future
earnings changes; see Kothari and Sloan (1992:147).
25
Collins and Kothari (1989) find that unexpected earnings from the year (t+1)
correlate with returns from year t. The implication is that the market obtains alternative
information in year t, which is a substitute for some „news‟ of earnings in year (t+1). This
may be explained by the conservative nature of accounting, which recognises events but
considers them only after verification. Kothari and Sloan (1992) include leading-period
returns in their regressions and, based on their estimated earnings response coefficients,
suggest that the market, on average, views earnings changes as largely permanent.
This study examines the degree of association between price relatives (one plus
buy-and-hold return), with and without dividend inclusion, and the earnings-to-price ratio
(accounting earnings yield). Following the methodology suggested by Kothari and Sloan
(1992), the degree of association between the price relatives and earnings-to-price ratio
(earnings yield) is tested, using the panel regression method (time series – cross-section
analysis). 4
First the model using contemporaneous earnings and prices is employed:
t


 t
t
t
t
X
X
P
P
(1)
The dependent variable on the left-hand side Pt/Pt- is the price relative, or one plus
buy-and-hold return, (a) inclusive of dividends 5
and (b) exclusive of dividends over the
period from the end of t- to the end of t. 6
On the right-hand side, the fraction Xt / Pt- represents the earnings-to-price ratio
(earnings yield). The numerator consists of accounting earnings per share before
extraordinary items for firm i for the period t. Earnings are divided by the stock price at the
beginning of the period t- (which is the same as the price at the end of the previous
period). The slope coefficient  represents the earnings response coefficient.
The implications of a positive dividend pay-out policy are that the estimated
coefficient in the case of a dividend inclusive model will be biased upwards, according to
the order of the firm‟s dividend yield. This bias arises because of the asymmetric effect of
26
dividends on dependent and independent variables.7
This approach is applied in this paper to yearly as well as quarterly estimating
windows. After completing contemporaneous estimates, the estimation interval is
lengthened and the new coefficients are estimated over a longer measurement interval. It
has been observed in the literature that the relation is more pronounced as the lag
increases.
In the next step the independent variable is modified, and earnings are summed for
(t-) periods. This measure should reduce the proportion of unexpected earnings variation,
as shown by Easton et al. (1992), Kothari and Sloan (1992) and others.
The equation (1) then changes to:
t
t
tt
t
t
P
X
P
P








,
(2)
In the numerator on the right-hand side of equation (2), earnings are summed over
(t-) periods.
In the case of both model (1) and (2), however, there is a constraint that the sample
size reduces as the price earnings measurement window increases due to the short time
series of yearly. Both regression models were repeated using quarterly data; but in
emerging capital markets a reference period of less than a year may not be long enough to
bring out the responses between variables. Conversely, in an emerging market the
environment changes rapidly, so that it may be quite different from one observation to
another, if the interval between observations is too long.
b) Alternative model specifications
In order to explore the nature of the relationship between prices and earnings in this
emerging market two other different model specifications have been tested in the similar
fashion using one- to four-year estimation window.
The model testing the relation between earnings changes and price changes
27
deflated by price at the beginning of the testing period is expressed as:
t
t
t
t
t
P
X
P
P







(3)
Price change is regressed on earnings change deflated by the price at the beginning of the
previous period. It would be expected that the resulting coefficient would be smaller than in
the level specification, ie it would be biased towards zero. 8
Another specification tests the relation running this regression when earnings were
deflated by the preceding years' earnings.
t
t
t
t
t
X
X
P
P




(4)
A problem arises with this type of regression, when company earnings are negative. If
earnings were negative in the denominator, the absolute value of earnings was used as in
Kothari (1992). It is expected that earnings deflated model is more biased than price
deflated.
These two alternative models have also been tested using pooled regression with three
different estimation methods, and their results were compared.
5. Description of the data
Data obtained from a sample of 63 Czech publicly traded industrial companies for
the period 1993-98 is used in this study. The accounting data have been taken from the
database of the Washington-based research firm PlanEcon, whose data are often used by
brokerage firms and other companies in the Czech Republic‟s financial industry. The
security prices are downloaded from Datastream. For cross-sectional time series analysis,
the econometric package EViews (previously TSP) as developed by Quantitative Micro
Software in Irvine (California), is used.
28
Sample and descriptive estimates
The choice of representative companies was driven by the availability of corporate
data and security prices. In 1993 only 33 of the companies from the sample were listed on
the Prague Stock Exchange and in 1994 only 38, so that the time series for the remaining
25 companies is shorter by one year (four quarters) and for 5 companies by 2 years. In
1998 only 60 companies published their earnings. In three cases, therefore, the time
series is shorter by the final year. More details of the number of observations for each year
and each estimation window are included in Table 1 in the Appendix.
In addition to the yearly estimates, the relationship between accounting data and
market price returns using quarterly data was measured. Prices were available from the
end of 1993 for 33 companies, from the first quarter of 1994 for 38 companies and from the
first quarter of 1995 for 63 companies. The following earnings numbers were available: 61
for the end of 1993, 63 from the first quarter of 1994 and only 58 for 1998. Details of the
number of observations for each quarterly window are included in Table 2 in the Appendix.
In this sample, there is a sizable subset of utilities (22 in total). The rest of the
sample represents a selection of companies from various industries: pulp and paper (1),
trade (2), machinery (7), glass and ceramics (4), transport (2), construction (4), chemicals
(6), electronics (3), mining (4), food processing (5), and metallurgy (3). More detailed
analysis distinguishing between the industries was not attempted, because the resulting
sample would be too small and the conclusions would not be valid or at best tenuous.
Descriptive statistics for earnings yields and price relatives computed from yearly
data, from the end of 1993 until the end of 1998, and reported for measurement intervals
ranging from one to four years are summarised in Table 1 at the end of the paper.
The first and second Panels of Table 1 contain descriptive statistics for earnings
yield. In the first Panel, both the mean values and medians increase with time; however,
variability measured by standard deviation is much higher than in the Kothari and Sloan
(1992) study. In Panel 2 non-cumulative earnings yield show an almost steady, modestly
29
increasing mean over the years. The median, on the other hand, fell by about 20 per cent
in value between 1 and 4 year lags. Unlike in Panel 1 the standard deviation decreases
with the smaller number of observations. In Kothari and Sloan (1992), mean as well as
median earnings yields increase monotonically with .
Panel 3 of Table 1 contains descriptive statistics about price relatives, including
dividends. Variability increases with the measurement interval (). After annualising, a
decreasing yearly return is apparent. The annualised return is fairly steady in Kothari and
Sloan (1992) at the level of 16.8-17% p.a. In the Czech case, the value varies between
3.3% and -3.1% with increasing see Panel 3, last column). When dividends are
disregarded (Panel 4, last column), the annualised mean is positive only in the case of the
one-year window, otherwise the annualised mean is negative, falling to -4.8 per cent in the
four-year window. This shows that returns were declining on the Czech market during the
period 1993-98. Another indicator of falling returns and declining prices is the median value
of price relatives, which is less than one and decreasing for price relatives in both dividendinclusive
and dividend-exclusive cases. This decline may partly be due to problems created
by illiquidity on the Czech stock market (PSE), weak corporate governance, lack of
transparency and other infrastructural shortcomings.
The mean price relative (in Panel 4, second column) decreases with the
measurement window interval. It ranges from just under +1.9% in the shortest
measurement interval containing no lags to almost -17.9% over the four-year period. The
median price relative reported in the fourth column of the last panel decreases with the
measurement interval. This indicates that most share prices were falling during the period
of measurement.
30
6. Regression results and implications
a) Length of the response window
Yearly models
Pooled regressions constrain the earnings response coefficient to a cross-sectional
constant, so that the explanatory power of the model could be expected to be rather low
(Teets and Wasley, 1996). This is particularly true in the case of contemporaneous
estimates. As the response window lengthens, the explanatory power of the model,
measured by R2
, improves. Similar conclusions were found in Collins and Kothari (1989),
Easton et al. (1992), and Kothari and Sloan (1992), among others. To take into
consideration the possible influence of firm specific characteristics the model was tested in
three estimation forms: the pooled regression with common intercept, with fixed effects,
and with random effects.9
Subsequently, the results were compared – see Table 2 at the
end of the paper.
Results of the pooled univariate model (1) are reported in Table 2. All coefficients
are significant at the one per cent level, except of the case of four-year estimation window
with fixed effect specification. The ERC coefficients are similar using all three methods.
The explanatory power measured by adjusted R2
is increasing as the window increases.
The earnings response coefficient measured using the pooled least squares with
common intercepts increases from 0.56 for the contemporaneous regression to 1.13, when
one-year-leading returns are included, to 1.79 when returns over two-leading-years are
included and to 3.61 for the four-year estimation window. All three estimation methods give
similar results. The results in this section suggest that estimates of the earnings response
coefficient increase with the inclusion of leading-period returns in accordance with the
hypothesis tested. Amongst the estimation methods the model which uses random effects
in each window has the highest explanatory power. The coefficients are larger if dividends
are taken into consideration.
31
Cumulative models
When earnings are aggregated and earnings measurement window increased the variation
of results should be smaller. In the Table 2 results were significant for each estimation
window. The results of cumulative model (2) are in Table 3 at the end of the paper.
In Table 3 the prices are regressed on aggregated earnings. All ERC coefficients
are significant and growing as the estimation window increases, however there are
apparent differences between individual estimation methods. In the case of
contemporaneous estimates the explanatory power is very low (the same as in Table 2).
The estimated coefficients using fixed effects are larger than the other two methods for
windows longer than one year. The model with random effects has the biggest explanatory
power as measured by adjusted R2
. Generally the models with fixed and random effect
have bigger explanatory power than the first model with common intercepts. As in the
previous table the coefficients are larger if dividends are taken into consideration. The level
of significance of individual coefficients is higher than in the previous model, which implies
that the variability of estimates is lower.
Quarterly models
After completion of the annual estimates, the quarterly results were assessed.
Pooled data was used again in estimating the common slope coefficient according to
equation (1). 10
The first Panel of Table 4 (see Table 4 at the end of the paper) contains estimates
for non-cumulative models. Results of pooled regression suggest that there is a highly
significant relationship between returns and accounting results on a quarterly
contemporaneous basis the highest value of the coefficient being 1.49 in the case of
estimation with common intercept. The pattern of these quarterly coefficients, however, is
rather puzzling. For two quarters the coefficient drops to 0.17, while for the three-quarters
32
window it rises to 0.40. All estimates are significant at the one per cent level. The only
model, which is not statistically significant is ERC estimate for two quarters using the fixed
effects estimator. The coefficient R2
is very low for two- and three-quarter estimation
windows.
The results of cumulative models are in Panel 2. All results are significant at the one
per cent level and follow similar pattern as in Panel 1. The highest R2
is again in onequarter
window. The coefficient of determination is rather low for two- and three-quarter
estimation windows, but it is slightly higher than in non-cumulative model, which is in
accordance with previous literature.
These findings could lead to the conclusion that the relation holds in the longer run, more
than one year, but not in the short run. Prices do not lead accounting data in the short run –
up to one year. The reasons for this may be inter alia that the Czech quarterly results are
not fully reliable and so they may not be entirely value relevant. 11
b) Alternative model specifications
In order to capture fully the relationship between accounting earnings and returns
on the Czech market, other model specifications were tested: a model using earnings
changes rather than levels (the „difference model‟):
(Pt-Pt-)/P t- =  +  (X t -X t-)/P t- + t; (3)
and a model using different deflator - past earnings (the „earnings deflator model‟):
Pt/P t- =  +  X t / X t- + t; (4)
The descriptive statistics of additional variables in these models for estimation
period ranging form one- to four-year estimation window, are shown in Table 5 at the end
of this paper.
Panel 1 of Table 5 contains the characteristics of earnings yield measured in
changes, and Panel 2 provides characteristics for earnings deflated by earnings at the
33
beginning of the testing period. The values are generally rather low and the standard
deviations are very large. The median value of earnings yield measured in changes is
basically equal to zero. In the case of the second specification, which uses the previous
year's earnings as a deflator, the median value is bellow one for all measurement windows
periods and the mean value is higher than one only in one-year window. 12
The third and fourth panels of Table 5 enclose price relatives in their change
specification including and excluding dividends respectively. The differences between the
characteristics in Panels 3 and 4 are negligible. Therefore it can be seen that there is no
substantial difference whether dividend yield is taken into consideration or not. This is
consistent with prior research (eg Collins and Kothari, 1989). Judging from the median
rather than the mean values, price relatives measured in changes are slightly negative in
all measurement intervals. This indicates that prices were falling during the sample
period.13
But again, the standard deviations are rather large which casts some doubts upon
the reliability of the estimates.
The results of pooled regression for model specification in changes are reported in
Table 6 at the end of the paper.
The estimates in Panel 1 include dividends while estimates in Panel 2 they do not.
For one-year estimation window this model is not significant using any estimator. It is
significant for two and three-year estimation window in regression with common intercept
and random effects. Generally, the coefficient is biased towards zero, which is in
accordance with previous literature. The explanatory power of this specification is generally
low. The coefficients differ slightly with different estimation methods, but during the first
three periods they follow the usual rising pattern. The negative coefficient for a four-year
window is puzzling and statistically insignificant in spite of relatively high adjusted R2
in the
model with random effects estimator.
The last model that has been tested is the specification, which uses previous
34
earnings as a deflator instead of prices. Results of this test are in Table 7 at the end of the
paper.
The ERC coefficients in this model are insignificant apart from one case: the threeyear
measurement interval using regression with fixed effects. The results in both panels of
Table 7 show that the intercept coefficient is equal to one in the first three windows and it is
highly significant, whereas the slope coefficient is practically always equal to zero and not
statistically significant. The coefficient of determination is basically equal to zero in the first
two intervals, which implies that the behaviour of the Czech capital market cannot be
adequately explained by this model specification.
7. Conclusions and implications for future research
This study has investigated the nature of the relationship between accounting earnings and
returns on the Czech market. It has been conducted using a data sample covering the
years 1993-1998.
The results of pooled regression models in levels suggest that:
a) for a short estimation window of up to three-quarters, there is a statistically significant
relationship between earnings-to-price ratios and price relatives. However, the
coefficients do not behave as expected; the relation is not monotonic.14
b) if lagged variables and a longer response window of one year or more are employed,
however, it can be seen that the postulated relationship holds and that it is significant at
the 1% confidence level.
Alternative model specifications of earnings response coefficients were tested in the
second part of this study. The results of pooled regressions suggest that:
c) the model using specification in changes was biased towards zero and not statistically
significant in the one and four-year windows.
d) the model using past earnings as a deflator of current earnings, was not statistically
significant.
35
A number of provisos need to be added. During the period under review (1993-98),
prices on the Prague stock exchange were mostly falling. This trend may have influenced
the earnings / return relation, which has been central to this analysis. Such a trend is
unlikely to persist and, when a similar analysis is repeated later one can expect to find that
more pronounced and stable results obtain.
Secondly, the analysis has been performed on a relatively short time series, and
due to that the sample size decreased as the lag between observed price response
coefficients increased. Note that, in the case of a 4-year interval, there are only 71
observations, so that no definitive conclusions can be drawn from it.
A fruitful direction for further research would include replicating similar models on a
longer time series to see whether these results continue to hold for the Czech capital
market. It could be also tested whether different relation exists for loss-making and profitmaking
firms. It would also be appropriate to examine whether incorporation of other
financial statement data in addition to accounting earnings would enhance the explanatory
power of the model.
36
Notes
1
Privatisation „vouchers‟ represented the artificial one-purpose money that could be exchanged only
for shares; these vouchers were not tradable.
2
This was a one-off mechanism for distribution of shares in the Voucher Privatisation in the early
1990s. More details on the actual distribution mechanics used in the privatisation can be found in
(Hingorani et al., 1997). Now, the RM-System works like an over-the–counter market with real
money and real shares.
3
See, for example, IMF, Czech Republic – 1999, Article IV Consultation Mission Concluding
Statement, April 16, 1999, International Monetary Fund.
4
Firms-specific analysis was also performed, but due to the limited number of observations of
earnings and prices for each company and resulting limited explanatory power of models the details
were not included in this paper.
5
Dividends and dividend yields are adjusted for stock splits. Generally, dividends tend to be small
and irregular, and few companies in the Czech Republic pay them, the major reason for this
appearing to be double taxation of dividends. This question requires more in-depth research, and it
is not central to this analysis.
6
For contemporaneous estimates =1.
7
According to Kothari and Sloan (1992), earnings in the Xt /Pt- variable reflect earnings on the firm‟s
initial equity investment and on the increase in investments over time through earnings retention.
But Xt does not reflect earnings on past dividend payments. The return variable Pt /Pt-, on the other
hand, assumes dividends are reinvested.
8
According to Kothari and Zimmerman (1995), Christie (1987) and Donnelly and Walker (1995),
models in levels and models in changes have the same economic interpretation. However, the
return models have fewer econometric problems.
9
Both fixed effects and random effects models are used to consider the possible influence of firm
specific characteristics. Fixed effect model treats intercepts as fixed separately for each individual
firm. Random effect model assumes that intercepts of the individual firms are different but they can
be treated as drawings from some underlying distribution. These drawings are assumed to be
independent from Xit. The fixed effect model is conditional upon the values for i. It considers the
37
distribution of Yit given i where is can be estimated. It can be used if the individuals in the sample
are of one kind, and cannot be viewed as a random draw from some underlying population. The
random effect approach allows one to make inference with respect to the population characteristics.
However, if the Xit and i are correlated, the random effect model, ignoring this correlation, leads to
inconsistent estimators. (see Verbeek, 2000: 318-9)
10
Table 4 contains results of non-cumulative and cumulative quarterly models which include
dividends. The results excluding dividends are not presented because there were no apparent
differences due to a very short estimation period involved.
11
Although the Czech companies listed on the Prague Stock Exchange are obliged to publish the
quarterly accounting results this data is not audited, therefore it may not form price relevant
information.
12
The earnings pattern is that in later years there were more losses in the sample. For example,
there were two loss-making companies out of 63 in 1994, five out of 63 in 1995, nine out of 63 in
1996, ten out of 63 in 1997 and seven out of 60 in 1998. In 1993 there were ten loss-making
companies out of 63, but this is immaterial to this analysis because, where the deflator in the fourth
model is negative its absolute value has been used (see Kothari 1992). (Note: There are no specific
tests in this study about loss-making versus profit-making companies.)
13
The price decline, however, is more visible in Table 1 Panels 3 and 4, where the price relatives
are computed in levels.
14
The model was also estimated as firm specific but due to the limited number of observations
detailed analysis was not included in this paper. In the case of firm-specific estimates, the
differences are much smaller but the general pattern of non-monotonicity remains the same.
38
References
Ball, R. and Brown, P. (1968) ‘An Empirical Evaluation of Accounting Income Numbers’,
Journal of Accounting Research, 6: 159-78.
Basu, S. (1997) ‘The Conservatism Principle and the Asymmetric Timeliness of Earnings’,
Journal of Accounting and Economics, 24: 3-37.
Booth, G. G., Broussard, J. and Loistl, O. (1997) ‘Earnings and Stock Returns: Evidence
from Germany’, The European Accounting Review, 6(4): 589-603.
Christie, A. A. (1987) ‘On Cross-sectional Analysis in Accounting Research’, Journal of
Accounting and Economics, 9: 231-58.
Collins, D. W. and Kothari, S. P. (1989) ‘An Analysis of Intertemporal and Cross-sectional
Determinants of Earnings Response Coefficients’, Journal of Accounting and Economics, 11:
143-81.
Collins, D. W., Kothari, S. P. and Rayburn, J. D. (1987) ‘Firm Size and the Information Content
of Prices with Respect to Earnings’, Journal of Accounting and Economics, 9: 111-38.
Dechow, P. M., (1994) ‘Accounting Earnings and Cash Flows as a Measure of Firm
Performance: The Role of Accounting Accruals’, Journal of Accounting and Economics, 18:
3-42.
Donnelly, R. and Walker, M. (1995) ‘Share Price Anticipation of Earnings and the Effect of
Earnings Persistence and Firm Size’, Journal of Business Finance and Accounting, 22: 5-18.
Easton P. D. and Harris, T. S. (1991) ‘Earnings as an Explanatory Variable for Returns’,
Journal of Accounting Research, 29: 19-36.
Easton, P. D., Harris, T. S. and Ohlson, J. A. (1992) ‘Aggregate Accounting Earnings can
Explain Most of Security Returns’, Journal of Accounting and Economics, 15: 119-42.
Easton, P. D. and Zmijewski, M. D. (1989) ‘Cross-sectional Variation in the Stock Market
Response to Accounting Earnings Announcements’, Journal of Accounting and Economics,
11: 117-41.
39
Filacek, J., Kapicka, M. and Vosvrda, M. (1998) ‘Efficiency Market Hypothesis: Testing on
the Czech Capital Market’, Finance a uver, 48(9): 554-66.
Hingorani, A., Lehn, K. and Makhija, A. K. (1997) ‘Investor Behaviour in Mass Privatisation:
The Case of the Czech Voucher Scheme’, Journal of Financial Economics, 44: 349-96.
IMF (1999) ‘Czech Republic – 1999 Consultation Mission, Concluding Statement’, April 16,
1999.
Kormendi, R. and Lipe, R. C. (1987) ‘Earnings Innovations, Earnings Persistence and Stock
Returns’, Journal of Business, 60: 323-45.
Kothari, S. P. (1992) 'Price-earnings Regressions in the Presence of Prices Leading
Earnings', Journal of Accounting and Economics, 15: 173-202.
Kothari, S. P. and Sloan, R. G. (1992) ‘Information in Prices about Future Earnings’, Journal
of Accounting and Economics, 15: 143-71.
Kothari, S. P. and Zimmerman, J. L. (1995) ‘Price and Return Models’, Journal of Accounting
and Economics, 20: 155-92.
Martikainen, T., Kallunki, J-P. and Perttunen, J. (1997) ‘Finnish Earnings Response
Coefficients: the Information Content of Losses’, The European Accounting Review, 6(1): 69-
81.
Mejstrik, M. (ed) (1997) The Privatization Process in East-Central Europe, Kluwer Academic
Publishers, Norwell, Massachusetts.
Ohlson, J. A. (1991) ‘The Theory of Earnings and an Introduction to the Ball-Brown
Analysis’, Contemporary Accounting Research, 8: 1-19.
Patton, J. and Zelenka, I. (1997) 'An Empirical Analysis of the Determinants of the Extent of
Disclosure in Annual Reports of Joint Stock Companies in the Czech Republic', European
Accounting Review, 6: 605-26.
40
Sommer, M. (1996a) ‘Market Model on the Czech Equity Market’, Prague Economic Papers,
3: 12-48.
Sommer, M. (1997b) ‘Beta as an Indicator of the Risk on the Czech Capital Market’, (part 1),
Finance a uver, 7: 397-402, [Translation from Czech].
Sommer, M. (1997) ‘Beta as an Indicator of the Risk on the Czech Capital Market’, (part 2),
Finance a uver, 8: 463-76 [Translation from Czech].
Svejnar, J. (ed) (1995) The Czech Republic and Economic Transition in Eastern Europe San
Diego, Academic Press.
Teets, W. R. and Wasley Ch. E. (1996) ‘Estimating Earnings Response Coefficients: Pooled
versus Firm-Specific Models’, Journal of Accounting and Economics, 21: 279-95.
Verbeek, V. (2000) A Guide to Modern Econometrics. Chichester: John Wiley and Sons.
White, H. (1980) ‘A Heteroscedasticity Consistent Covariance Matrix Estimator and a Direct
Test for Heteroscedasticity’, Econometrica, 48: 817-38.
41
Appendix Table 1
Number of observations in yearly models. The table provides a breakdown of number of
observations for each measurement window ranging form one to four years.
Number of firm-years
year No of Xt No of Pt =1 =2 =3 =4
1993 61 33 33 33 33 33
1994 63 38 38 38 38 38
1995 63 63 63 63 60
1996 63 63 63 60
1997 63 63 60
1998 60 63
Total 257 194 131 71
a
Xt is annual earnings per share before extraordinary items for year t.
b
Pt is a share price at the end of year t.
c
 is a length of the measurement period ranging from one to four years.
Number of estimations in each interval is determined as a minimum number of observations
of either prices or earnings in the estimated models.
(1) Pooled model Pt/P t- =  +  X t /P t- + t
(2) Cumulative model Pt/P t- =  +  X t-t /P t- + t
(3) Difference model (Pt-Pt-1)/P t-1 =  +  (X t -X t-1)/P t-1 + t
(4) Earnings deflator model Pt/P t-1 =  +  X t / X t-1 + t
This more detailed explanation starts with the longest four-year window:
For the four-year estimation window there were only 33 prices available (end of the year
1993); hence there are only 33 observations comparing the earnings from 1997 with the final prices of
1993 and 38 observations comparing earnings from 1998 and prices from 1994. In total there are 71
estimates for the four-year estimation window.
For the three-year window there are 33 prices available for the year 1993, 38 prices for the
year 1994, and 63 prices for the year 1995. However, in this estimation window, the prices from 1995
are compared with earnings of 1998 and there are only 60 observations of earnings at the end of
1998. Altogether there are 131 observations for the three-year window.
In case of a two-year window the limitation is similar. Firstly the observations are limited by
availability of prices for 1993 and 1994 and in the final year 1998 the limiting factor is earnings.
Altogether there are 194 observations for the two-year window.
In case of one-year window there are 258 observations. The limitations are the same as
above.
The same number of observations indicated in this table is used in measuring the earnings
response coefficient in all yearly pooled models. The results are summarised in tables 2, 3, 6 and 7.
42
Appendix Table 2
Number of observations in quarterly models. This table provides a breakdown of number of
observations for each measurement window ranging form one to three quarters.
Number of firm-quarters
Quarter No of Xt No of Pt  = 1  = 2  = 3
1993:4 63 33 33 33 33
1994:1 63 38 38 38 38
1994:2 63 38 38 38 38
1994:3 63 38 38 38 38
1994:4 63 38 38 38 38
1995:1 63 63 63 63 63
1995:2 63 63 63 63 63
1995:3 63 63 63 63 63
1995:4 63 63 63 63 63
1996:1 63 63 63 63 63
1996:2 63 63 63 63 63
1996:3 63 63 63 63 63
1996:4 63 63 63 63 63
1997:1 63 63 63 63 63
1997:2 63 63 63 63 58
1997:3 63 63 63 58 58
1997:4 63 63 58 58 58
1998:1 58 63 58 58 58
1998:2 58 63 58 58
1998:3 58 63 58
1998:4 58 63
Total 1110 1047 984
a
Xt is quarterly earnings per share before extraordinary items for year t.
b
Pt is a share price at the end of quarter t.
c
 is a length of the measurement period ranging from one to three quarters.
Number of estimations in each interval is determined as a minimum number of observations
of either prices or earnings in the estimated model.
1) Quarterly pooled model Pt/P t- =  +  X t /P t- + t
(2) Cumulative model Pt/P t- =  +  X t-t /P t- + t
More detailed explanation starts with the longest three-quarter window:
There are 984 firm quarters on three-quarter window models. Limiting factors are initially
number of observations of prices in 1993:4 when only 33 prices were available. In all four quarters of
1994 there were only 38 prices available. In all four quarters of 1998 there were earnings for only 58
companies available. For the rest of the quarters there were 63 observations available.
In two-quarter window models there are 1047 observations available. The limitations are at
first number of prices in 1993:4 and in all four quarters of 1994 where only 38 prices were available. In
1998 there were earnings for only 58 companies available. Otherwise 63 observations were available.
In one-quarter model there are 1110 observations available. The limitations are the same as
above.
This number of observations is used in measuring the earnings response coefficient in the
quarterly pooled models. The results are summarised in table 4.
43
Table 1
Descriptive statistics for earnings yield cumulative (X t- ,t /P t- ), and non-cumulative (X t /P t- ) and price
relative ( Pt/P t- ). Variables have been measured over the period of one up to four years using data over the
period 1993-1998. Sample sizes vary from 257 to 71 firm-years. Price relatives distinguish between whether
dividends are taken into consideration (Panel 3), and excluded from observations (Panel 4). e
Panel 1
Earnings yield cumulative
a
Period length MEAN STDEV MEDIAN MIN MAX COUNT
1 year (no lag) 0.030 0.309 0.050 -3.765 1.150 257
2 years 0.058 0.485 0.095 -4.164 2.531 194
3 years 0.128 0.544 0.149 -2.461 3.464 131
4 years 0.193 0.670 0.179 -2.461 3.491 71
Panel 2
Earnings yield non-cumulative
b
Period length MEAN STDEV MEDIAN MIN MAX COUNT
1 year (no lag) 0.030 0.309 0.050 -3.765 1.150 257
2 years 0.034 0.253 0.045 -1.864 1.382 194
3 years 0.048 0.194 0.048 -1.058 0.933 131
4 years 0.049 0.104 0.032 -0.229 0.477 71
Panel 3
Price relatives including dividend yield
c
Period length MEAN STDEV MEDIAN MIN MAX COUNT an. mean
1 year (no lag) 1.033 0.697 0.919 0.172 8.827 257 1.033
2 years 0.990 0.862 0.841 0.076 9.723 194 0.995
3 years 1.001 0.958 0.809 0.110 7.521 131 1.000
4 years 0.883 1.074 0.514 0.064 6.862 71 0.969
Panel 4
Price relatives excluding dividend yield
d
Period length MEAN STDEV MEDIAN MIN MAX COUNT an. mean
1 year (no lag) 1.019 0.684 0.905 0.172 8.581 257 1.019
2 years 0.961 0.819 0.814 0.076 9.061 194 0.980
3 years 0.957 0.909 0.762 0.110 7.009 131 0.986
4 years 0.821 1.009 0.482 0.063 6.394 71 0.952
a
Earnings yield (X t- ,t /Pt- ) annual earnings per share before extraordinary items over the years t -  to t
divided by the price at the beginning of the return measurement interval.
b
Earnings yield (X t /P t- t divided by the price
at the beginning of the return measurement interval.
c
Price relative (Pt /P t- ) one plus buy and hold return inclusive (and exclusive respectively) of dividends over
the years t -  to t.
All earnings per share numbers, dividends and prices are adjusted for stock splits.
d
Annualized price relative is calculated as the geometric mean annual price relative for the particular
measurement interval.
e
Details about number of observations for each estimation window are in Table 1 in the Appendix.
44
Table 2
Estimated parameters of pooled regression of price relatives on earnings deflated by price. Earnings
measurement interval is one year and price relatives are for contemporaneous one year and for periods
that also include leading years. Sample sizes vary from 257 to 71 firm-years. Variables have been
measured using pooled regression with three different estimation methods over the period of one up to
four years using data over the period 1993-1998.
Pt/P t- =  +  X t /P t- + t
a,b,c
Yearly models (estimates include dividends)
period
length    = 3  = 4
estimation
method
d
CI FE RE CI FE RE CI FE RE CI FE RE
intercept
e
1.017 1.017 0.939 0.939 0.921 0.922 0.710 0.710
(24.082) (27.638) (16.322) (16.214) (11.412) (10.492) (5.282) (4.918)
ERC 0.562 0.557 0.563 1.133 1.113 1.133 1.788 1.607 1.764 3.609 3.485 3.600
(4.106) (3.081) (4.272) (4.991) (3.613) (4.983) (4.382) (2.520) (4.241) (3.056) (1.359) (2.927)
adj R2 0.058 -0.041 -0.092 0.110 0.056 0.119 0.122 0.127 0.316 0.108 0.240 0.390
N 257 257 257 194 194 194 131 131 131 71 71 71
Yearly models (estimates exclude dividends)
intercept 1.004 1.004 0.925 0.925 0.885 0.886 0.666 0.666
(24.195) (27.591) (16.510) (16.199) (11.475) (10.392) (5.237) (4.832)
ERC 0.540 0.539 0.540 1.073 1.053 1.073 1.628 1.441 1.598 3.237 3.171 3.231
(4.012) (3.039) (4.157) (4.849) (3.523) (4.828) (4.177) (2.398) (4.019) (2.897) (1.329) (2.767)
adj R2 0.055 -0.043 -0.087 0.104 0.057 0.130 0.112 0.140 0.340 0.097 0.255 0.416
N 257 257 257 194 194 194 131 131 131 71 71 71
a
Earnings yield (X t /P t-annual earnings per share before extraordinary items for year t divided by
the price at the beginning of the return measurement interval.
b
Price relative (Pt /P t-) one plus buy and hold return inclusive (exclusive respectively) of dividends
over the years t -  to t.
c
Details about number of observations for each estimation window are in Table 1 in the Appendix.
d
The three estimation methods are pooled regression with (1) common intercept (CI), (2) fixed effects
(FE) and (3) random effects (RE).
e
t - values are in parentheses.
Details about number of observations for each estimation window are in Table 1 in the Appendix.
All earnings per share numbers, dividends and prices are adjusted for stock splits.
45
Table 3
Estimated parameters of pooled regression of price relatives on earnings deflated by price. Earnings
measurement interval ranges from one to four years and price relatives are for contemporaneous one
year and for periods that also include leading years. Sample sizes vary from 257 to 71 firm-years.
Variables have been measured using pooled regression with three different estimation methods over the
period of one up to four years using data over the period 1993-1998.
Pt/P t- =  +  X t-t /P t- + t
a,b,c
Yearly cumulative models (estimates include dividends)
period
length    = 3  = 4
estimation
method
d
CI FE RE CI FE RE CI FE RE CI FE RE
intercept
e
1.017 1.017 0.932 0.928 0.878 0.855 0.588 0.525
(24.082) (27.638) (16.652) (15.175) (11.803) (9.531) (5.998) (4.305)
ERC 0.562 0.557 0.563 0.733 1.403 0.788 0.930 1.620 1.091 1.294 1.905 1.565
(4.106) (3.081) (4.272) (6.058) (6.635) (6.208) (6.830) (7.292) (7.533) (8.328) (9.978) (10.512)
adj R2 0.058 -0.041 -0.092 0.155 0.223 0.269 0.258 0.462 0.593 0.498 0.809 0.879
N 257 257 257 194 194 194 131 131 131 71 71 71
Yearly cumulative models (estimates exclude dividends)
intercept 1.004 1.004 0.919 0.915 0.845 0.822 0.550 0.489
(24.195) (27.591) (16.832) (15.204) (11.806) (9.414) (5.831) (4.144)
ERC 0.540 0.539 0.540 0.697 1.337 0.754 0.849 1.492 1.011 1.187 1.763 1.453
(4.012) (3.039) (4.157) (5.904) (6.491) (6.071) (6.482) (7.048) (7.254) (7.932) (9.817) (10.264)
adj R2 0.055 -0.043 -0.087 0.149 0.219 0.273 0.238 0.458 0.599 0.473 0.808 0.881
N 257 257 257 194 194 194 131 131 131 71 71 71
a
Earnings yield (X t-t /P t-is earnings per share before extraordinary items are summed over the
relevant measurement interval (t - and then divided by the price at the beginning of the return
measurement interval.
b
Price relative (Pt /P t-) is one plus buy-and-hold return inclusive (exclusive respectively) of dividends
over the interval t -  to t.
c
Further details about number of observations for each estimation window are in Table 1 in the
Appendix.
d
The three estimation methods are: pooled regression with (1) common intercept (CI), (2) fixed effects
(FE) and (3) random effects (RE).
e
t - values are in parentheses.
All earnings per share numbers, dividends and prices are adjusted for stock splits.
46
Table 4
Estimated parameters of pooled regression of price relatives on earnings deflated by price. Earnings
measurement interval in Panel 1 is one quarter, and in Panel 2 ranges from one to three quarters. Price
relatives are for contemporaneous one quarter and for periods that also include leading quarter. Sample
sizes vary from 1110 to 984 firm-quarters. Variables have been measured using pooled regression with
three different estimation methods over the period of one up to three quarters using data over the period
1993-1998.
Pt/P t- = t + t X t /P t- + t and
Pt/P t- =  +  X t-t /P t- + t
a,b,c,d
Panel 1
Quarterly models (estimates include dividends)
period
length  = 1  = 2  = 3
estimation
method
e
CI FE RE CI FE RE CI FE RE
intercept
f
1.004 1.003 0.998 0.998 1.003 1.003
(63.660) (122.637) (73.210) (63.260) (54.001) (40.501)
ERC 1.489 1.505 1.459 0.169 0.066 0.144 0.399 0.267 0.342
(40.780) (37.592) (45.998) (4.273) (1.579) (3.618) (6.478) (4.179) (5.535)
adj R2 0.601 0.583 0.527 0.016 0.059 0.061 0.040 0.100 0.123
N 1110 1110 1110 1047 1047 1047 984 984 984
Panel 2
Quarterly cumulative models (estimates include dividends)
intercept 1.004 1.003 0.998 0.998 1.001 1.001
(63.660) (122.637) (76.043) (64.017) (55.939) (40.129)
ERC 1.489 1.505 1.459 0.307 0.254 0.293 0.397 0.352 0.379
(40.780) (37.592) (45.998) (10.120) (7.426) (9.438) (10.991) (8.262) (9.858)
adj R2 0.601 0.583 0.527 0.089 0.107 0.125 0.109 0.146 0.178
N 1110 1110 1110 1047 1047 1047 984 984 984
a
Earnings yield (X t /P t-annual earnings per share before extraordinary items for year t divided by
the price at the beginning of the return measurement interval.
b
Earnings yield (X t-t /P t-earnings per share before extraordinary items are summed over the
relevant measurement interval (t - and then divided by the price at the beginning of the return
measurement interval.
c
Price relative (Pt /P t-) is one plus buy-and-hold return inclusive of dividends over the interval t -  to
t.
d
Further details about number of observations for each estimation window are in Table 2 in the
Appendix.
e
The three estimation methods are: pooled regression with (1) common intercept (CI), (2) fixed effects
(FE) and (3) random effects (RE).
f
t - values are in parentheses.
All earnings per share numbers, dividends and prices are adjusted for stock splits.
47
Table 5
Descriptive statistics for two alternative models.
Descriptive statistics for earnings change deflated by the price at the beginning of the measurement
period ((Xt-Xt-)/Pt-), and earnings deflated by the earnings at the beginning of the measurement period
(Xt/Xt-) are in Panel 1 and 2 respectively. Panel 3 contains descriptive statistics for price change
including dividends ((Pt – Pt-)Ct /P t-) deflated by the price at the beginning of the period. Panel 4
contains descriptive statistics for price change excluding dividends ((Pt – Pt-) /P t-) deflated by the
price at the beginning of the period. All variables have been measured over the period of one year using
data over the period 1993-1998. Sample sizes vary from 257 to 71 firm-years. e
Panel 1
Earnings difference variable ((Xt-Xt-)/Pt-)
a
period length MEAN STDEV MEDIAN MIN MAX N
1 year (no lag) 0.013 0.455 -0.001 -3.767 5.423 257
2 years -0.016 0.265 -0.004 -2.032 1.051 194
3 years -0.027 0.178 -0.019 -1.063 1.050 131
4 years -0.032 0.107 -0.028 -0.346 0.355 71
Panel 2
Earnings deflated by previous year earnings (Xt/Xt- ) b
period length MEAN STDEV MEDIAN MIN MAX N
1 year (no lag) 1.550 12.750 0.886 -62.051 163.305 257
2 years -0.340 8.831 0.834 -90.173 25.668 194
3 years 0.669 5.220 0.731 -39.384 36.246 131
4 years 0.657 9.223 0.457 -50.566 56.875 71
Panel 3
Price difference relatives including dividend yield ((Pt-Pt-)*Ct/Pt- ) c
period length MEAN STDEV MEDIAN MIN MAX N
1 year (no lag) 0.018 0.698 -0.096 -0.864 7.798 257
2 years -0.039 0.860 -0.193 -0.948 8.650 195
3 years -0.045 0.957 -0.260 -0.920 6.448 131
4 years -0.192 1.073 -0.553 -0.986 5.789 71
Panel 4
Price difference relatives excluding dividend yield ((Pt-Pt-)/Pt- ) d
1 year (no lag) 0.019 0.684 -0.095 -0.828 7.581 257
2 years -0.039 0.819 -0.186 -0.924 8.061 194
3 years -0.043 0.909 -0.238 -0.890 6.009 131
4 years -0.179 1.009 -0.518 -0.937 5.394 71
a
Earnings change deflated by the beginning of the year price ((Xt-Xt-)/Pt-). Annual earnings per share
are before extraordinary items for year t and t- respectively. The difference is deflated by the price at
the beginning of the measurement period.
b
Earnings deflated by previous year earnings (Xt/Xt-). Annual earnings per share before extraordinary
items for year t divided by the absolute value of the annual earnings per share at the beginning of
estimation period. Annual earnings per share are before extraordinary items for year t and t-1
respectively.
c
Price change inclusive of dividends divided by the price at the beginning of the measurement interval
((Pt – Pt-)Ct /P t-). Where Ct is one plus current yield (Dt/Pt-) i.e. dividend paid in year t divided by
price at the beginning of the year t.
d
Price change exclusive of dividends divided by the price at the beginning of the measurement interval
((Pt – Pt-) /P t-).
e
Details about number of observations for these yearly pooled models are in Table 1 in the Appendix.
All earnings per share numbers, dividends and prices are adjusted for stock splits.
48
Table 6
Difference model: Estimated parameters of pooled regression of price relatives on earnings deflated by
price. Both, prices and earnings are measured in first differences and deflated by price at the beginning
of the estimation period. Earnings measurement interval is one year and price relatives are for
contemporaneous one year and for periods that also include leading years. Sample sizes vary from 257
to 71 firm-years. Variables have been measured using pooled regression with three different estimation
methods over the period of one up to four years using data over the period 1993-1998.
(Pt-Pt-)/P t- =  +  (X t -X t-)/P t- + t
a,b,c
Difference models (estimates include dividends)
period
length    = 3  = 4
estimation
method
d
CI FE RE CI FE RE CI FE RE CI FE RE
intercept
e
0.020 0.020 -0.024 -0.024 0.015 0.014 -0.196 -0.212
(0.470) (0.519) (-0.400) (-0.400) (0.187) (0.157) (-1.429) (-1.410)
ERC 0.059 0.067 0.057 0.645 0.489 0.645 1.597 1.404 1.562 -0.137 -2.621 -0.598
(0.684) (0.726) (0.648) (2.793) (1.913) (2.794) (3.015) (1.942) (2.950) (-0.109) (-1.435) (-0.475)
adj R2 -0.002 -0.089 -0.113 0.034 -0.010 0.032 0.058 0.096 0.298 -0.015 0.238 0.395
N 257 257 257 194 194 194 131 131 131 71 71 71
Difference models (estimates exclude dividends)
intercept 0.021 0.021 -0.025 -0.025 0.014 0.013 -0.179 -0.195
(0.497) (0.548) (-0.431) (-0.428) (0.182) (0.145) (-1.387) (-1.372)
ERC 0.056 0.065 0.054 0.595 0.442 0.593 1.502 1.271 1.453 -0.046 -2.388 -0.518
(0.659) (0.710) (0.623) (2.702) (1.816) (2.694) (2.981) (1.868) (2.894) (-0.039) (-1.411) (-0.438)
adj R2 -0.002 -0.089 -0.112 0.031 -0.007 0.042 0.057 0.113 0.323 -0.015 0.260 0.421
N 257 257 257 194 194 194 131 131 131 71 71 71
a
Earnings yield ((X t -X t-)/P t-
divided by the price at the beginning of the return measurement interval.
b
Price relative ((Pt-Pt-)/P t-) is change of the share price inclusive (exclusive respectively) of
dividends over the interval t -  to t divided by the price at the beginning of the return measurement
interval.
c
Further details about number of observations for each estimation window are in Table 1 in the
Appendix.
d
The three estimation methods are: pooled regression with (1) common intercept (CI), (2) fixed effects
(FE) and (3) random effects (RE).
e
t - values are in parentheses.
All earnings per share numbers, dividends and prices are adjusted for stock splits.
49
Table 7
Earnings deflator model: Estimated parameters of pooled regression of price relatives on earnings
deflated by earnings at the beginning of the measurement period. Earnings measurement interval is one
year and price relatives are for contemporaneous one year and for periods that also include leading
years. Sample sizes vary from 257 to 71 firm-years. Variables have been measured using pooled
regression with three different estimation methods over the period of one up to four years using data
over the period 1993-1998.
Pt/P t- =  +  X t / X t- + t
a,b,c,
Earnings deflator models (estimates include dividends)
period
length    = 3  = 4
estimation
method
d
CI FE RE CI FE RE CI FE RE CI FE RE
intercept
e
1.037 1.037 0.996 0.983 1.010 1.009 0.872 0.872
(23.941) (26.992) (15.874 ) (15.878) (12.057) (10.869) (6.551) (5.954)
ERC 0.000 0.000 0.000 -0.001 0.000 0.001 0.000 -0.004 -0.001 0.006 0.066 0.006
(0.739) (-0.022) (0.921) (-0.132) (0.088) (0.881) (-0.069) (-0.799) (-0.283) (0.406) (0.507) (0.391)
adj R2 -0.002 -0.092 -0.142 -0.005 -0.039 0.038 -0.008 0.056 0.260 -0.012 0.201 0.374
N 257 257 257 194 194 194 131 131 131 71 71 71
Earnings deflator models (estimates exclude dividends)
intercept 1.023 1.023 0.967 0.968 0.965 0.885 0.808 0.807
(24.074) (27.069) (16.194) (15.931) (12.139) (17.311) (6.460) (5.830)
ERC 0.000 0.000 0.000 -0.001 0.000 0.001 0.000 -0.007 -0.004 0.007 0.068 0.007
(0.736) (-0.021) (0.913) (-0.117) (0.093) (0.878) (-0.110) (-2.631) (-1.665) (0.505) (0.556) (0.482)
adj R2 -0.002 -0.092 -0.137 -0.005 -0.032 0.053 -0.008 0.126 0.256 -0.011 0.221 0.398
N 257 257 257 194 194 194 131 131 131 71 71 71
a
Earnings yield (X t /X t-
the beginning of the return measurement interval. If earnings in denominator is negative the absolute
value is used.
b
Price relative (Pt /P t-) one plus buy and hold return inclusive (exclusive respectively) of dividends
over the years t -  to t.
c
Further details about number of observations for each estimation window are in Table 1 in the
Appendix.
d
The three estimation methods are: pooled regression with (1) common intercept (CI), (2) fixed effects
(FE) and (3) random effects (RE).
e
t - values are in parentheses.
All earnings per share numbers, dividends and prices are adjusted for stock splits.
50
Část 3
Accounting for Good News and Accounting for Bad News:
Some Empirical Evidence from the Czech Republic
[Účetnictví „Dobrých zpráv“ a účetnictví „Špatných zpráv“:
Empirické důkazy z České republiky]
Irena Jindrichovska & Stuart McLeay
European Accounting Review 14(2), 2005, 1-21.
51
Accounting for Good News and Accounting for Bad News: Some Empirical Evidence
from the Czech Republic
ABSTRACT
This paper is motivated by the links that continue to be forged between security pricing and
accounting, building on recent findings that firms tend to be asymmetrically conservative in the
timeliness of earnings recognition. The evidence is that firms in the European Union tend to recognise
unrealised losses more quickly in their earnings than unrealised gains (Giner and Rees, 2001; Raonic,
McLeay and Asimakopoulos, 2004), and there is evidence of even greater accounting conservatism in
the US (Basu, 1997; Ball, Kothari and Robin, 2000; Givoly and Hayn, 2001). This article
investigates whether the Czech market exhibits conformity with the behaviour that has been
documented elsewhere by examining the earnings/returns relationship, focusing to begin with on the
impact of losses on earnings response coefficients and then considering the asymmetric timeliness of
income recognition in the Czech market.
The findings indicate that the Czech market is similar to more developed markets, at least in one
respect: There is statistically significant evidence of different market effects of profits and losses, in
that profits are more persistent than losses. However, contrary to the findings in more developed
markets, there is no statistically significant evidence of earnings conservatism in the Czech market.
These results are most probably due to the continuing influence of restrictive tax regulations that
mitigate any tendency towards conservatism, as well as the transitional nature of the economy. A
further reason is likely to be that the regulatory environment in the Czech Republic is close to the kind
of stakeholder corporatism that is described by Ball et al. (2000), who show that conservatism tends to
be less pronounced in such regimes where there are fewer managerial incentives to bias current
earnings. In conclusion, if changes in market prices signal good news and bad news about future risky
outcomes, there is no evidence of asymmetry in the Czech market in accounting for such risks.
Keywords: asymmetry, conservatism, impact of losses, emerging markets, Czech capital market.
52
1. INTRODUCTION AND MOTIVATION
In developed markets, the impact of capital market regulations and growing risk aversion is
ever more traceable in accounting through the effect of increased conservatism of corporate
earnings (Givoly and Hayn, 2001). Since there has been relatively little research on the link
between accounting and the stock market in Central Europe, this paper looks at the current
situation regarding the accountability of listed Czech corporations in order to investigate this
issue in the context of an economy in transition.
The recent intensive move to develop standard stock markets and to introduce proper
financial reporting standards in transitional economies should create more constructive and
more informative environment for investors and allow them to gain greater insight into
corporate operations and results. Furthermore the accountability of corporate management to
investment public is reinforced. (Sucher and Jindrichovska, forthcoming).
The informationally efficient market implies that all relevant information is immediately
captured by prices, including expectations concerning future losses and future profits that are
anticipated from the risky ventures in which firms are engaged. Conservatism in accounting
for a firm‟s activities arises from the regulatory requirement that anticipated losses should be
incorporated in current earnings, whereas anticipated gains must be recorded only when
realised. Following Basu (1997), conservatism thus results in earnings reflecting bad news
more quickly than good news, as unfavourable information is more likely to be reflected fully
in current earnings than favourable information. It has also been recognised that the effect of
accounting losses differs in the earnings/returns relationship from that of profits, with losses
being less persistent than profits (Hayn, 1995), which is consistent with pervasive
conservatism.1
Indeed, conservative accounting has been shown to be an important feature of
developed markets (Pope and Walker, 1999; Ball, Kothari and Robin, 2000; Givoly and
53
Hayn, 2001; Giner and Rees, 2001; Raonic, McLeay and Asimakopoulos, 2004)
The relation between accounting earnings and returns has been examined in a number of
markets that are under transformation, including Poland (Jermakowicz and Gornik–
Tomaszewski, 1998) and the Czech Republic (Jindrichovska, 2001), and to some extent
Hungary (Chun, 2000). This research has concentrated however on different specifications of
the earnings/returns model, rather than accounting conservatism and the different effects of
accounting losses and profits. The present paper enhances the previous understanding of the
earnings/returns relationship by exploring the nature of the impact of negative accounting
earnings on earnings response coefficients in the Czech market, together with the different
effects of good and bad news.2
Given the long tradition of central planning, Czech accounting was not influenced by the
capital market until recently, but only by state regulations. Currently, Czech accounting is
very closely linked to tax rules (Holeckova, 1996). These are restrictive, detailed and specific,
and still do not allow companies to make material discretionary provisions nor to depart from
inflexible depreciation rules in spite of some recent changes in accounting law (Act on
Accounting, 2001). Given this institutional context, the present paper examines whether the
market effect of earnings differs in the Czech market between loss-making and profit-making
firms, as previously shown in developed markets, and whether or not there is any evidence of
asymmetric timeliness.
The rest of this paper is organised as follows: The paper begins with background information
on the accounting system and capital market of the Czech Republic and then summarises the
prior literature on the impact of losses and the timeliness of earnings. After explaining the
research methodology, a description of the sample and data is provided, together with the
empirical findings and conclusions.
54
2. CZECH ACCOUNTING AND ITS CAPITAL MARKET CONTEXT
Since the early 1990s, there have been many changes in accounting practices in the
transforming economy of the Czech Republic (e.g. Schroll, 1995; Seal et al., 1995; Sucher et
al., 1996, Zarova, 2001). There were similar developments in other East European countries
as they attempted to harmonize and conform to international accounting standards and
practices (e.g. Garrod and McLeay, 1996; Jermakowicz and Rinke, 1996; Craner et al., 2000;
Daniel et al., 2001).
The new framework for financial accounting in the Czech Republic comprises three major
legislative instruments: the Commercial Code, the Act on Accounting and the Charts of
Accounts. The Commercial Code and the Act on Accounting were passed in 1991. Whilst the
Commercial Code provides the general background to the operation of businesses in the
country, the Act on Accounting broadly follows the European Union‟s Fourth Directive (Seal
et al., 1995) and specifies which companies must prepare an audited annual report to be filed
with the Chamber of Commerce. Following on from this legislation, details concerning charts
of accounts and the format for year-end financial statements, and guidance on notes on
consolidated accounts have been issued by the Ministry of Finance.3
In spite of the legal changes that have occurred, Czech accounting is still very closely linked
to taxation rules (Seal et al., 1996). The first tax law was passed in 1992, and further
amendments allowed some discretion but to a very limited extent. There is therefore little
scope for conservatism in financial reporting, given the limitations on provisioning,
depreciation, and deferred taxation. Moreover, enterprises are likely to be penalized if they
overprovision and reduce the tax base (Holeckova, 1996; Zarova, 2001). Hence, Czech
corporations chose to follow tax rules even though it is not absolutely obligatory according to
current Czech accounting legislature. This reflects the overall tax orientation of many Czech
55
companies, rather than orientation to investors that can be seen in other economies.
The Securities Act was passed in 1992, and it relates to issuers of publicly traded securities
and their duty to provide information to market participants. All companies traded on the
Prague Stock Exchange have to publish summary accounting statements in the Business
Bulletin (Sucher and Zelenka, 1998) using internationally accepted accounting principles.
Although reporting on the basis of International Accounting Standards (IAS) is not
compulsory as yet, companies with securities listed on the Prague Stock exchange have to
provide IAS reports to the stock exchange.4
Differences between Czech Accounting
Standards (CAS) and IAS are expected to diminish in the future as more IAS standards are
incorporated into Czech Accounting Standards (Zarova, 2001; Sucher and Alexander, 2001).
The stock market has developed rapidly. After the major political change in 1989, state
property was privatised using a voucher method. This method enabled rapid privatisation of a
large volume of assets despite the lack of domestic capital (Lastovicka et al., 1994). The
Czech capital market was re-established in 1993 following the first wave of voucher
privatisation, and after a 50 year break. At the outset, about one thousand Czech companies
were listed on the market.5
The results of this rapid transformation were questioned later
(Filer and Hanousek, 1999; Desai, 1995), as the privatisation process and the subsequent
listing of shares brought about problems of market rigidity, high agency costs due to the
behaviour of investment funds and protective strategies by commercial banks (Desai, 1995).
Trading on the market suffered from illiquidity and low volume. The decision to list as many
companies as possible took precedence over the usual listing requirements, and the stock
market lacked transparency to begin with (Claessens, 1997; Filer and Hanousek, 1999; and
Hanousek and Nemecek, 2001).6
In 1996, significant administrative and legal changes were introduced into the infrastructure
56
of the Czech capital market. A change of listing rules at the Prague stock exchange required
greater disclosure, and, in an attempt to address some of the problems, illiquid shares and
companies ignoring the reporting requirements were delisted (around 1300 in 1997). This has
had a major impact on the disclosure and reporting behaviour of Czech corporations (Patton
and Zelenka, 1997). Furthermore, the establishment of the independent Securities
Commission in April 1998 was followed by the amendment of the Law on investment funds
in June 1998. One of the most important contributions of the Securities Commission since
then has been to improve transparency in the Czech capital market, with the aim of
maintaining the interest of international investors.7
3. LITERATURE REVIEW
A standard approach in assessing the relation between accounting income and stock return is
to regress a measure of return on accounting income and estimate the earnings response
coefficient, or ERC (see Kothari and Sloan, 1992). The commonly estimated earnings/returns
regression model8
has been expanded by introducing additional factors to explain abnormal
returns (Ou and Penman, 1989), and cross sectional and intertemporal variations of earnings
response coefficients have also been examined (Collins and Kothari, 1989; Donnelly and
Walker, 1995). An extensive account of this empirical research is given in Kothari (2001),
with particular reference to valuation, market efficiency and the political processes
underlying regulation.
3.1 The influence of losses on earnings response coefficients
Previous research on the association of earnings and returns has found that earnings explain
57
only a small fraction of the total variation in returns. To understand this phenomenon, it has
been suggested that earnings contain transitory components that are either value irrelevant or
have only a small valuation impact. As Hayn (1995) shows, this is particularly the case for
losses, which are not expected to perpetuate given that shareholders have a liquidation option,
and the informativeness of losses with respect to future cash flows of the firm is limited. As a
result, if profit and loss observations are pooled in samples when estimating the information
content of earnings, this leads to a downward bias in the estimated earnings response
coefficient and earnings/returns association. In this context, Hayn (1995) shows that reported
losses are less strongly associated with returns than profits and are thus perceived as
temporary by investors. In addition to the extent that there is variation in the incidence of
losses across firms and over time, such variation will affect the cross sectional and
intertemporal variation in the measured earnings response coefficients.9
A further suggestion concerning the low association between earnings and returns is that
different specifications of the earnings variable (levels versus changes, deflation by price
versus earnings, etc.) also have an effect on the measured earnings response coefficient
(Kothari, 1992). More fundamental, however, is the notion that earnings do not reflect the
underlying economic events in a timely manner, and therefore are not synchronised with
stock price movements. This reasoning has given rise to a stream of research on the
timeliness of earnings, which is discussed below.
3.2 Timeliness and conservatism
Asymmetry of earnings timelines was investigated for the first time by Basu (1997), and to a
certain extent it follows from the previous recognition of the different behaviour of losses by
Hayn (1995). Basu changed the perspective and adopted a Beaver et al. (1980) „reverse
58
regression‟ with earnings as the dependent variable. Notably, the OLS standard errors and test
statistics are better specified when the leading variable is specified as independent and the
lagging variable as dependent. In the regression of annual earnings on concurrent annual
returns, the slope coefficient reflects the sensitivity of earnings to contemporaneous
unexpected returns where positive and negative returns are assumed to reflect good and bad
news respectively.10
Basu demonstrates for the US market that the response to good news is
greater, more persistent and less timely than the response to bad news. Good news is less
timely because company accountants require more verifiable information before they
recognise it. The fact that good news is more persistent than bad news is reflected in
subsequent earnings, which materialises as a positive intercept in the earnings/returns
regression. Basu (1997) finds that earnings are 4.5 times more sensitive to negative returns
than to positive returns over the period 1963-1990. Givoly and Hayn (2000) report results that
are consistent with a trend of increased conservatism in the US over this period and into the
1990s, and Raonic, McLeay and Asimakopoulos (2004) show this trend to be present
throughout the European Union as well. Holthausen and Watts (2001) point out that
accounting conservatism existed prior to formal standard settings in the US and increased
after the establishment of FASB in 1973 to the point where virtually all of the association
between earnings and stock prices is driven by bad news.
Ball et al. (2000) have identified international differences, suggesting a high degree of
conservatism in US accounting. They propose that differences in the demand for income
information from one institutional context to another will cause the properties of earnings to
vary internationally. Their results show that accounting income in regulatory regimes with a
shareholder orientation is significantly more timely than in those with a stakeholder
orientation, which is due to the quicker incorporation of economic losses that characterises
conservative accounting. Information asymmetry is resolved in stakeholder regimes through
59
institutional features other than timely and conservative financial statements, notably by
closer relations between management and other major stakeholders.
4. METHODOLOGY
This study considers timeliness and conservatism in the Czech Republic. First, we assess the
effect of losses on the earnings/returns relationship, using two different specifications of the
regression
tttt PXR   1/ [1]
where:
Rt is the return over a 12-month period commencing at the end of fiscal year t-1 including
dividends,
Xt is the earnings variable specified either as earnings per share (EPS) for the year t or the
change in EPS in that year,
t is a normally distributed error term with zero mean and constant variance.
Following Hayn (1995: 134), if losses have a dampening effect on the slope coefficient β that
measures informational content, the following relation between the regression parameters of
the subgroups should hold: ERC (R2
) losses < ERC (R2
) whole sample < ERC (R2
) profits
The reverse regression capturing the relationship between earnings and returns can be
expressed as:
ttt uREY   , [2]
where
EYt is the earnings yield, Xt / Pt-1
60
Rt is market return over period t including dividends,
 is the return response coefficient (RRC), the indicator of timeliness, and
ut is a regression error assumed to be normally distributed with zero mean and constant
variance.
Model [2] describes the relationship between the earnings yield, EYt=Et/Pt-1, and the market
return, Rt=(Pt+Dt-Pt-1)/Pt-1 over the same period. Earnings are less timely if value changes
that are recognized by the market in the present period are not incorporated in the accounting
computations until some time in the future.11
As discussed above, accounting conservatism anticipates losses more quickly than gains, so
that stock prices reflect bad news in the form of contemporaneous market losses earlier than
good news in the form of market gains. By introducing a dummy variable, D, which has a
value of one if return Rt is negative and zero otherwise, the estimation of earnings yield, may
be expressed as:
tttttt uDRRßDEY  2121  . [3]
The coefficient ß1 captures responsiveness of earnings to contemporaneous good news (i.e.
positive market returns) whereas the total of ß1 and ß2 can be interpreted as responsiveness to
bad news (i.e. negative market returns). Conservative accounting implies that ß2 is a positive
coefficient. The relative explanatory power of this regression in periods of bad news and
good news is assessed by the ratio of R2
in the bad news sub-sample to the R2
in the good
news sub-sample. This ratio is expected to be greater than one under conservative reporting12
.
Basically, timeliness and persistence are different ways of viewing the same phenomenon.
Basu summarises this as follows: ‟More timeliness means that more current value relevant
news is recognized contemporaneously in earnings, leaving less current value relevant news
61
to be recognized in future earnings. More persistence means that less current value relevant
news is reported in current earnings, and more of it will be reported in future earnings.‟
(Basu, 1997, 19). Similarly, we expect losses to be less persistent than profits, since the
recognition of bad news is predicted to be more timely than it is in the case of good news. We
hypothesize that the slope coefficient and R2
from a regression of annual earnings on annual
unexpected returns are higher for negative unexpected returns than for positive unexpected
returns.
5. SAMPLE DATA
The sample consists of yearly accounting data and stock prices for 63 industrial companies
listed at the Prague Stock Exchange over the period from 1993 to 1999. The data was
provided by Cekia Ltd., an information agency of the Prague Stock Exchange that is now
independent and assembles data for market research in the Czech Republic. Share prices were
obtained from DataStream.13
Observations are grouped by fiscal year-end, which is identical
to calendar year in the Czech Republic. All accounting variables are measured per share and
deflated by the opening stock price, which controls for heteroscedasticity (Christie, 1987).
The observed data for each firm-year are the current year‟s earnings per share, current and
previous fiscal year-end stock price and current dividend per share. The econometric software
used in this study is E-Views 3.
Table 1 “Earnings and returns in the Czech Republic: descriptive statistics” (see Table 1 at
the end of the paper) provides an overview of the data used in this study. Panels 1 and 2 show
the distribution of earnings yield and earnings changes, and Panel 3 the corresponding
returns. Of the 317 earnings yield observations, 44 are negative, and the average is +3.0%.
Earnings yield is negatively skewed across the whole period, similar to the data in other
62
studies. Earnings changes were measured as the differences in yearly earnings divided by
price at the beginning of the year, and Panel 2 shows that half (157 out of 317) are negative,
with an average of –1.0%. This compares with an average return of +0.7%, as shown in Panel
3, although 175 returns out of 317 (55%) are negative.
Table 2A “Negative earnings yields, corresponding changes and returns” gives a similar
analysis for negative earnings, and table 2B “Positive earnings yields, corresponding changes
and returns” for positive earnings (both tables are at the end of the paper). The average
negative earning yield is –43.1%, with a higher variance than for the sample as a whole.
Earnings changes show a similar pattern to earnings yield when losses are reported, with an
average fall of –32.1% and relatively high variance. Returns corresponding to negative
earnings yields are mainly negative (32 out of 44), and the average is –18.5%.
There are considerably more positive than negative earnings yields, and Table 2B shows that
the mean positive earning yield is +10.4%. The variance of positive earnings yield is much
smaller than that of the negative yields. Average earnings changes are all positive except for
1996, with an average value overall of +4.1%. About one half of the corresponding returns
are negative (143 out of 273), the average return corresponding to a positive earnings yield
being +3.8%.
A summary of the data can be obtained by cross-tabulating positive and negative earnings
yield with positive and negative returns.
The first quadrant (Q1) of Figure 1 “Cross tabulation of earnings and returns” (see Figure 1
at the end of this paper) shows that 41.0 per cent of all observations are positive earnings
contemporaneous with positive return. In Q2, positive earnings and negative return account
for 45.1 per cent of observations. From a naïve viewpoint, the market appears to be more
63
sceptical in its expectations than the accountants are in determining the year‟s earnings. In the
third quadrant, 10.1 per cent are negative returns that are contemporaneous with negative
earnings yield, leaving a frequency of just 3.8 per cent when earnings are negative and returns
are positive.14
6. EMPIRICAL FINDINGS
6.1. The impact of losses
As noted earlier, the first part of this study concentrates on the impact of losses on the
earnings/returns relationship. The model is estimated in levels and in changes of the earnings
variable using yearly data, first for the pooled sample. Subsequently, the sample is divided
into profit-making and loss-making sub-groups and regression [1] is estimated for each subgroup,
again using both the earnings levels and earnings changes specifications. The response
coefficient should be biased downwards for the pooled sample by comparison with the
separate regression of positive earnings. As mentioned earlier, the following relation between
the regression parameters of the subgroups should hold: ERC(R2
) losses < ERC(R2
) whole sample <
ERC(R2
) profits. The results for the Czech data are summarised in Table 3 “The impact of
losses on the earnings/returns relationship” (see Table 3 at the end of this paper).
In levels, the slope coefficient is 0.219 and statistically significant at the 5 per cent level. The
intercept is positive but insignificant and adjR2
is 0.015. A low coefficient of determination
such as this indicates poor specification of this regression model, as often found in the
previous literature on the earnings/returns relationship. The expectation is that the
specification improves after separation of positive and negative earnings.
When the model is re-estimated on the loss-making sub-sample, the slope coefficient is –
0.341 and the intercept coefficient –0.332, both 5 per cent significant. Negative coefficients
64
are not in line with results on other markets. However, it can be seen that this model has the
highest explanatory power, with AdjR2
of 12.1 per cent.15
When the model is again reestimated,
this time on the sub-sample of positive earnings, the slope coefficient is 0.855 and
significant, and AdjR2
is 0.078, which is higher than for the whole sample but lower than for
the loss sub-sample.16
Overall, the results show that the estimation improves as expected after the separation of
profit and losses. This is also the case for earnings changes, where estimates of slope and
intercepts are insignificant in the whole sample and in the case of profits, but significant for
losses. Again, the coefficients have negative values (β=-0.199), which is not in accordance
with previous research in more developed markets, and the coefficient of determination is
still the largest in the case of the loss-making sub-sample.
6.2. Asymmetric timeliness
Two models were estimated, the first being the reversed earnings/returns regression [2] that
evaluates timeliness, and the second the model of asymmetry and conservatism specified in
equation [3] above. The results are summarised in Table 4 “Timeliness and conservatism in
the Czech Republic” at the end of this paper.
Model [2] is estimated for the whole sample, and the slope coefficient (0.084) is positive and
5 percent significant, indicating a degree of timeliness in Czech reporting. In Model [3],
however, there is no direct evidence of conservatism in the sample as a whole as all
coefficients are statistically insignificant.17
A graphical representation of the relationship
between earnings and returns in the „bad news‟ and „good news‟ sub-samples is provided in
Figure 2 “Bad news vs good news on the Czech market” (see Figure 2 at the end of this
paper).
65
The relationship between market returns and accounting earnings that is shown in Figure 2
for the Czech market depicts all observations in the sample, with separate regression lines for
good news and bad news sub-samples (i.e. the last two estimates in Table 4 at the end of this
paper). Neither of the regression models have statistically significant coefficients at
conventional levels. We could thus conclude that there is no different effect of bad news and
good news in the Czech market and there is no evidence of conservatism.
Finally, the sample was partitioned into the positive and negative earnings sub-samples that
were used earlier when evidence was found of the different effect of profits and losses on
earnings response coefficients. We estimated the timeliness [2] and conservatism [3] models
again for these sub-samples, and the results of these tests are summarised in Table 5
“Timeliness and conservatism in profit- and loss-making sub-samples” at the end of this
paper.
The results for the model of timeliness [2] show that both slope coefficients and intercepts are
statistically significant and positive in the positive earnings yield sub-sample (ß1=0.095)
whereas they are negative and significant in the negative earnings yield sub-sample (ß1=-
0.414). The slope coefficient ß1 is significant and positive when earnings yield is positive, but
negative (and less significant) when earnings yield is negative, the latter not being in
accordance with expectations. The results for the model of conservatism [3] show that the
timeliness coefficient ß1 is significant and negative when earnings yield is negative
(ß1=-0.617) but significant and positive when earnings yield is positive (ß1=0.158). However,
the conservatism coefficient ß2 is not significant in either case. Both models show that,
contrary to the situation in more developed markets, good news appears to be more timely
than bad news, and there is no evidence of conservatism on the Czech market.18
66
7. CONCLUSION
This study extends the literature on timeliness and conservatism by further investigation of
the relationship between accounting earnings and returns on the emerging market of the
Czech Republic. We examine the market effect of profit-making and loss-making on returns
and consider the asymmetric timeliness of earnings recognition of bad news and good news.
The results reported in this study are based on a sample of current earnings and
contemporaneous market returns for the period from 1993 to 1999.
Our findings point to differences in the earnings response coefficients of profit making and
loss making firms that are similar to results reported in more developed markets, suggesting
that profits are more persistent than losses in the Czech market as elsewhere. However, we
find no evidence of conservatism in earnings recognition, in the sense that the recognition of
bad news in earnings is not more timely than the recognition of good news, as has been
shown to be the case in more developed markets. Indeed, a more detailed analysis of the subsample
of firm years in which profits were reported led to similar inferences, and the results
therefore are not due to the kind of inconsistent behaviour of negative earnings yields that
would distort the whole picture. In fact, there is some evidence that good news in the Czech
market is recognised more quickly than bad news.
These findings are most likely due to economic transition and regulatory conditions that limit
market influences on accounting behaviour, and therefore the general tendency to
conservatism in more developed markets is not yet apparent. Czech accounting rules have
been relatively inflexible during the period examined, and also firmly linked to fiscal
requirements, and therefore this has not permitted significant discretion. A further reason is
likely to be that the regulatory environment in the Czech Republic is close to the kind of
stakeholder corporatism that is described by Ball et al. (2000), who show that conservatism
67
tends to be less pronounced in such regimes where there are fewer managerial incentives to
bias current earnings.
In conclusion, if changes in market prices signal good news and bad news about future risky
outcomes, there is little evidence of asymmetry in the Czech market in accounting for such
risks. It would be interesting to explore whether accounting conservatism becomes a more
significant feature over time on the Czech market, following the separation of accounting
rules and regulations from tax rules, and the progressive internationalisation of accounting
standards and financial market integration that are now taking place, particularly in Europe.
68
NOTES
1
Conservatism in accounting as described by Basu (1997) is partially explained by the risk adverse
behaviour of corporate accountants. The speed with which bad news is recognized in earnings differs
from the speed of good news recognition. As previously recognized by Hayn (1995), the slope of the
earnings response regression changes with negative earnings which is consistent with the transitory
nature of corporate losses.
2
Timeliness reflects the strength of the link between earnings and returns, as the speedy recognition
of the bad news that is appreciated by the market implies cautiousness on the part of corporate
accountants. Greater conservatism implies risk aversion and risk-responsible accountability.
3
The greatest change in Czech accounting took place in 1991 when the new Act on Accounting came
into effect. Since then there have been several frequent amendments (1994, 1997, 2000, 2001 and
2002) as well as by-laws that address issues of detail, but no further fundamental changes. A new Act
on Accounting comes into effect in 2004. This research does not make ex-post adjustments with
regard to these legislative changes.
4
At this time, the situation is not entirely uniform. Companies that are listed have to prepare IAS
accounts and submit them to the Securities Commission. Other companies are not obliged to do so,
although all companies have been obliged to report according to Czech standards. This does not have
an impact on the dataset used in the present study.
5
Subsequently, about 700 new companies were listed following the second wave of voucher
privatisation in 1995.
6
The efficiency of the Czech capital market has been empirically tested (e.g. Sommer, 1996; Filacek
at al., 1998). Such studies have concentrated on different aspects of market efficiency and are not
congruent. However, the results mostly imply that the market is not efficient, with low turnover and a
lack of transparency. Although this may have an impact on the results of our study, since we use only
end-of-the-year data, low turnover will not affect the results.
7
Administrative changes positively affect data availability, but not the results. Indeed, it is worth
69
noting that such changes did not stem the outflow of international capital investment (Fact Book
Prague Stock Exchange 2002, pp 32-41).
8
The univariate model has the form tttt PXR   1/ , where Rt is the return over a one-year
period, Xt is the earnings per share (EPS) for the year t, t is a normally distributed error term with
zero mean and constant variance.
9
In a previous study on the Czech market (Jindrichovska, 2001), the models did not distinguish
between profits and losses.
10
Note that asymmetry in the timeliness of incorporating economic events in reported earnings may
be measured in different ways, not only the incremental response to bad news relative to good news
but also the relative sensitivity of earnings to bad and good news and the relative explanatory power
of negative and positive return regressions.
11
To improve the explanatory power of this model to explain more of the earnings variance, the
introduction of lagged returns on the RHS of the model would be necessary.
12
The ratio (ß1 + ß2)/ß1 is also considered as a measure of the relative sensitivity of earnings to bad
news compared with their sensitivity to good news (Givoly and Hayn, 2000). This ratio is expected to
be greater than one under conservative reporting, but its magnitude should be interpreted with care
given that zero is the expected value of ß1 when accounts are not timely.
13
This sample represents data on all industrial companies that were available simultaneously in both
databases on the Czech market.
14
During testing, we also excluded these 12 observations from our sample, but the impact was not
material.
15
This result is not in line with other markets. During the period under investigation, negative returns
in the market were far more frequent than negative earnings yields (175 vs. 44) – see Tables 1 and 2.
Therefore, much of the bad news (negative returns) corresponds contemporaneously with reported
profits rather than reported losses. This causes a negative sign on the slope coefficient.
70
16
We control for influential values of the observed variables by weighting out of the regression those
observations with high Cook statistics (>2.25 for this sample). As a result, 5 unduly influential
observations (3 loss makers and 2 profit makers) were omitted from the regression. The ERCs remain
consistent with the full data, increasing from 0.219 (t=2.43) to 0.380 (t=3.46) overall, from -0.341 (t=-
2.63) to -0.334 (t=-1.84), in the case of losses, and from 0.855 (t=4.89) to 0.997 (t=3.21) in the case of
profits, with a change in significance level only in the case of losses, where the degree of confidence
decreases from 95 to 90 per cent, i.e. still 10% significant. Note however that the coefficient of
determination is still the largest in the case of the loss-making sub-sample.
17
Controlling again for observations that have a significant influence on the estimation (see Note 16
above), the timeliness coefficient ß2 in Model [2] of 0.084 (t=2.43) for the full data is relatively
unchanged at 0.081 (t=3.14) when unduly influential observations are weighted out. The estimations
from Model [3] are also robust in this respect, with the conservatism measure ß2 remaining
insignificant: ß2 is 0.045 (t=0.37) for the full data and 0.051 (t=0.53) for the reduced data.
18
We also excluded all negative earnings yield observations from the sample and tested the timeliness
model [2] separately on the sample of positive earnings yield (totalling 173 observations) partitioned
into good and bad news. The estimated slope coefficients are positive and statistically significant in
both cases. However, the coefficient of determination shows that the model is now less well specified.
71
REFERENCES
Ball, R., Kothari, S.P. and Robin A. (2000) ‘The effect of International Institutional Factors
on Properties of Accounting Earnings’, Journal of Accounting and Economics, 29: 1-51.
Basu, S. (1997) ‘The Conservatism Principle and the Asymmetric Timeliness of Earnings’,
Journal of Accounting and Economics, 24: 3-37.
Basu, S. (2001) ‘Discussion of On the Asymmetric Recognition of Good and Bad News in
France, Germany and the United Kingdom’, Journal of Business Finance and Accounting,
28(9&10): 1333-49.
Beaver, W.H., Lambert, R.A. and Morse, D. (1980) ‘The information Content of Security
Prices’, Journal of Accounting and Economics, 2: 3-28.
Claessens, S. (1997) ‘Corporate Governance and Equity Prices: Evidence from the Czech and
Slovak Republics’, The Journal of Finance, 52(4): 1641-58.
Christie, A.A. (1987) ‘On Cross-Sectional Analysis in Accounting Research’, Journal of
Accounting and Economics, 9: 231-58.
Collins, D. W. and Kothari, S. P. (1989) ‘An Analysis of Intertemporal and Cross-sectional
Determinants of Earnings Response Coefficients’, Journal of Accounting and Economics, 11:
143-81.
72
‘Commercial Code and Accounting Legislation in 2001 in the Czech Republic’, (Prague:
Linde).
Craner, J., Krzywda, D., Novotny, J. and Schroeder, M. (2000) ‘The Determination of a
Group for Accounting Purposes in the UK, Poland, and the Czech Republic in a
Supranational Context’, International Journal of Accounting, 35(3): 355-97.
Chun, R.M. (2000) ‘Compensation Vouchers and Equity Markets: Evidence form Hungary’,
Journal of Banking and Finance, 24: 1155-78.
Daniel, P, Suranova, Z. and de Beelde, I. (2001) ‘The development of Accounting in
Slovakia’, The European Accounting Review, 10(2): 343-59.
Desai, R.M. (1995) ‘Financial Market Reform in the Czech Republic 1991-1994: The
Revival of Repression’, Department of Government, Harvard University.
Donnelly, R. and Walker, M. (1995) ‘Share Price Anticipation of Earnings and the Effect of
Earnings Persistence and Firm Size’, Journal of Business Finance and Accounting, 22: 5-18.
Fact Book - Prague Stock Exchange, 2002. on www.pse.cz accessed on 12.12.2003.
Filacek, J., Kapicka, M. and Vosvrda, M. (1998) ‘Efficiency Market Hypothesis: Testing on
the Czech Capital Market’, Finance a uver, 48(9): 554-66 [in Czech].
73
Filer, R. and Hanousek, J. (1999) ‘The Relationship between Economic Factors and Equity
Markets in Central Europe’, Working Paper Series, 119, CERGE-EI, Prague.
Garrod, N., McLeay, S. (1996) Accounting in Transition: the Implication of Political and
Economic Reform in Central Europe, (London: Routledge).
Giner, B. and Rees, W. (2001) ‘On the Asymmetric Recognition of Good and Bad News in
France, Germany and the United Kingdom’, Journal of Business Finance and Accounting,
28(9&10): 1285-331.
Givoly, D. and Hayn, C. (2000) ‘The Changing Time-Series Properties of Earnings, Cash
Flows and Accruals: Has Financial Reporting Become More Conservative?’, Journal of
Accounting and Economics, 29: 287-320.
Hanousek, J. and Nemecek, L (2001) ‘Market Structure, Liquidity, and Information Based
Trading at the Prague Stock Exchange’, Discussion Paper 66, CERGE-EI, Prague.
Hayn, C. (1995) ‘The Information Content of Losses’, Journal of Accounting and Economics,
20: 125-53.
Holeckova, J. (1996) ‘Relationship between accounting and taxation in the Czech Republic’,
European Accounting Review, 5 Supplement: 859-70.
74
Holthausen, R.W. and Watts, R.L. (2001) The Relevance of the Value-Relevance Literature
for Financial Accounting Standard Setting, Journal of Accounting and Economics, 31(1-3):
3-75.
Jermakowicz, E. and Gornik–Tomaszewski, S. (1998) ‘Information Content of Earnings in
the Emerging Capital Market: Evidence from the Warsaw Stock Exchange’, Multinational
Finance Journal, 2(4): 245–67.
Jermakowicz, E. and Rinke, D. (1996) ‘The new accounting Standards in the Czech Republic,
Hungary and Poland vis-à-vis International Accounting Standards and European Union
Directives’, Journal of International Accounting Auditing and Taxation, 5(1): 73-88.
Jindrichovska, I. (2001) ‘The Relationship between Accounting Numbers and Returns: Some
Empirical Evidence from the Emerging Market of the Czech Republic’, The European
Accounting Review, 10(1): 107-31.
Kothari, S. P. (1992) 'Price-earnings Regressions in the Presence of Prices Leading Earnings',
Journal of Accounting and Economics, 15: 173-202.
Kothari, S.P. (2001) ‘Capital Markets Research in Accounting’, Journal of Accounting and
Economics, 31: 105-231.
Kothari, S. P. and Sloan, R. G. (1992) ‘Information in Prices about Future Earnings’, Journal
of Accounting and Economics, 15: 143-71.
75
Lastovicka, R, Marcincin, A. and Mejstrik, M (1994) ‘Privatisation and Opening the Capital
Markets in the Czech and Slovak Republics’, Working Paper Series 54, CERGE-EI, Prague.
Ou, J. A. and Penman, S. H. (1989) ‘Financial Statement Analysis and the Prediction of
Stock Returns’, Journal of Accounting and Economics, 11: 295-329.
Patton, J. and Zelenka, I. (1997) ‘An Empirical Analysis of the Determinants of Disclosure in
Annual Reports of Joint Stock Companies in the Czech Republic’, European Accounting
Review, 6(4): 605-26.
Pope, P. and Walker, M. (1999) ‘International Differences in the Timeliness, Conservatism
and Classification of Earnings’, Journal of Accounting Research, 37 Supplement: 53-100.
Raonic, I., McLeay, S. and Asimakopoulos, I. (2004) ‘The Timeliness of Income Recognition
by European Companies: An Analysis of Institutional and Market Complexity’, Journal of
Business Finance and Accounting, 31(1-2): pp.115-48.
Schroll, R. (1995) ‘The New Accounting System in the Czech Republic’, The European
Accounting Review, 4(4): 827-32.
Seal, W., Sucher, P. and Zelenka, I. (1995) ‘The Changing Organization of Czech
Accounting’, The European Accounting Review, 4(4): 659-81.
76
Seal, W., Sucher, P. and Zelenka, I. (1996) ‘Post-Socialist Transition and the Development of
Accounting Profession in the Czech Republic’, Critical Perspectives on Accounting, 7: 485-
508.
Sommer, M. (1996) ‘Market Model on the Czech Equity Market’, Prague Economic Papers,
3: 12-48.
Sucher, P. and Alexander, D. (2001) 'International Accounting Standards, Why, How and
Where, a Case Study', working paper presented at the BAA SIG conference in London, 2001.
Sucher, P. and Jindrichovska, I. ‘Implementing IFRS: A Case Study of the Czech Republic’,
Accounting in Europe, 1: 109-41.
Sucher, P., Seal, W. and Zelenka, I. (1996) ‘True and Fair in the Czech Republic: A Note on
Local Perceptions’, The European Accounting Review, 5(3): 545-57.
Sucher, P. and Zelenka, I. (1998) ‘The Development of the Role of the Audit in the Czech
Republic’, The European Accounting Review, 7(4): 723-51.
Zarova, M. (2001) ‘Emerging Issues in the Czech Accounting System and Proposals to
Solutions’, conference paper presented at the 24th
European Accounting Congress 2001.
77
Figure 1 Frequencies of observations in quadrants
Q2 EYt Q1
Total
143 130 273
45.11% 41.01%
Rt
32 12 44
10.09% 3.79%
Q3 Total 175 142 Q4
78
Figure 2 Bad news versus good news on the Czech market
Bad news vs Good news
-1.0
-0.8
-0.6
-0.4
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
-1.0 -0.5 0.0 0.5 1.0
Return (Rt)
EarningsYield(EYt)
Regression equation for good news is EYt = 0.031 + 0.076Rt.
Regression equation for bad news is EYt = 0.044 + 0.121Rt.
79
Table 1 Earnings and returns in the Czech Republic: descriptive statistics
Descriptive statistics for earnings yield (Et /Pt-1), earnings change ((Et –Et-1)/Pt-1) and returns
((Pt-Pt-1DtPt-1). Variables have been measured over a one-year period using data over the
period 1993-1999. Sample sizes vary from 33 to 63 firm-years. a,b,c
Panel 1
Earnings yield: EYt=Et/Pt-1
Negative % of neg.
Year MEAN STDEV MEDIAN MIN MAX COUNT earnings earnings
1994 0.086 0.081 0.058 -0.065 0.330 33 1 3
1995 0.031 0.235 0.059 -1.232 0.315 38 4 11
1996 0.022 0.203 0.051 -1.156 0.396 63 9 14
1997 0.016 0.180 0.034 -0.659 0.757 62 10 16
1998 0.057 0.191 0.046 -0.739 0.662 61 7 11
1999 -0.007 0.608 0.093 -3.018 1.858 60 13 22
1993-9 0.030 0.314 0.052 -3.018 1.858 317 44 14
Panel 2
Earnings change measure: DEYt=(Et-Et-1)/Pt-1 Negative % of neg.
change change
Year MEAN STDEV MEDIAN MIN MAX COUNT measures measures
1994 0.016 0.096 -0.000 -0.155 0.392 33 17 52
1995 -0.032 0.190 0.004 -1.036 0.226 38 15 39
1996 -0.043 0.209 -0.017 -1.248 0.591 63 45 71
1997 0.017 0.198 0.001 -0.372 0.848 62 30 48
1998 0.050 0.228 0.005 -0.411 1.051 61 24 39
1999 -0.063 0.633 0.008 -3.122 1.682 60 26 43
1993-9 -0.010 0.328 0.000 -3.122 1.682 317 157 50
Panel 3
Return: Rt=(Pt-Pt-1+Dt)/Pt-1
Bad % of bad
Year MEAN STDEV MEDIAN MIN MAX COUNT news news
1994 0.226 0.583 0.225 -0.798 1.388 33 11 33
1995 -0.120 0.366 -0.176 -0.658 0.725 38 25 66
1996 0.247 0.590 0.251 -1.000 1.584 63 22 35
1997 -0.136 0.331 -0.108 -0.712 1.519 62 45 73
1998 -0.249 0.390 -0.349 -0.828 1.126 61 47 77
1999 0.125 0.522 0.113 -1.000 2.077 60 25 42
1993-9 0.007 0.507 -0.057 -1.000 2.077 317 175 55
a
Earnings yield (Et /Pt-1 is annual earnings per share for year t divided by the price at the
beginning of that year.
b
Earnings change((Et-Et-1)/Pt-1 is the difference in earnings per share between years t and t-1
divided by share price at the beginning of year t.
c
Return ((Pt-Pt-1DtPt-1) is return over year t measured as difference in prices at the end and
beginning of year t including dividends paid in year t divided by price at the beginning of
year t.
All earnings per share numbers, dividends and prices are adjusted for stock splits.
80
Table 2A Negative earnings and corresponding earnings changes and returns: descriptive
statistics
Descriptive statistics for earnings yield (Et /Pt-1), earnings change ((Et –Et-1)/Pt-1) and returns
((Pt-Pt-1DtPt-1). Variables have been measured over a one- year period using data over the
period 1993-1999. a,b,c
Panel 1
(Negative) Earnings yield: EYt=Et/Pt-1
Year MEAN STDEV MEDIAN MIN MAX COUNT
1994 -0.065 NA -0.065 -0.065 -0.065 1
1995 -0.425 0.561 -0.225 -1.232 -0.016 4
1996 -0.322 0.354 -0.174 -1.156 -0.007 9
1997 -0.261 0.226 -0.247 -0.659 -0.011 10
1998 -0.299 0.280 -0.213 -0.739 -0.056 7
1999 -0.739 0.806 -0.389 -3.018 -0.065 13
1993-9 -0.431 0.541 -0.272 -3.018 -0.007 44
Panel 2
Corresponding earnings change measures: DEYt=(Et-Et-1)/Pt-1
Year MEAN STDEV MEDIAN MIN MAX COUNT
1994 -0.060 N/A -0.060 -0.060 -0.060 1
1995 -0.402 0.460 -0.264 -1.036 -0.042 4
1996 -0.225 0.499 -0.233 -1.248 0.591 9
1997 -0.128 0.161 -0.142 -0.372 0.152 10
1998 -0.103 0.152 -0.049 -0.352 0.105 7
1999 -0.648 0.954 -0.403 -3.122 0.328 13
1993-9 -0.321 0.614 -0.191 -3.122 0.591 44
Panel 3
Corresponding return: Rt=(Pt-Pt-1+Dt)/Pt-1 No of bad
Year MEAN STDEV MEDIAN MIN MAX COUNT news
1994 -0.669 N/A -0.669 -0.669 -0.669 1 1
1995 -0.227 0.411 -0.211 -0.658 0.174 4 2
1996 -0.160 0.419 -0.247 -0.743 0.605 9 7
1997 -0.219 0.639 -0.383 -0.655 1.519 10 9
1998 -0.544 0.146 -0.605 -0.737 -0.376 7 7
1999 0.068 0.469 0.115 -0.597 0.838 13 6
1993-9 -0.185 0.491 -0.316 -0.743 1.519 44 32
a
Earnings yield (Et /Pt-1 is annual earnings per share for year t divided by the price at the
beginning of that year.
b
Earnings change ((Et-Et-1)/Pt-1 is the difference in earnings per share between years t and t-1
divided by share price at the beginning of year t.
c
Return ((Pt-Pt-1DtPt-1) is return over year t measured as difference in prices at the end and
beginning of year t including dividends paid in year t divided by price at the beginning of
year t.
All earnings per share numbers, dividends and prices are adjusted for stock splits.
81
Table 2B Positive earnings and corresponding earnings changes and returns: descriptive
statistics
Descriptive statistics for earnings yield (Et /Pt-1), earnings change ((Et –Et-1)/Pt-1) and returns
((Pt-Pt-1DtPt-1). Variables have been measured over a one-year period using data over the
period 1993-1999. a,b,c
Panel 1
(Positive) Earnings yield: EYt=Et/Pt-1
Year MEAN STDEV MEDIAN MIN MAX COUNT
1994 0.090 0.078 0.074 0.001 0.330 32
1995 0.085 0.071 0.062 0.003 0.315 34
1996 0.080 0.077 0.059 0.001 0.396 54
1997 0.070 0.109 0.043 0.001 0.757 52
1998 0.103 0.117 0.061 0.005 0.662 54
1999 0.195 0.334 0.115 0.001 1.858 47
1993-9 0.104 0.168 0.064 0.001 1.858 273
Panel 2
Corresponding earnings change measures: DEYt=(Et-Et-1)/Pt-1
Year MEAN STDEV MEDIAN MIN MAX COUNT
1994 0.018 0.097 0.001 -0.155 0.392 32
1995 0.012 0.051 0.008 -0.093 0.226 34
1996 -0.013 0.082 -0.015 -0.237 0.338 54
1997 0.045 0.193 0.003 -0.221 0.848 52
1998 0.070 0.230 0.010 -0.411 1.051 54
1999 0.099 0.391 0.015 -0.658 1.682 47
1993-9 0.041 0.217 0.003 -0.658 1.682 273
Panel 3
Corresponding return: Rt=(Pt-Pt-1+Dt)/Pt-1
no of bad
Year MEAN STDEV MEDIAN MIN MAX COUNT news
1994 0.254 0.569 0.268 -0.798 1.388 32 10
1995 -0.108 0.365 -0.176 -0.636 0.725 34 23
1996 0.315 0.590 0.318 -1.000 1.584 54 15
1997 -0.120 0.239 -0.085 -0.712 0.405 52 36
1998 -0.211 0.396 -0.238 -0.828 1.126 54 40
1999 0.141 0.539 0.110 -1.000 2.077 47 19
1993-9 0.038 0.504 -0.036 -1.000 2.077 273 143
a
Earnings yield (Et /Pt-1 is annual earnings per share for year t divided by the price at the
beginning of that year.
b
Earnings change ((Et-Et-1)/Pt-1 is the difference in earnings per share between years t and t-1
divided by share price at the beginning of year t.
c
Return ((Pt-Pt-1DtPt-1) is return over year t measured as difference in prices at the end and
beginning of year t including dividends paid in year t divided by price at the beginning of
year t.
All earnings per share numbers, dividends and prices are adjusted for stock splits.
82
Table 3 Impact of losses on earnings–return relationship
Estimated parameters of a pooled regression of annual returns on earnings yield for the same
period deflated by price. Sample sizes vary from 44 to 317 firm-years. Variables have been
measured over a period of one-year using data for 63 companies over the period 1993-1999.
The estimated regression model is tttt PXR   1/ a
Specification Levels Changes
Sample  ß AdjR
2
 ß adjR
2
N
Whole sample 0.001 0.219 0.015 0.008 0.049 -0.002 317
t-statistic (0.024) (2.428)* (0.270) (0.561)
Loss cases -0.332 -0.341 0.121 -0.249 -0.199 0.040 44
t-statistic (-3.724)* (-2.628)* (-3.031)* (-1.665)
+
Profit cases -0.051 0.855 0.078 0.033 0.141 0.000 273
t-statistic (-1.467) (4.887)* (1.055) (1.002)
t-statistics are in parenthesis: *) the coefficient is statistically significant at the 5 per cent
level; +
) statistically significant at the 10 per cent level.
a
In the model, Rt is the annual return of fiscal year t, Xt is the earnings per share variable in
year t (specified either in levels or changes), Pt-1 is the share price at the end of year t-1, t is
the error term,  is the intercept, and  is the earnings response coefficient (ERC).
83
Table 4 Timeliness and conservatism: good news versus bad news
Estimated parameters of a pooled regression of one-year earnings yield on annual returns of
the same period. Sample sizes vary from 142 to 317 firm-years. Variables have been
measured over the period of one-year using data for 63 companies over the period 1993-
1999.
The timeliness model has the following specification: ttt uREY   .a
The conservatism model is specified as follows: tttttt uDRRßDEY  2121  .b
α1 α 2 ß1 ß2 AdjR
2
R
2
N
Timeliness - whole sample
coefficient 0.029 0.084 0.015 0.018 317
t-statistic (1.670)
+
(2.428)*
Conservatism - whole sample
coefficient 0.031 0.013 0.076 0.045 0.090 0.019 317
t-statistic (0.780) (0.219) (1.141) (0.370)
Timeliness - good news only
coefficient 0.031 0.076 -0.001 0.006 142
t-statistic (0.624) (0.915)
Timeliness - bad news only
coefficient 0.044 0.121 0.009 0.015 175
t-statistic (1.384) (1.608)
t-statistics are in parenthesis: *) the coefficient is statistically significant at the 5 per cent
level; +
) statistically significant at the 10 per cent level.
a
The variables in the timeliness model are defined as follows: EYt is earnings yield for year t
(specified either in levels or changes), Rt is the annual return of fiscal year t,  is the intercept
and  is the return response coefficient measuring timeliness, and ut is the error term.
b
The conservatism model has the following variables:is specified as follows: EYt is earnings
yield for year t (specified either in levels or changes), Rt is the annual return of fiscal year t,
Dt is dummy variable which has value of 1 in case of negative returns and zero otherwise, and
ut is the error term. 1 is the intercept 2 is the marginal change in the intercept of negative
return cases. 1 is the return response coefficient measuring timeliness of good news earnings
and 2 is the marginal change of bad news earnings timeliness measuring conservatism.
84
Table 5 Timeliness and conservatism in profit and loss making sub-samples
Estimated parameters of a pooled regression of one-year earnings yield on annual returns of
the same period. Sample sizes vary from 44 for loss making firms to 273 for profit making
observations. Variables have been measured over the period of one-year using data for 63
companies over the period 1993-1999.
The timeliness model has the following specification: ttt uREY   .a
The conservatism model is specified as follows: tttttt uDRRßDEY  2121  .b
specification α1 α2 ß1 ß2 adjR
2
R
2
N
Timeliness - profit making firms
coefficient 0.101 0.095 0.078 0.081 273
t-statistic (10.281)* (4.887)*
Timeliness - loss making firms
coefficient -0.508 -0.414 0.121 0.141 44
t-statistic (-6.208)* (-2.628)*
Conservatism - profit making firms
coefficient 0.062 0.037 0.158 -0.094 0.086 0.096 273
t-statistic (2.939)* (1.179) (4.403)* (-1.406)
Conservatism - loss making firms
coefficient -0.373 -0.154 -0.617 0.190 0.086 0.149 44
t-statistic (-1.606)
+
(-0.474) (-1.658)* (0.313)
Timeliness - profits on good news sample
Coefficient 0.062 0.158 0.070 0.077 130
t-statistic (2.181)* (3.264)*
Timeliness - profits on bad news sample
Coefficient 0.099 0.064 0.027 0.034 143
t-statistic (8.529)* (2.226)*
t-statistics are in parenthesis: *) the coefficient is statistically significant at the 5 per cent
level; +
) statistically significant at the 10 per cent level.
a
In the timeliness model the variables are as follows: EYt is earnings yield for year t
(specified either in levels or changes), Rt is the annual return of fiscal year t,  is the intercept
and  is the return response coefficient measuring timeliness, and ut is the error term.
b
In the conservatism model the variables are as follows: EYt is earnings yield for year t
(specified either in levels or changes), Rt is the annual return of fiscal year t, Dt is dummy
variable which has value of 1 in case of negative returns and zero otherwise, and ut is the
error term. 1 is the intercept 2 is the marginal change in the intercept of negative return
cases. 1 is the return response coefficient measuring timeliness of good news earnings and 2
is the marginal change of bad news earnings timeliness measuring conservatism.
85
Část 4
Implementing IFRS: A Case Study of the Czech Republic
[Zavádění Mezinárodních standardů finančního výkaznictví:
Případová studie České republiky]
Patricia Sucher & Irena Jindrichovska
Accounting in Europe, 1. 2004, 109-141.
86
ABSTRACT
IMPLEMENTING IFRS: A CASE STUDY OF THE CZECH REPUBLIC
This empirical paper presents a study of the implementation process for international financial
reporting standards (IFRS) in one of the accession countries, the Czech Republic. Based upon a
review of the legislation, institutional framework and context, and drawing upon recent interviews
with Czech companies required to prepare IFRS accounts, auditors and institutional players in the
Czech Republic, the paper highlights some of the key issues that are arising with the move to the
implementation of IFRS for listed group companies and other enterprises in the Czech Republic.
The paper considers the issues that arise when implementing new accounting regulations, some of
which are not new and have been well covered in the literature, but others of which are particular to
the implementation of IFRS. The method of implementation, the scope of IFRS, particular issues
with local accounting practice and IFRS, the issue of enforcement of compliance with IFRS and its
relationship with audit, the link between IFRS and taxation and the provision of education and
training are all considered. There is also a review of the state of preparedness of local group listed
entities with respect to the implementation of IFRS.
There are many potentially rich areas for accounting research where the work could also inform the
practice of IFRS accounting. The paper provides a contribution by highlighting how one country has
moved to implement the requirement for group listed enterprises to prepare IFRS accounts and the
issues that then arise for legislators, preparers and users.
Pat Sucher Irena Jindrichovska
Royal Holloway – University of London Charles University, Prague
Egham, Surrey, TW20 0EX, UK Opletalova 26
P.sucher@rhul.ac.uk 110 00 Prague
Czech Republic
87
Introduction
In May 2004, ten new countries (the accession countries)1
will join the European Union (EU). As
they join the EU they will accede to many of the EU regulations, which could and will involve many
dramatic changes to their national laws, local regulation and, eventually, local practices. One of the
major changes for accession countries will involve the need for group-listed companies to prepare
financial statements according to international financial reporting standards (IFRS). Over the last
fourteen years, the transitional economies that comprise eight of the accession countries have had to
implement a multitude of changes to their legislation, which have been well documented (Bailey,
1995; Bailey and Alexander, 2001; Schroeder, 1999; and Sucher and Zelenka, 1998 writing on
transitional economies). In many ways this latest change is but one more which has constituted a
revolution in financial reporting for many transitional economies. However, unlike the changes in
national legislation towards the 4th
and 7th
Directives, the moves by the transitional countries are
happening at the same time as many other countries make a similar change – not least those already
within the EU. Therefore any research on what has happened/is happening in transitional economies
could also illuminate the issues that are likely to arise in other economies as they move to implement
the June 2000 EU communication, The EU’s Financial Reporting Strategy: The Way Forward (EU
2000).
In this paper, following a review of some of the relevant literature on international accounting
harmonisation and how appropriate IFRS might be for transitional economies, the key general and
specific themes and issues that might arise when implementing IFRS within a transitional economy
are considered. Then, using these themes and issues as a structure for analysis, the authors analyse
the legislative and institutional approach to implementing IFRS within the Czech Republic. This
analysis is supported by interview work with prospective preparers, auditors and other institutional
representatives. The authors highlight the current situation in the Czech Republic concerning the
enabling legislation for companies to prepare IFRS and the state of preparedness amongst the relevant
enterprises. In the fourth and final section, the problems and prospects of the move to implement
IFRS for group-listed enterprises in the Czech Republic are delineated.
Literature review
There are various strands in the literature on international accounting harmonisation which cover
such areas as, the degrees of harmonisation (Tay and Parker, 1990; Emenyonu and Gray, 1996;
Murphy, 2000), whether international accounting harmonisation is appropriate (Hove, 1986; Cairns,
1997: Flowers, 1997) and the degree of compliance with IFRS2
(Cairns, 2001; Street and Gray, 2001).
Of most relevance to this paper on the implementation of IFRS would seem to be the research on the
88
appropriateness of IFRS for developing or transitional economies, the practical implications of
implementing IFRS within a country located in the continental tradition of accounting and research on
the changes to national accounting in transitional economies.
In his 1998 study, based on a review of the literature on accounting and developing countries, Nobes
noted various points about the appropriateness of IAS for developing countries. He suggested that
given the likely users of most enterprise financial statements in developing countries were the tax
authorities and owners and lenders to private enterprises, adopting IAS in developing countries, where
IAS are stated to be investor oriented (IASB, 2003; F9), may not be appropriate. In particular many
of the later IAS (IAS 22-38) might not be relevant for presenting accounting information for taxation
purposes. Nobes also suggested that, given the lack of a developed accountancy profession to
interpret and apply the more judgemental aspects of IFRS, (either as auditors or accountants) there
could also be issues of the reliability of accounting information. Though he provides examples, he
does not make explicit that this could be a particular issue for any accounting information based to a
large extent on market data such as fair value (Nobes, 1998a). It is unlikely that the Czech Republic
would be classified as a developing economy of any of the criteria suggested by Nobes (e.g. number
of public companies; number of public companies deflated by (put into context of) the size of the
economy), but some of the issues Nobes raises may still be relevant given the short time in which an
accounting profession has developed. Further research on transitional economies has indicated
structural issues with a lack of active stock markets, and corruption, which may affect the reliability
of IFRS financial statements and estimations of fair values (EBRD, 2000; Sucher and Bychkova,
2001).
The historical differences in accounting thought, context, ethos and practice between a number of
countries that may affect the de facto preparation of IFRS financial statements have been analysed in
the work of Nobes (1983). Nobes distinguishes two broad traditions, the European and the AngloSaxon.
The difficulties of moving from one to the other, or seeking some harmonisation among the
different traditions, are well illustrated by the relative failure of the European 4th
Directive to achieve
great similarity of thought and practice. In particular, the concept of „true and fair‟ which may be
identified with the Anglo-Saxon tradition 3
has been applied as an override or as a complimentary part
of the accounting depending on the accounting tradition within the country. See for example
Alexander (1993, 1996) and Ordelheide (1990, 1993, 1996), which demonstrate how local traditions
exercise a strong influence over the implementation of true and fair.
There has been surprisingly little academic empirical research on what has happened when IFRS are
implemented within a country, though there are recent World Bank reports on the subject (ROSC,
89
20014
;ROSC, 2002a, ROSC, 2002b; ROSC, 2002c; ROSC, 2002d; ROSC, 2003a; ROSC, 2003b).
The objectives of the World Bank reports
„are to assess the comparability of national accounting and auditing standards with International
Accounting Standards (IAS) and International Standards on Auditing (ISA), respectively; and the degree
to which corporate entities comply with established accounting and auditing standards in the country‟.5
In the reports it is suggested that there may be issues regarding the availability of current translations
of the IFRS which affect how able enterprises are to prepare IFRS financial statements (ROSC, 2002d
and ROSC, 2003b). In Bulgaria, Croatia and Romania, the requirement to prepare IFRS financial
statements has been extended beyond listed enterprises to all other enterprises. The World Bank
queries whether it is appropriate for small and medium sized non-listed enterprises to prepare IFRS
financial statements as it may be an „undue burden‟ for such enterprises (ROSC, 2002a).
The World Bank studies have also highlighted the lack of mechanisms for enforcement of compliance
either with national accounting standards (ROSC, 2002d) or IFRS, where they have already been
implemented (ROSC, 2002a). This is reflected in the World Bank reviews of samples of published
financial statements where they note non-compliance in various areas. The same studies also
comment on the need for substantial education and training in IFRS to prepare for the move to IFRS
in the countries studied (e.g. in ROSC, 2003a; ROSC, 2003b). These observations are supported in the
research of Sucher and Alexander, (2002).
On the basis of a review of the IFRS accounts and interviews in Russia and the Czech Republic with
large companies implementing IFRS and their auditors, Sucher and Alexander (2002) suggested that
there were particular issues that arose when companies implemented IFRS. The companies they
researched were preparing IFRS financial statements voluntarily. The impetus for this was usually the
requirement to obtain funding from overseas investors. In their study, the authors highlighted the lack
of accountants trained and fully familiar with IFRS and the degrees of non-compliance with IFRS,
particularly in the area of group accounts and substance over form. There was also some evidence that
in cases where companies were preparing IFRS accounts, the auditors were substantially involved in
preparing the accounts to such an extent that there might be issues concerning the infringement of
auditor independence (Sucher and Alexander, 2002). Concerns were also raised about the complexity
of certain IFRS and the difficulties of estimating and applying fair values in company accounts
(Sucher and Alexander, 2002).
Other research has studied the way in which national accounting systems have changed towards IFRS.
Much of this has been descriptive and has recounted the degree to which national accounting
regulations have approximated to IFRS (an example is Jermakowicz and Rinke, 1996 or the IFAD6
90
reports of 2000 and 2001). This research has not indicated the nature of any issues that might arise in
implementing IFRS nationally or for individual enterprises, though one might suppose that in any
areas where there are substantial differences between local financial reporting and IFRS, there may be
more problems in preparing IFRS financial statements as the information may not be available e.g.
segmental reporting in IAS 14.7
The changes in financial reporting in the transitional economies of Central Europe since 1989 have
been the focus of a variety of research. It has been suggested that, even with the changes towards
substance over form and a focus on investors in financial reporting, the emphasis on compiling proper
accounting records and on adhering to tax regulations rather than fairly presenting financial
statements has continued, despite contrary requirements in the legislation (Bailey, 1995; Sucher, et
al., 1996). Large differences between local financial reporting practices as they have developed since
1991, with an orientation to tax reporting and form, and the IFRS emphasis on investor oriented
financial reporting may also affect the degree of compliance with IFRS in locally prepared IFRS
financial statements.
Other writers have used organisational change theory to reflect on the levels of change in financial
reporting and establish the coherence and depth of the change (Seal et al, 1995). Their suggestions
were that the changes in financial reporting were generally coherent, and had achieved a „second
order‟ depth through „colonisation.‟ With colonisation, „a small group “ on the back of an
environmental disturbance, create lasting and fundamental change in both the visible and invisible
elements in an organisation.”‟ (Seal et al, 1995: 662). However, much change has been state
sponsored, rather than spontaneous, and the Ministry of Finance has continued to play a key role (Seal
et al., 1995). This may therefore indicate that change may be very slow.
Garrod and McLeay (1996) also highlighted how governments have adapted the previous state
controlled accounting systems to meet the envisaged demands of new users as well as state demands
for tax collection and dealt with the general issues of conflicting objectives for financial reporting.
They comment on the problems of establishing the right balance between state and professional
involvement in the regulation of accounting and the difficulties that have arisen in valuation of assets,
in the absence of fully functioning markets, and the scope of financial statements (Garrod and
McLeay, 1996).
In much of the previous research there has been an emphasis on the de jure situation with accounting
change in Central Europe, and less focus on the de facto situation which might be quite different (as
suggested by Bailey, 1995). There is a need for further research on the actual operation of financial
91
reporting within enterprises to ensure that any implementation process pays due regard to the de facto
situation rather than merely the de jure situation.
There has also been a tendency in the academic research to study individual countries as comprising
enterprises all with similar financial reporting objectives. However, within a country there are
different groups of enterprises with different reporting objectives depending on their size, funding and
ownership structure. Indeed, some authors have suggested that, within the classification literature,
there may be different „types‟ of accounting systems in operation in any one country (Nobes, 1998b).
This could be taken further to suggest that there are different groups of professional auditors and
accountants trained under different reporting systems (e.g. some auditors trained under an
international system of exams in accounting; some under a local examination system). Any research
on the changes to national accounting should reflect this diversity of needs and resources.
The Big Four audit firms have played a large role in the move to IFRS around the world, as they are
often the auditors of IFRS statements and they have the large resources needed to maintain
appropriate knowledge and expertise in IFRS. There has been relatively little research on their role in
transitional economies (though Seal, et al., (1995), and Sucher and Alexander, (2002) are exceptions
and some concerns about their roles are raised in some of the World Bank Reports). Some
consideration needs to be given to their role in the move to IFRS within the transitional economies.
Following from the review above, we focus our study of the implementation of IFRS in the Czech
Republic as follows. The literature suggests that there may be contextual factors which affect the
appropriateness and effectiveness of the implementation of IFRS in a transitional economy such as the
Czech Republic. These factors may be divided into the position of the Ministry of Finance in
initiating and implementing reform; the role of the particular users of financial statements, and the
relationship between financial reporting and taxation which in turn may relate to the national tradition
in accounting (Anglo Saxon vs European). Then it is suggested that there are specific factors which
may affect the effectiveness of and the ability to implement IFRS: differences between local and
IFRS accounting standards, particular issues with individual IFRS and the reliability and availability
of market data for fair values, and the approach to enforcing compliance, with particular regard to the
role of auditors. There may also be additional factors concerning how the IFRS are translated and the
extent of education and training in IFRS. Obviously some of the specific factors are interrelated with
the contextual factors, but they are separated here for ease of discussion. Practically, it would also be
useful to ascertain how prepared group-listed enterprises in the Czech Republic are for the transition
to IFRS.
92
In the next section we outline how we have approached our analysis of the situation in the Czech
Republic.
Approach to analysis
Having identified the general and specific issues that may arise with the transition to IFRS in a
transitional economy, we used this as a structure to inform our review of the legislative and
institutional developments in the Czech Republic. As we were also keen to identify how enterprises,
auditors and other professionals were interpreting these developments and experiencing the changes
in accounting, we undertook interviews in September and December, 2003, to try and identify what
was happening as well as to support out analysis of the legislation. Tables of those interviewed are
noted below:
Table one: Companies interviewed
Company Activity Individual(s)
A Import/Export Director
B Property rental Financial Director
C Wine production Economic director
D Hotels Economic director
and Auditor
E Construction Financial controller
F Bakery Economic director
G Printer Economic director
H Construction Financial controller
I Manufacturer and
distributor
Financial controller
Table two: Other interviewees
Organisation Individuals
International accounting firm (Big Four) Manager J
International accounting firm (outside Big
Four)
Manager K
Union of Accountants President L
Union of Accountants Head of Accounting
methodology
M
Previous research on the use of IFRS in the Czech Republic has focused on the state of IFRS
implementation with the „blue chip‟ companies (Sucher and Alexander, 2002). In the company
interviews for this research we sought to obtain information about the state of preparedness and
general issues that might arise with the smaller listed enterprises that would have to prepare IFRS
financial statements after 2005. Therefore the RM system population of companies was used as a
basis for selecting a suitable sample of companies for interview.8
The companies were identified from
93
the RM system on two different dates, 4th
August 2003 and 29th
September 2003.9
A sample of
companies was selected that met the following criteria:
 They were not a financial institution;
 They were not majority owned by overseas investors; and
 Their registered office was in Prague10
This led to a sample of 19 companies that covered a diverse range of activities from hotels, to
construction to the manufacture of food. From this sample of 19 companies, four companies stated
that they were not doing anything about IFRS, there was nothing to discuss and they would await the
regulation from the Ministry of Finance. Six companies declined to be interviewed for a variety of
other reasons. For example, one company was facing an imminent privatisation, another had de-listed
from the RM system and another was undergoing an audit. This left nine companies who were willing
to be interviewed in either early September or December 2003. Given the small sample, though it was
a cross-section of Czech companies, we cannot infer that they are representative of all Czech
companies. However the interviewee comments do provide indications of some of the issues with the
transition to IFRS in the Czech Republic.
The interviews were semi-structured in format, with open-ended questions to prompt discussion. A
note of the questions used in the company interviews in is in appendix one. There were two
interviewers present at each interview, one of whom was a native Czech speaker and fluent in
English. The majority of the interviews were conducted in Czech with English interpretation where
necessary. Detailed notes were taken during the interview by both interviewers. These notes were
written up very soon after each interview. In one company, the auditor, a sole practitioner, was
interviewed with the financial director, as they would be working closely together on the move to
IFRS. After initial introductions, the auditor discussed most of the issues, as they were better
informed about IFRS than the director.
We now outline the role of the stock exchange and the situation with overall funding for enterprises,
the development of the institutional framework for financial reporting and the process that has been
followed in developing the necessary legislation for implementing IFRS. We then consider the
relationship between financial reporting, taxation and the tax inspectorate, and other users of
published financial reports. Subsequently, we discuss the role of the true and fair requirement as an
indication of the influence of the Anglo-Saxon versus European tradition approach to financial
reporting in the Czech Republic, and its implications for IFRS. We then study particular issues with
IFRS and how enforcement is likely to operate. We also comment on the local professions and
education and training. Finally we discuss the state of preparedness for the implementation of IFRS
94
amongst the smaller listed group enterprises. Where appropriate, we include comments from the
interviewees.
The Stock Exchange and overall Enterprise funding
With the Velvet revolution of 1989, the Czech Republic (then Czechoslovakia) commenced the
process of transition from a command economy to a market economy. Though the initial impetus for
foreign direct investment slowed down as investors became concerned about the security,
accountability and transparency of their investments after 1995, by the end of 2002 the Czech
Republic had the highest amount of foreign direct investment per capita in European transitional
economies (EBRD, 2003:65).
Since 2000, companies listed on the main market of the Prague Stock Exchange (PSE) have had to
produce consolidated financial statements in accordance with IFRS as well as legal entity statements
in accord with Czech accounting standards.11
At the end of 2002, there were 70 companies listed on
the PSE with a market capitalisation of U$15.8 billion. 237 companies were listed on the RM-system,
with a market capitalisation of U$12.3 billion. Together these represented about 37% of the GDP for
2002 (ROSC, 2003b). This represents one of the highest market capitalisation:GDP ratios in any of
the accession countries in Central Europe12
. However, there has not been a single initial public
offering of equity since the Prague Stock Exchange was re-established, though there have been some
issues of bonds. Most investors only trade in six „blue chip‟ shares (Patria, 2001). Instead, large listed
enterprises have obtained funding from overseas (e.g. CEZ) or from local banks.
The development of the mid-term and long-term financing provided by the Czech capital market is
summarized in Table three. In relation to bank loans, capital market financing represents just a
fraction, though it has increased over time.
Table three: Total Market capitalisation of corporate bonds at the Prague Stock Exchange
Mil CZK
1998 1999 2000 2001 2002 2003
Market capitalisation of bonds 198,107 231,116 280,075 319,400 372,500483,800
No of issues 98 95 93 82 72 77
As compared to total bank credits 22.08% 26.57% 32.33% 32.1% 36.6% 48.3%
Source: Prague Stock Exchange Annual report 2000, Fact Book Prague Stock Exchange, 2002 p49
and monthly statistics Prague Stock Exchange, 2003, p 6, CNB statistics
wdb.cnb.cz/cnb/contrib710_prezentace.STAT_DATA_ARAD, and author‟s calculation
This would seem to indicate, generally, that there is not a large pool of equity investors providing
finance to enterprises.
95
Implementation of IFRS requirements into local legislation
A key focus of all legislative change has been to ensure that it conforms to the acquis communautaire
for accession countries to the European Union (EU) i.e. that the Czech Republic fulfils the
requirements to become a member of the EU. In the Monitoring report on the status of the Czech
Republic‟s progress towards harmonising with EU legislation, issued by the European Commission in
late 2003, it was stated that,
„The Czech Republic‟s commitments in the field of accounting have been met and it will be in a position to
implement the acquis in this field from the date of accession.‟ (Commission, 2003:22)
In other words, the legislation and regulations in accounting in the Czech Republic conform to the EU
directives and regulations. How has this process been undertaken? The table below summarises the
developments since 1991.
Table Four: Developments in Financial Reporting Regulation in the Czech Republic
Date Aim Developments
Act on
Accounting 1991
New Act appropriate to
new economic conditions
and oriented to EU
directives
Based on IVth Directive; True and Fair
introduced. Substantial penalties for noncompliance
with tax law. Limited penalties
for non-compliance with accounting law.
Followed by
detailed
regulations
Czech companies still continued to prepare
financial statements, which were more oriented to
the requirements of the tax office as they were
often the main user of financial reports (Sucher et
al 1996; Holeckova, 1996).
Accounting Act
Amendment 2001
Increase harmonisation
with EU directives and
with IFRS in preparation
for accession to EU
More detail about true and fair and it seems to
have been given the status of an „override‟ (article
7.2). However, a reference to the need for
accounts also to be correct (article 8.1) may seem
to contradict the override. In addition, finance
leases were still not to be capitalised in the
financial statements of the lessee enterprise (Act,
2001)13
.
The concept of the „fair value‟ of certain assets
was introduced, including the situation where
companies were reorganised (article 27), to bring
Czech accounting more into line with IFRS14
. A
provision for “supranational companies” to
prepare a consolidated financial statement in
accordance with international recognized
accounting standards was also introduced (article
23.6). If such a company produces such financial
statements, it will not have to provide group
accounts according to Czech accounting
standards. A substantial increase in fines for noncompliance
with the Act on Accounting was also
96
enacted (article 37).
November 2002 to
apply from 1.1.2003,
Seven general,
accounting decrees
covering different
sectors of the
economy (e.g.
entrepreneurs and
insurance companies
(Decrees 500/2002
and 501/2002).
„The substance of these decrees is expressly
stated by law and involves charts of accounts and
accounting procedures, the arrangement and
content of financial statement items, the scope of
information to be publicly disclosed and
procedures for preparing a consolidated financial
statement‟ (Ernst and Young, 2001:1).
New amended Act
on Accounting,
December 2003
(Act, 2003)
Increase further
harmonisation with EU
directives and regulation
1606/2002 on IFRS
„Czech companies trading securities on a stock
exchange in an EU member state would be
obliged to prepare accounts in accordance with
IAS. IAS will also be compulsory for the purposes
of their consolidated statements. This provision
will become effective for periods starting after 30
April 2004.‟15
The option for other companies to
apply IAS for their consolidated accounts would
remain.16
The seven decrees noted above (the
„implementing or operational directives‟) have
also been modified. In particular, the decree for
entrepreneurs (legal entities) has now been
modified to state that legal entities that are not
obliged to use double entry bookkeeping17
, are not
obliged to use fair values for certain assets and
liabilities as per article 27 of the Act on
Accounting. In other words, for smaller legal
entities, there is no need to use fair values.
In this way, the Czech Republic is providing the legislative framework to ensure that group listed
companies prepare IFRS compliant financial statements by 1 January 2005, that other group
companies may be permitted to do so, and there is some move to preparing legal entity financial
statements which may follow IFRS, to some extent, in such areas as fair value. However, there is still
the issue of exactly to what extent the Czech regulations would incorporate IFRS requirements into
the financial reporting requirements for unlisted Czech companies, whether they are groups or
individual entities. In the next section the findings from our analysis of the implementation of IFRS
in the Czech Republic are structured as laid out in the summary at the end of the literature review
above.
97
Findings
Drafting the legislation
The amendments to the Accounting Act, 1991 and subsequent proposals have been drafted within the
Ministry of Finance. An advisory body, the Finance Commission for Issues of Regulation and Future
Development of Accountancy and Audit, was set up in September 2000 to advise the Ministry of
Finance on the move to IFRS. It had 22 members, drawn from most organisations involved or
interested in accounting and audit, and met about four times a year. However, there still seemed to be
concerns that the process for drafting such amendments within the Ministry of Finance was a little
opaque (ROSC, 2003b). In particular, as accounting regulation is taken forward by the Ministry of
Finance, which also regulates and collects taxes, there might be concern that the Ministry of Finance
would not take account of the views of all prospective users of financial statements. The opacity and
slowness of the processes for drafting regulations within the Ministry of Finance was also a concern
for some of the other interviewees e.g. at December 2003, the Czech accounting standards for
implementation in 2004, had not yet been published.
This inability to prepare new decrees may arise from the slow process of drafting all new regulations
within the Ministry of Finance and the large debate about whether all aspects of IFRS were
appropriate for all entities. There was a large concern about how smaller legal entities would be able
to prepare IFRS compliant accounts, the costs, and whether it was appropriate given that their users
were mainly the tax inspectorate and local fund providers. The latter debate was referred to in our
interviews at the Union of Accountants, but it was difficult to see how extensive was the debate.
However, as we will see, this debate was not just relevant to the smaller legal stand-alone entities, but
also various group companies which were part of our survey and were listed on the RM system.
It has been suggested locally that the accounting profession could provide support for interpreting
financial reporting, by issuing accounting regulations. As one interviewee stated,
„It‟s written in the law that Ministry of Finance can give powers to prepare accounting standards to someone else,
(based on the) French model. That‟s why we composed this National Board and we wanted to be the one setting
national accounting standards, but the Ministry of Finance did not let us.‟ (interviewee L)
This national accounting board was set up in 1995. It was supposed to represent public opinion on
accounting issues and provided advice to the Minister of Finance. It had representatives from
Chamber of Auditors, Association of Accountants, Chamber of Tax advisors and Prague School of
Economics. In 2001, an authorized Czech translation of the International Accounting Standards at
31.12.2000 was produced. This was co-ordinated through the Ministry of Finance working with the
98
National Board. In late 2003, a new Czech translation of the IFRS was expected, but there was
concern in interviews that the changes in IFRS would mean that this translation would already be out-
of-date18
.
In summary, the implementation of the requirement to prepare IFRS and its ramifications for other
Czech enterprise financial reporting, has been dealt with within the Ministry of Finance which has
seemed disinclined to share the power to develop and implement the changes with other organisations.
This may also follow from the influential position of the tax inspectorate in the Ministry of Finance as
a key user of financial reports.
Users of financial reports
One of the issues that must be addressed in any research on developments in financial reporting,
assuming that the developments should be influenced by user demands, is who the users are and their
needs (Nobes, 1998a). It is difficult to obtain information on user needs for financial reports in the
Czech Republic, though previous research has already queried whether there is a large pool of
investors in the Czech Republic requiring financial statements under Czech GAAP or IFRS (Sucher
and Alexander, 2002). We asked the preparers in interviews about their experience with users of their
statements.
The main theme in the interviews, though this issue was addressed in various ways, was that it was
the regulators who required the financial statements. The interviewees stressed that the financial
statements they currently produced were for the regulators such as the Stock Exchange though
sometimes the banks expressed an interest in the accounts if the company had bank loans. It was
stressed that shareholders very rarely showed an interest in the financial statements. As one
interviewee stated,
„Unfortunately when we published our annual report, no one came and picked it up. We had to prepare annual
reports for the RM system, the Prague Stock Exchange and two others. It was a little bit frustrating that no one
was keen to know.‟ (Interviewee B)
None of the interviewees anticipated using IFRS accounts for internal management accounting
information. Two interviewees were at the stage of considering setting up integrated computer
systems that would provide the IFRS information as well as other desired financial information.
However, they were concerned about the expense and were working closely with their auditors (in
both cases one of the Big Four firms) to look at what was possible.
99
Relationship between financial reporting and taxation
In the interviews, taxation was much more of a concern for many interviewees, and how the move to
IFRS would affect the calculation of company tax.
In all countries the calculation of taxable income begins with the profit calculated according to
financial reporting rules. However, the number and type of adjustments needed to move from
accounting income to taxable income then varies greatly. Germany is an example of a country where
a large number of companies have no adjustments, and an Einheitsbilanz is prepared for both financial
reporting and tax purposes. In other countries, such as the United States, there can be many
adjustments, and tax accounting is seen as a separate field of expertise from financial reporting (Lamb
et al., 1998). In between are most other countries where various types and numbers of adjustments
are made between accounting and taxable income e.g. for timing and particular types of expenses. In
the Czech Republic, financial reporting has traditionally been heavily influenced by taxation
requirements, but this has been at the level of the individual legal entity. Consolidated statements are
not directly relevant for tax.
If IFRS were imposed for the financial reporting of individual tax-paying entities, this would change
financial reported income in a number of ways. For example, unrealized gains on investments and
long-term contracts would be treated as income. There would be much less scope to include
deductions for provisions in financial reported income (IAS 37). On the other hand marking to
market (IAS 39) might increase financial reported income. Then the issue would be how the tax
authorities would deal with the calculation of taxable income from financial reporting income.
A move to IFRS amongst individual companies in the Czech Republic could have a substantial impact
on taxable income unless:
a) There are a large number of adjustments made between IFRS income and taxable income
or;
b) The accounting for taxation is completely separated from that for financial reporting.
Currently19
there is a European Union funded project where PWC and ACCA are working together to
analyse the impact of an investor oriented accounting system on the determination of the tax base in
the Czech Republic.
In the interviews with Czech companies listed on the RM system, unprompted, three of the
interviewees expressed some concern about how the move to IFRS would affect the calculation of
taxation. They presumed that profits under IFRS would be higher than under Czech accounting
standards, and also possibly more subjective and were concerned how the Czech tax authorities would
100
deal with this. One interviewee was particularly concerned that, with the move to percentage
completion method under IAS 11, compared to the completed contract method under Czech
accounting, Czech tax authorities would use IFRS accounts rather than Czech accounts as a basis for
the tax computation. Certainly, the Income Tax Law does make many references to the Act on
Accounting whenever there are issues with more complex aspects of accounting such as treatment of
financial instruments. Individuals at one of the professional bodies, Union of Accountants, felt that
though financial reporting for those companies affected would be according to IFRS, corporate taxes
would be calculated based on specific Czech regulations. However, the detail of this is still not clear.
It is also difficult to predict what is likely to happen with the relationship between the tax authorities
and IFRS. The Advisory Commission to the Ministry of Finance seems to have spent quite a lot of
time reviewing the relationship between IFRS and income tax20
. It is possible that the delay in
implementing new regulations on accounting for stand-alone entities has arisen because of the
interaction with taxation. One interviewee commented that one of the proposed amendments to the
Act on Accounting was to extend the requirement for IFRS statements to standalone subsidiaries.
However, this had been rejected by the parliament21
.
It is also important to understand the power of the tax inspectorate in the Czech Republic. It has
already been noted that company preparers discussed their importance for the companies. Two of the
other interviewees commented that the Tax inspectorate rather then the department for accounting
methodology was the stronger power within the Ministry of Finance. One interviewee evidenced this
by reference to the key role played by the tax inspectorate in imposing penalties for non-compliance
with accounting regulations22
. The tax inspectorate was also often more interested in ensuring the
formal aspects of financial reporting had been complied with (e.g. at proper bookkeeping) rather than
whether the financial statements showed a true and fair view under s.7 of the Act on Accounting.
(Interviewee J) This interaction between s.7 and the tax inspectorate is discussed below.
Role of True and Fair
The status of true and fair in Czech legislation could be emblematic of the extent to which Czech
accounting has embraced an approach to accounting which is more focused on „substance over form‟
than one where legal form is more important in deciding such issues as control (e.g. of subsidiaries)
which might be seen to be indicative of the „European tradition.‟ Though this is a crude distinction, it
gives some indication of the differences between traditions in financial reporting which might affect
how successfully IFRS are implemented23
. The issue of true and fair and how it related to other
aspects of the accounting regulation was discussed with the other interviewees.
101
There seemed to be two different views of true and fair. On the one hand, true and fair was seen as an
override, as suggested by the Act (2001), though constrained in certain ways. On the other hand, true
and fair was interpreted as „conformity with other accounting regulations‟. Interviewees at the
international firms espoused the first view. The second view was suggested by the other interviewees.
In the first view there seemed to be a belief that true and fair was now an override in the act, that
could be used to deal with any inconsistencies, or lack of clarity in parts of the accounting legislation,
though not necessarily to deal with issues of substance over form. As the Act, (2001) specifically
stated that finance leases could not be capitalised, this was not an area where local financial reporting
could differ from the Act‟s requirement, even with the true and fair override.
As one interviewee commented, with the changes, the balance between the True and Fair requirement
and the rules issued by the Ministry of Finance was now
„A little better because of paragraph 7, the True and Fair override. Now you can diverge [from the rules]‟
(Interviewee K).
This interviewee then commented that it was not possible to use the override with the leasing
requirements, but highlighted other areas of accounting, which were confusing in the regulations, and
therefore the use of the override seemed appropriate. An example was the treatment of exchange
gains on the hedging of debtors.
The other interviewee at an international firm thought that it was very dangerous to state that the True
and Fair requirement was an override, as financial reporting was very closely tied to the tax legislation
e.g. it would not be applied to leasing. However, though it was not used often in practice it might be
used „in cases where Czech law is not prescriptive and there are grey areas.‟ (Interviewee J)
One of the interviewees at the Union of Accountants suggested that „the Ministry of Finance was
against using paragraph 7. Instead, the problem of any conflicts [in the regulations] is discussed
between the different professional bodies and the Ministry of Finance. The differences [between
regulations, Acts and the Commercial Code] arise because the Ministry of Finance is not speedy
enough in applying new regulations.‟ (Interviewee L)
It is probable, since the calculation of taxable income is still very intertwined with local financial
reporting, and given the emphasis amongst preparers on the importance of taxation for financial
reporting, that True and Fair does not really have the status of an override in practice, amongst most
102
preparers. However, this would have to be tested in a Czech court of law, before any final judgement
could be made. It would seem that the larger international firms which have expertise in applying a
„true and fair‟ approach elsewhere are treating it as a „limited override‟ to deal with areas of
accounting which are not clear.
There are also issues concerning how an override could be interpreted locally as there is little
guidance in the Act (2001) or any accounting regulations about such matters as qualities of financial
statements, or definitions of the elements of financial statements i.e. there is no conceptual framework
similar to the IASB framework. The principles of going concern (section 7.3) and materiality (section
19.5b) were included in the Act (2001). The provisions regarding materiality stated that enterprises
should include in their financial statements any facts „of their consequences have a significant
(material) effect on the view of the accounting unit‟s financial position.‟ (Section 19.5b) This would
seem to reflect some view of users in general, but this must be balanced against the comments above
about taxation and true and fair.
In summary, it would seem that true and fair, as an override, will only have a limited effect on the
financial reporting of Czech companies, and that taxation will be a greater influence. The intertwining
of financial reporting with taxation will complicate the use of IFRS for individual Czech legal entities.
Other complications may arise from specific differences between local Czech accounting regulations
and IFRS.
Differences between local and IFRS financial reporting
The various differences between local Czech accounting and IFRS have been documented elsewhere
(IFAD, 2000; IFAD, 2001; ROSC, 2003b). These differences may affect the capacity of local
companies to prepare IFRS financial statements (for example with such areas as revenue recognition
and definitions of subsidiaries). Therefore a summary of some of the key differences between Czech
financial reporting regulations and IFRS extant at April 2002, relevant to this research, is shown in
Table five.
103
Table Five: Some of the Key differences between Czech and IFRS accounting
Topics Czech accounting IFRS
True and Fair override If exceptionally the given
accounting method hides
reality, an enterprise must
adopt an alternative method
and proceed in such a way
that a true and fair view is
provided
In very rare cases override
the standards where essential
to give true and fair view
Historical cost Historical cost, except for
certain assets and liabilities
Historical cost, but intangible
assets, PPE and investment
property must be revalued.
Derivatives, certain
agricultural assets and most
securities must be revalued.
Substance over form Mostly legal form over
substance e.g. all leases are
regarded as operating leases
and are not capitalized in
lessee balance sheet.
Accounting records must
reflect the economic
substance of transactions
despite their legal form. E.g.
Finance leases are capitalized
in balance sheet of lessee.
Construction contracts % of completion method is
not allowed. Accounting
follows contractual
agreements. In most cases
completed contract method is
used.
% of completion method is
used.
Definition of subsidiary Power to exercise dominant
influence and full control
over entity‟s operations, by
virtue of holding more than
50% of the ordinary shares or
based on a contract or
Articles of Association.
Consolidated financial
statements seen as
„secondary‟ compared to
stand alone financial
statements in practice.
Based on voting control or
power to exercise influence.
Property, Plant and
Equipment (PPE)
Use historical cost (in certain
cases replacement cost).
Revaluations allowed only if
acquired as part of
acquisition of a business or
investment. Spare parts not
part of PPE.
Use historical cost or
revalued amounts. Frequent
revaluations of entire classes
of assets required.
Impairment Only general requirement to
reflect possible impairment.
No detailed guidance on
measurement.
If impairment is indicated,
write down assets to higher
of net selling price and value
in use based on discounted
cash flows
Financial assets -
measurement
Similar to IAS. Held for
trading securities are limited
Depends on classification of
investment – if held to
104
to those traded on stock
exchanges.
maturity or originated by
entity then carry at amortised
cost, otherwise at fair value.
Revenue recognition No separate guidance Recognise revenue if meets
specific criteria
Provisions – general
Record provisions relating to
present obligations from past
events if probable outflow of
future resources. Also legal
provisions permitted for
future repairs and
maintenance on property,
plant and equipment.
Record provisions relating to
present obligations from past
events if probable outflow of
resources can be reliably
estimated.
Segmental information No specific requirement. All
companies must provide a
split of revenue between
domestic and export sales
and between main ordinary
activities
Report extensive information
on primary and secondary
segments based on risks,
returns and internal reporting
structure
Adapted from IAS, US GAAP and CZ GAAP, PWC, April 2002.
In the interviews with preparers, leasing, and the need to capitalise leases, was seen as a problem for
preparing IFRS accounts. However, on further discussion it was agreed that this was less problematic
once the basic numbers and values had been derived and a spreadsheet could be developed.
The auditors raised the issue of IAS 36, and impairment, as a problem when dealing with enterprises
preparing IFRS statements. One auditor commented that the,
„Drive for impairment comes from the auditors. The auditor asks the company to perform the appraisal [he]…had
never seen a company do it of its own volition.‟ (Interviewee K)
Issues with particular IFRS
The treatment of financial derivatives and provisions under IAS 39 and IAS 37, respectively, and the
issue of impairment under IAS 36 are all aspects of financial reporting which are likely to cause
problems for accountants in transitional economies, just as, given their complexity, they are likely to
cause problems for accountants in many companies (e.g. Paterson, 2001 discussing the situation in the
UK). There are however, other specific issues with transitional economies, which have been identified
already (Sucher and Alexander, 2002; ROSC 2002a; ROSC 2002b). These may broadly be organised
into three key aspects; interpretation of fair value, use of estimates and the preparation of group
accounts.
105
Fair value is a concept included in several IFRS (Sucher and Alexander, 2002; ROSC, 2002b), though
guidance on its application does vary between IFRS (Nobes, 2001). The definition used throughout
the IFRS is,
“The amount for which an asset could be exchanged between knowledgeable, willing parties in an arm‟s length
transaction” (IAS 16: IASB, 2003).
In IAS 16, Property, plant and equipment, it is stated that “fair value of land and buildings is usually
its market value” (IAS 16:para 30) and that “the fair value of items of plant and equipment is usually
their market value determined by appraisal.” (IAS 16:para 31). However, in many transitional
economies there are no deep markets for many of the fixed assets so described.
Other research has indicated that the use of fair values can be an excuse for manipulation of financial
accounts (Xiang, 1998), where supposed “arms length” transactions are enacted between related
parties. This could also be an issue in the Czech Republic, given the corporate governance issues of
the mid 1990s.
With the implementation of „fair value‟ into the Act (2001), there has now been further experience of
using fair values in the Czech Republic. In our interviews we asked about the use of fair values. The
difficulties that were noted revolved around obtaining market values and the cost of using external
valuers.
In one company where they were already preparing IFRS statements, the interviewee‟s comments
about the inability to estimate fair values for equity investment was at variance with the statement on
valuation in the IFRS statements24
. The interviewee commented that the auditor had been happy with
the published version, even though it did not reflect the actual problems of fair valuation in the
company. The amounts involved were not material to the consolidated balance sheet.
Six of the companies made reference to using ‟court appointed appraisers‟ to assess fair values for
property in particular, but not exclusively. The perception seemed to be that though costly, if
necessary, each company could (or was already doing so) use a court appointed appraiser to estimate
fair values.
One of the auditors raised the issue of the conflict between the Commercial Code and Act on
Accounting requirements regarding when fair values were required in company re-organisations. He
also stated that it was unclear whether any losses on revaluation in such circumstances would be
allowable for taxation. (i.e. there were conflicts and confusion between different laws). The other
106
auditor commented that there was a problem with fair values as Czech markets were not so developed,
particularly with the PSE, and one had to „engage a reputable valuer, speak to the appraiser and say
why we want them to do the fair value.‟ (Interviewee M)
Another interviewee commented that two different expert appraisers could provide different opinions
about a fair value and that it was therefore not objective. In particular in the Czech Republic, there
was a lack of stability (in the stock market) and prices could fluctuate wildly. These issues with fair
values could affect the reliability of financial statements prepared under Czech accounting of IFRS,
where fair values are used.
Many IFRS are built around the use of estimates in valuing assets and liabilities for inclusion in the
balance sheet, and therefore what constitutes income and expenditure in the financial statements.
This is the case in particular with IAS 11, Construction Contracts. This was an issue for two of the
preparers interviewed, working in construction companies, who were already preparing IFRS
statements. Obtaining the management information necessary to estimate the stage of completion of
contracts for IAS 11, construction contracts, was considered a particular problem. Applying IAS 11
meant a large increase in the amount of information that needed to be generated internally on such
issues as „contract variations‟ and there was not enough guidance in IAS 11 about what should be
done25
. One firm had turned to its auditors for detailed guidance.
Previous research has commented on the problems concerning consolidated accounts. There has not
been a long tradition of preparing group accounts in the transitional economies. Deciding which
companies are controlled, under IAS 27, and therefore need to be consolidated is particularly
problematic for auditors of large companies. Recent interviews with audit partners indicated that
clients did not see why such companies needed to be consolidated, or were not keen to provide the
information (Sucher and Alexander, 2002). It was often difficult to identify which companies should
be consolidated in the group financial accounts.
“In this part of the world there have evolved very circular complex ownership structures. Groups are often not
organised on straight divisional lines. Which means that ensuring all related parties in the group (are accounted
for) is a problem. A mix of taxation and a lack of transparency as to who exercises control makes for complex
group structures.” (Sucher and Alexander, 2002:52)
Enforcement
Any enforcement of compliance with IFRS in the Czech Republic is likely to build upon the system
for enforcement already in place. There is a requirement to file audited financial statements with the
Trade Registry. However, as has been noted in previous research and reports (Sucher and Zelenka,
107
1998; ROSC, 2003b) though this is a requirement in many cases it is not observed. In those financial
statements, which were filed, some substantial non-compliance with Czech regulations was also
observed (ROSC, 2003b).
The current enforcement of compliance with financial reporting standards, noted in the Act, 200126
,
was dealt with by the tax inspectorate. There was some concern about this amongst the interviewees,
as different tax offices were not consistent in their treatment
„So tax inspectorate imposing penalties on companies not doing accounting according to standard. Some
companies took tax inspectorate to court. Now have decisions by court that [on the one hand] can enforce and
[on the other hand] cannot enforce it. No precedence here. District Courts for Ceske Budejovice and Brno
decided differently [on the same accounting issue]‟ (Interviewee L)
Therefore under the current Czech taxation system, regional courts interpret the tax law in different
ways in different parts of the country. As noted by the deputy finance minister Yvona Legerska,
„In one region you can be fined for accounting mistakes while in another you can‟t.‟ (Carey, 2003)
However, there is also the issue that case law is very rarely used. Therefore a decision could be made
by one regional tax office, but this would not be binding or relied upon. If a decision was changed, it
could then be retroactively applied. This has caused much uncertainty for companies operating in the
Czech Republic (Carey, 2003).
In the Act, 2003, it is now stated that penalties will be levied on accounting units, which are obliged
to prepare IFRS financial reports, and do not do so27
and these penalties should be collected by the
Tax Inspectorate. The penalties for non-compliance with financial reporting regulations had been
increased to a maximum of 6% of balance sheet gross assets. There seems to be no requirement to
monitor actual compliance with IFRS28
The Czech Securities Commission also reviews the financial statements of listed companies for
completeness and obvious errors. On its website it carries all the latest publications from the Centre
for European Securities Regulation (CESR) which suggest how IFRS compliance should be enforced.
On the website, there is a comment that,
„The Commission, together with the Ministry of Finance and the Czech National Bank, is creating conditions for
the performance of supervisory activities over the so-called consolidated financial groups.‟
29
However, it is not yet clear what these conditions might be. The Commission has certainly not had
the technical resources to enforce accounting standards until now (ROSC, 2003a). It is unclear how
108
this is likely to improve in the short-term. Therefore a large burden of the enforcement of compliance
with IFRS will fall on the auditors.
Role of the audit firms
In the Czech Republic, the audit market is divided into the Big Four international audit firms, audit
firms affiliated to the smaller international networks of audit firms, medium sized local Czech audit
firms and local sole practitioners. The larger, listed Czech enterprises and international companies are
usually audited by the Big Four audit firms. The smaller Czech enterprises are audited by the smaller
Czech audit firms and sole practitioners. However, many of the smaller Czech enterprises may still be
listed entities on the RM system, and therefore subject to the requirement to prepare IFRS statements
from 2005.
Concerns have been raised in previous and forthcoming research about the ability and capacity of
smaller Czech audit firms to be independent of their clients (Sucher and Zelenka, 1998; Sucher and
Kosmala MacLullich, forthcoming). In particular, local auditors focused to a greater degree on
auditing financial statements to ensure that the statements are correct for the tax computations rather
than establishing their truth and fairness as required by the Act on Accounting (Sucher and Kosmala
MacLullich, forthcoming). This may seem appropriate when the main user of published financial
statements is the tax inspectorate.
Whilst it has been suggested that only the international Big Four audit firms may have the ability to be
truly independent of their clients in transitional economies (Xiang,1998) there is research that
indicates that sometimes these firms themselves may have problems in maintaining their
independence, which might affect compliance with IFRS. For example there is evidence that in some
cases, with IFRS reporting, the Big Four firms both prepare30
and audit the IFRS accounts (Sucher
and Alexander, 2002). However the degree to which this is the case does seem to vary between firms.
It could be argued that, in the first few years of the enterprise‟s transition to IFRS, it might be
necessary for the audit firm to work closely with the client in preparing the IFRS statements (ROSC,
2003a). However, it is therefore important to ensure that there are suitable controls in place to ensure
the proper separation of the preparation and audit of the financial statements. Previous research in the
Czech Republic has indicated that this might not always be the case. Partly this might be because,
„Assistance in preparation and audit of IAS financial statements can be quite time consuming (especially the
deferred tax calculations and presentation) and therefore it can be more economic to have one team both
preparing and auditing the IAS financial statements.‟ (Sucher and Alexander, 2002)
109
On the other hand one could be rather more cynical and suggest that a large amount of income can be
derived from advising on IFRS accounting and therefore, as commented on by an interviewee who
had worked in one of the Big Four firms,
„Some financial managers and accountants [in enterprises] do not know what is going on [with IFRS]. They are
only the passive receivers. In [X audit firm], there are templates that transform Czech accounting to IAS. The
knowledge is kept in the audit company. It is big business [for the audit firm]. You can train companies to do the
supporting sheets but the final bit is done by the audit firm.‟ (interviewee J).
The relationship with auditors and the preparation of IFRS was discussed with the company
interviewees. Six of the companies were audited by individual or smaller Czech audit firms. Two
were audited by Big Four audit firms, and a third by a Czech audit firm, which was part of an
international network. As many of the interviewees had not made any preparations at all for the
transition to IFRS, there was little issue of auditor involvement. In those cases where the interviewees
had made some preparation for IFRS the situations varied. In one case, the interviewee had been on
IFRS courses, had undertaken a test run with the 2002 financial statements, would work with their
auditors on IFRS financial statements for 2003, and would „go live‟ on their own for 2004.
One company was already preparing IFRS financial statements and had done so since 2002. Initially,
their auditors, one of the Big Four firms, had provided a lot of support and schedules to prepare the
IFRS accounts. The company had found it particularly helpful to have support in the setting up of the
necessary management information systems as the main issue they had had in preparing IFRS
financial statements was the lack of necessary management information, particularly for contract work
and leases. The company had changed auditor between two of the Big Four audit firms. The
interviewee drew an unfavourable comparison between the full support offered by the previous Big
Four firm, where the firm had provided full support with IFRS, with the current Big Four audit firm,
where the interviewee had been told that it was up to the enterprise to prepare the IFRS statements,
and not for the auditor.
Local professions, education and training
Since 1989, the local professions have grown and developed. There are three main organisations, the
Union of Accountants, the Chamber of Auditors and the Chamber of Tax Advisors. Here we will
focus on the first two organisations, which are very active in accounting.
110
The Union of Accountants (Union) is a professional organisation of accountants where membership is
voluntary. At the end of 2002, it had about 4,900 members (ROSC, 2003b). It co-operates with the
Ministry of Finance in drafting legislation through its membership of the National Council, and its
membership of the advisory body on accounting and audit. It publishes a monthly magazine,
Ucetnictvi, which includes practical articles concerning the application of IFRS. It also organises
training courses on IFRS for local accountants.
All registered auditors and audit firms must be members of the Chamber of Auditors. At 31st
December 2003, there were 1,239 registered auditors.31
There is a committee within the Chamber of
Auditors, which deals with IFRS, and submits comments on IFRS exposure drafts to the IASB. The
Chamber of Auditors also provides training courses on IFRS for its auditors.
Aside from the training courses on IFRS provided by the Union and the Chamber of Auditors, courses
are provided by the Prague School of Economics. Some of the larger accountancy firms also provide
courses on IFRS. As an example, Ernst and Young operates an IFRS Academy, in collaboration with
the department of financial accounting at the Prague School of Economics, which provides training in
both the theory and practice of various IFRS32
.
The Association of Certified Accountants (ACCA) is also active in the Czech Republic. Many of the
largest international accountancy firm‟s train their accounting staff under the ACCA qualification,
studied and examined in English, which covers IFRS. Currently there are 500 members of the ACCA
and 1,000 ACCA students in the Czech Republic.33
Amongst those companies interviewed, five of the interviewees had undertaken no training yet in
IFRS. Three of those interviewees who had undertaken no training so far did not seem to anticipate
undertaking any future training. The others had already arranged future training programmes with
either the Union of Accountants or the Prague School of Economics.
One interviewee had been to a course where there had been a general overview of IFRS. All the other
three interviewees had undertaken training courses on IFRS with Big Four audit firms as well as with
other providers such as the Union of Accountants. These interviewees contributed most to the
discussion on issues and concerns with the transition to IFRS.
The interviewees commented that the most helpful training was that which addressed the overall
principles of IFRS and provided detailed practical examples, with associated workbooks. How to
establish an appropriate management information system to produce the necessary information for the
IFRS statements was also of concern.
111
Several interviewees commented that they obtained most of their information from the internet, and
there was little information on the internet in the Czech Republic about the move to IFRS. A few
interviewees also subscribed to Ucetnictvi, which they found helpful. However, not all interviewees
were members of the Union and therefore they did not receive the journal. Those interviewees who
seemed to have a greater knowledge of IFRS, commented that they would very much like to be able to
access a web-site which carried information on IFRS and which provided support for any queries they
might have.
In summary, there would seem to be a large pool of courses available on IFRS from commercial
courses run by the large accountancy firms to complete accountancy qualifications, which include
education in IFRS. However, some of the qualifications, such as the ACCA training, might be
regarded as expensive for local Czechs. The provision of the new ACCA Diploma in IFRS may help
training in IFRS for some accountants, but it is only offered in English, and is therefore likely to have
a market only amongst the larger accountancy firms and listed enterprises. A review of the
professional bookshops in Prague also indicated that there were several books available, which
provided detailed guidance on how to prepare IFRS financial statements. However, in the small
sample of companies listed on the RM system that we interviewed, only four out of nine of the
interviewees had been on training courses, and some did not yet anticipate going on any such courses.
This reflected their overall preparedness for implementation of IFRS for the company.
State of preparedness for implementing IFRS
In the majority of the companies interviewed, the interviewees stated that they would generally wait
until they had a detailed regulation from the Ministry of Finance on IFRS before they took any action
over preparing IFRS accounts. There were three exceptions to this; one company, which was the
largest in its sector, another company that was a subsidiary of an overseas34
company and a third
company, which was already preparing IFRS accounts with a view to encouraging future foreign
investment. These latter three companies were already testing out aspects of IFRS accounts, or had
already prepared IFRS accounts. They were the companies, which were also audited by international
audit firms.
Though, given the many changes in accounting and tax legislation, the „wait and see‟ attitude could be
understood, it could be inferred that, taken with the general responses when we were setting up the
interviews35
, many smaller Czech listed companies have not yet taken any action over preparing IFRS
accounts.
112
Conclusions
In this paper, as the Czech Republic prepares for its accession to the EU, we have studied how the
requirement to prepare IFRS for group listed entities, and the general move to bring Czech accounting
more into line with IFRS has been implemented. Following from a review of the relevant literature,
we structured our study to consider the role of the Ministry of Finance; the influence of likely user
needs on the move to IFRS, in particular the influence of the tax inspectorate, and the place of true
and fair, as an indicator of the accounting tradition. Given the recent history, as the literature suggests,
we noted how the locus of the implementation process has remained in the Ministry of Finance, which
has seemed disinclined to share its regulatory power. This has possibly slowed down the
implementation process and must cast doubt on whether the changes in financial reporting will benefit
a wider group of users outside the tax inspectorate.
Currently, individual legal entities will not have to prepare IFRS financial statements, but
will follow Czech accounting. How this will interface with the requirement for listed group
companies, which may own such entities, to prepare IFRS statements is unclear. Though
Czech accounting is moving closer to IFRS in such areas as fair values, it still differs in certain
material respects from IFRS (see examples in table five). In might be that it will be left to the
individual group to ensure it obtains the necessary information in areas such as segmental
information.
There are also issues concerning how „grey areas‟ of accounting, which are either confusing in the Act
(2003) and its decrees, or are not covered, may be interpreted within the requirement to prepare
accounts which are True and Fair. In the absence of a conceptual framework to support True and Fair,
the interests of the tax inspectorate and an emphasis on form over substance may prevail.
There are different groups of entities in the Czech Republic at different stages of development with
the implementation of IFRS. As noted in previous research, there is a small group of large listed
Czech companies, usually funded by some level of overseas finance, which have already commenced
preparing IFRS financial statements, and there is previous research that has studied their
progress (Sucher and Alexander, 2002)36
. These companies are audited and serviced (e.g. with
IFRS up-dates) by the international audit firms (predominately the Big Four) where professional staff
usually speak English and have been trained in international accountancy qualifications. The situation
is probably similar for the Czech subsidiaries of overseas companies.37
113
Then there is a group of large Czech enterprises listed on the RM system, which have to apply IFRS
after 30 April 2004, and are not yet prepared to any extent, for the change. These firms were usually
audited by local Czech audit firms. In discussion about the accounting challenge faced by the
accession countries, the head of the IASB has commented that, „some companies could be rough and
ready in their application of the IASB‟s rules in 2005 because of the scale of the changes.‟(Parker,
2004) In the Czech Republic it is possible that there are other considerations. Since the „Velvet‟
revolution of 1989, there have been many changes to accounting and tax legislation. In a time of
constant change, it might be considered advisable to „wait and see‟. Of the group of firms listed on the
RM system, those, which were audited by the international audit firms, were better prepared than
other firms. It is likely that this situation may be found in many other accession countries.
There do seem to be a large number of training courses available on IFRS, of many different types
and from many different sources, which could help with the training process, in the short term, though
we were not able to assess the quality of these programmes. It would be particularly useful to have
some central database of all the programmes available (maybe on a web-site) with up to date
information of what is happening with IFRS implementation, and case studies of companies that have
already implemented IFRS in different business sectors.
With those companies that were already testing out the preparation of IFRS accounts, the need for
much more detailed management information to support the preparation of IFRS statements was
emphasised. This aspect of the transition to IFRS may have been underestimated in some of the
training. There was also a concern about the cost of moving to using fair value for more valuations of
assets and liabilities and the reliability of the fair values derived. As fair values are used more within
IFRS, the role of the valuation profession will become more important. It would be interesting to
research how valuers in the Czech Republic derive fair values for assets and liabilities where there are
no market prices.
Yet, there is also a more fundamental issue. Though one would have to undertake research with the
users of the accounts, with the majority of interviewees, though these were listed companies, there
was little concept of a large body of investors interested in information about the company. In the
three companies where they had progressed with training on IFRS they were very specific reasons
related to outside investors (current or potential), which seemed to have led the interviewees to be
more pro-active. It must be debateable whether it is to benefit of the individual company, or the
country (given the necessary costs of training and conversion) for all the companies listed on the RM
system to move to IFRS.
114
Though the majority of Czech preparers and auditors may not yet be ready to implement IFRS
properly, as the main users of Czech financial statements are the tax inspectorate, and the main basis
for the tax calculations are the individual Czech financial statements, this may not be such a problem.
However, the linkage between the preparation of IFRS statements at individual level and the
preparation of accounts for the calculation of taxation due needs to be clarified. This is not just an
issue for the Czech Republic but also for most members of the EU. In a recent report, reviewing the
move to IFRS in many countries and its impact on taxable income, it has been suggested that there
should be separate systems of tax and financial reporting such that taxation is not „polluted‟ by
financial reporting (Nobes, 2004). Though this has cost implications, it may be the most effective
approach to ensuring that the different objectives for taxation and financial reporting systems are met.
The audit firms are likely to continue to play a key role in enforcing the detail of compliance with the
IFRS. Concerns have been raised concerning compliance with IFRS in audited financial statements of
Czech companies (ROSC, 2003b and Sucher and Alexander, 2002). There have also been some high
profile cases of audit firms (local and international) signing off Czech financial statements, which has
subsequently been shown to be materially misstated. No audit firm has been sued or disciplined38
for
any infringements of audit regulations, which may cast doubt on how effective will be the
mechanisms for monitoring audit quality and how effective auditors will be in enforcing compliance
with IFRS. It will be necessary to set up some independent regulator, maybe attached to the
Securities Commission, to enforce compliance on a pro-active, risk assessed basis (CESR, 2002).
There are various limitations to our research. The implementation of IFRS into Czech legislation and
practice is in a state of flux, and we have looked at the situation at one point in time, at the end of
2003. Though we have tried to capture some of the detail of the implementation process, and the
forces at work, it is still not possible to identify all that is likely to happen, though we can speculate.
Our comments on what users may want from financial reporting in the Czech Republic is, necessarily,
anecdotal in the absence of a full scale survey of user needs, and substantially influenced by the views
of the preparers. In addition, we did not interview a large number of enterprises listed on the RM
system, though the overall responses from those we contacted, as well as those we interviewed, can
probably provide a reasonable picture of the situation with preparers at the end of 2003. We have also
used a structure in approaching our analysis, based on previous research in the area, which may have
sacrificed depth for breadth of the topic. However, this paper provides a pointer to areas for the
research that could be developed in greater depth and are likely to be of concern in other accession
countries.
115
Appendix one – Illustrative questions for companies
Personal background of interviewee
Post; education; age; position
Company
Ownership; Products; key issues? Who are the auditors?
In preparing financial accounts for your company, what audience do you have in mind and why?
Where is your chief source of funding?
IFRS
Are you already preparing IFRS?
Why do you think that IFRS are being implemented?
How much do you know about IFRS? Can you give examples of areas that cause you concern?
How will you plan the implementation?
What issues do you see in implementing IFRS for your company?
Where could training usefully be focused?
Have you been on any courses – which?
Have you access to the Czech translation of the IFRS?
A key term in IFRS is “fair value” . Could you, as far as possible, explain to us what it means to you?
To what extent will you be using IFRS accounts for management information?
Who will be using your IFRS accounts? Do you get any Qs from investors?
If you are not sure about the treatment under IFRS, how will you deal with it/who will you ask?
116
References
Act (1991), Act on Accounting Zákon č. 563/1991 Sb., o účetnictví, Trade Links translation
Act (2001), Act on Accounting Zákon č. 353/2001 Sb., o účetnictví, Trade Links translation
Act (2003), Act on Accounting Zákon č. 437/2003 Sb., o účetnictví, Trade Links translation
Alexander D. (1993) “A European true and fair view?” European Accounting Review Vol. 2 No. 1
pp59-80.
Alexander D. (1996) “Truer and Fairer; uninvited comments on invited comments”. European
Accounting Review Vol. 5 No.3 pp 483-493. .
Alexander, D. (1999) „A benchmark for the adequacy of published financial statements‟, Accounting
and Business Research, Vol. 29, No.3 pp239-53.
Alexander, D. and Archer, S. (2000) „On the “myth” of Anglo-Saxon Financial Accounting‟, The
International Journal of Accounting, Vol. 35 issue No. 4 pp539-557
Bailey, D. (1995) “Accounting in Transition in the Transitional Economy”, European Accounting
Review, Vol.4 No.4 pp596-623
Bailey, D. and Alexander, D. (2001), “Eastern Europe: Overview” in European Accounting Guide,
New York, Aspen
Cairns, D. (2001) International Accounting Standards Survey 2000, David Cairns, International
Financial Reporting.
Cairns, D (1997) “The future shape of harmonisation: a reply”, European Accounting Review, Vol. 6
No.2 pp305-348
Carey, N. (2003) „FinMin plans on major tax system shake-up‟, Prague Business Journal, 22nd
September. http://www/ pbj.cz
CESR, (2002) Proposed Statement of Principles of Enforcement of Accounting Standards in Europe,
Consultation paper, CESR/02 – 188b
117
Decree, (2002) Decree on Double Entry Bookkeeping for Businessmen No.500/2002 Coll Vyhláška,
kterou se mění vyhláška č. 500/2002 Sb., kterou se provádějí některá ustanovení zákona
č. 563/1991 Sb., o účetnictví, ve znění pozdějších předpisů, pro účetní jednotky, které jsou podnikateli
účtujícími v soustavě podvojného účetnictví (Trade Links translation)
EBRD (2000) Transition Report 2000, Economic Bank for Reconstruction and Development, London
EBRD (2003) Transition Report 2003, Economic Bank for Reconstruction and Development, London
Emenyonu, E.N. and Gray, S. (1996) “ International Accounting Harmonisation in Major Developed
Stock Market Countries: An Empirical Study”, International Journal of Accounting, Vol. 31 No. 3
pp269-279
Ernst & Young (2001) Accounting Act Amendment a First Step – Implementing regulations will be
key, Discussion document provided to the author.
EU (2000) The EU’s Financial Reporting Strategy: the Way Forward, COM(2000)359,13.06.2000
EU (2002) Regulation of the European Parliament and of the Council on the application of
international accounting standards, Brussels.
Commission (2003) Comprehensive Monitoring Report on the Czech Republic’s preparations for
membership (accessed at Europa.eu.int/comm./enlargement/report_2003/pdf/cmr_cz_final.pdf)
Flower, J. (1997) “The future shape of harmonisation: the EU verus the IASC versus the SEC”,
European Accounting Review, Vol. 6 No.2 pp281-303
Garrod, N. and McLeay, S. (1996) Accounting in Transition, London Routledge
Holeckova, J. (1996) „Relationship between accounting and taxation in the Czech Republic‟.
European Accounting Review, 5 Supplement: 859-870.
Hove, M.R. (1986) “Accounting practices in developing countries: colonialism‟s legacy of
inappropriate technologies”, International Journal of Accounting, Fall, pp81-100
IASB (2003) International Financial Reporting Standards, London
118
IFAD (2000) GAAP 2000, a survey of national accounting rules in 53 countries,
IFAD (2001) GAAP 2001, a survey of national accounting rules benchmarked against international
accounting standards,
Jermakowicz, E. and Rinke, D. (1996) “The New Accounting Standards in the Czech Republic,
Hungary, and Poland vis-à-vis International Accounting Standards and European Union Directives”,
Journal of International Accounting and Taxation, Vol. 5 No.1 pp73-87
Jindrichovska, I. (2001) The Relationship between Accounting Numbers and Returns: Some
Empirical Evidence from the Emerging Market of the Czech Republic” The European Accounting
Review, 10:1, pp.107-131
Lamb, M., Nobes, C., and Roberts, A.(1998) “International variations in the connections between tax
and financial reporting”, Accounting and Business Research, Summer
Lesenarova, H. (2003) „Ernst and Young fine too small by billions, Mlynar‟s joy too unreserved.‟
Prague Business Journal, accessed via http://www.pbj.cz/user/article.asp?ArticleID=179801
Murphy, A.B. (2000) „The impact of adopting IAS on the harmonization of accounting practices,
International Journal of Accounting, 35: 4pp 471-93.
Nobes C. (1983) “A judgmental international classification of financial reporting practices” Journal of
Business Finance and Accounting, Vol. 10 pp 1-19.
Nobes, C. (1998a) Accounting in Developing Economies, ACCA, London
Nobes, C. (1998b) „Towards a general model of the reasons for International Differences in Financial
Reporting‟, Abacus, Vol.34 No.2 pp162-187
Nobes, C. (2001) Asset Measurement Bases in UK and IASC Standards, ACCA, London
Nobes, C. (2004) A Conceptual framework for the taxable income of businesses and how to
apply it under IFRS, ACCA, London
Ordelheide D. (1990) Soft transformation of accounting rules of the Fourth Directive in Germany.
Paris. Les cahiers internationaux de la comptabilité.
119
Ordelheide D. (1993) “True and fair: a European and a German perspective”. European Accounting
Review 2(1) pp 81-90.
Ordelheide D. (1996) “True and fair: a European and a German perspective II”. European
Accounting Review 5(3) pp495-506.
Parker, A. (2004) „EU‟s new states face accounting challenge‟, Financial Times, January 5th
, p1
Paterson, R. (2001) “Widening the GAAP”, Accountancy, May
Patria, (2001) Czech equity Market focus, September, issued by Patria Research, Prague
ROSC (2001) „Report on the Observance of Standards and Codes: Slovakia‟, World Bank, November
27
ROSC (2002a) „Report on the Observance of Standards and Codes: Croatia‟, World Bank, June 20
ROSC (2002b) „Report on the Observance of Standards and Codes: Lithuania‟, World Bank, June 28
ROSC (2002c) „Report on the Observance of Standards and Codes: Poland‟, World Bank, July 25
ROSC (2002d) „Report on the Observance of Standards and Codes: Bulgaria‟, World Bank, December
16
ROSC (2003a) „Report on the Observance of Standards and Codes: Romania‟, World Bank, May 9
ROSC (2003b) „Report on the Observance of Standards and Codes: Czech Republic‟, World Bank,
July 17
Schroeder, M. (1999), “A description of the contents of the long-form statutory audit reports of a
sample of Polish listed companies for 1996” European Accounting Review, Vol.8 No.1 pp1-23
Seal, W. Sucher, P. and Zelenka, I. (1995) “The changing organisation of Czech accounting”,
European Accounting Review, Vol.4 No.4 pp659-684
Street, D.L. and Gray, S.J. (2001) Observance of International Accounting Standards: Factors
Explaining Non-compliance, ACCA Research Report No. 74, ACCA, London.
120
Sucher, P. and Alexander, D. (2002) IAS: issues of country, sector and audit firm compliance in
emerging economies, London: Centre for Business Performance of the Institute of Chartered
Accountants in England and Wales.
Sucher, P. and Bychkova, S. (2001) “Auditor independence in economies in transition: a Study of
Russia”, European Accounting Review, Vol.10 No.4 pp817-841
Sucher, P. and Kosmala MacLullich, K (forthcoming) “A construction of auditor independence in the
Czech Republic: Local insights”, Accounting, Audit and Accountability Journal
Sucher, P. and Zelenka, I. (1998) “The development of auditing in the Czech Republic”, European
Accounting Review, Vol.7 No.4 pp723-741
Sucher, P., Seal, W. and Zelenka, I. (1996) “True and Fair in the Czech Republic: a Note on Local
Perception”, The European Accounting Review, Vol 5:No.3, pp545-557.
Tay, J.S.W. and R.H. Parker (1990) “ Measuring International Harmonisation and Standardisation”,
Abacus, Vol.26 No.1 pp71-88
Xiang, B. (1998) “Institutional factors influencing China‟s accounting reforms and standards,”
Accounting Horizons, Vol.12 No.2 pp 105-119.
121
NOTES
1
Czech Republic, Estonia, Cyprus, Latvia, Lithuania, Hungary, Malta, Poland, Slovenia and Slovakia
2
Though international financial reporting standards (IFRS) were named international accounting standards
(IAS) before 2002, we will use IFRS as a phrase to encapsulate previous and current IAS, unless they are so
named in a paper.
3
Though see Alexander and Archer, 2000 for a discussion of the „myth‟ of Anglo Saxon accounting.
4
ROSC stands for Report on the Observation of Standards and Codes. The reports are prepared by the World
Bank, in collaboration with representatives of local organisations such as professional bodies. They focus on
such aspects as accounting and audit. The ROSC reviews for accounting and audit can be downloaded at
http://www.worldbank.org/ifa/rosc_aa.html
5
http://www.worldbank.org/ifa/rosc_aa.html; accessed 5
th
January 2004
6
IFAD stands for International Forum for Accountancy Development. The reports it produced can be accessed
at www.ifad.net
7
IAS stands for International Accounting Standard.
8
The RM-System was originally a one-off mechanism for distribution of shares in the Voucher Privatisation in
the early 1990s. During the privatisation company shares were administratively created, based on the book
information of privatised enterprises. The distribution algorithm assigning shares to privatisation vouchers was
matching supply and demand of these administratively created shares. After distribution the shares began to be
traded on the RM-System, which now resembles an over-the-counter market. (Jindrichovska, 2001) RM-system
has a bit less strict regulations and the population of RM system are mostly mid-size companies in comparison
to the Prague Stock Exchange. Most shares were, and still are, traded on both the RM-System and the Prague
Stock Exchange, even though some companies have delisted from the PSE as a reaction to introduction of
stricter reporting regulations.
9
http://www.rmsystem.cz
10
This was for the convenience of the researchers.
11
The World Bank Report on the Czech Republic suggests that companies listed on the PSE must prepare up to
four sets of audited financial statements – legal entity accounts and consolidated accounts under IFRS and
Czech accounting standards – which seems somewhat onerous (ROSC, 2003b).
12
Equivalent figures for other Central European accession countries in 2002 are as follows: Poland, 14.3%;
Hungary, 17.4%; Estonia, 33.6%; Latvia, 8%; and Lithuania, 9.5% (EBRD, 2003).
13
The Czech leasing association is very active in the Czech Republic. The following is stated on its web site, in
its annual report, 2002, ‟The Leasing Association wants to take part in preparing and amendments of the
national accounting standards on accounting for leases. It will strive for adoption of amendments on insurance
acceptable for non-banking financial products. Furthermore, it will strive for enforcement of non-applicability of
the Law
No 321/2001 on financial leasing, and for maintaining the current regime of leasing and other non-banking
financial products as it is in accordance with the Income Tax Law and with the VAT Law.‟
http://infos.eunet.cz/leasoc/rocenka/2002/3.htm. Author‟s translation.
14
According to an academic involved in the translation of the term „fair value‟ into Czech legislation, there was
a substantial debate about how the term should be translated. There were three possible translations. Adekvatni
realna hodnota and primerena hodnota, both of which might equate to „adequate value‟ and realna hodnota,
which equates to „real value‟ and is the term used in the Czech translation of the IFRS. The Amendment used
the term „realna hodnota‟
15
www.pwc.com/cz tax and legal alert, issue 22 30 September 2003, p1; Articles 19(9), 23a and 37b
16
Act, 2003 (article 23a(2)).
17
These are the very small entities
18
A separate issue which could be considered here is which version of the IFRS has superior status, the English
or Czech version? It could be that there is an error in the translation of the IFRS from English into Czech,
which leads to the preparation of Czech IFRS financial statements, which would differ, materially from the
English version of the IFRS financial statements. Is it possible to state that both the Czech and English IFRS
statements comply with IFRS?
19
January 2004
20
Discussion with member of the Commission
21
Interviewee L
22
see section on enforcement below
23
Though an issue here could also be how the IASB regards True and Fair. Indications are that true and fair is
not regarded, fully, as an override in the UK sense of True and Fair, but more as the American „Fairly State‟ i.e.
it is emphasised that the application of IFRS is presumed to achieve fair presentation. The revised IAS 1
122
requires departure from an IFRS in the „very rare circumstances where compliance would be misleading (except
where not permitted by the relevant regulatory framework) circumscribed by the financial reporting standards.‟
(IAS Plus, update January 2004, p3, available at www.iasplus.com).
24
Which commented that models that had been used to estimate fair value
25
Under Czech accounting, the completed contract method of accounting for contracts was used (see table 4),
and less information was necessary. Here we are presented with what might be called a „cultural difference‟
between how the equivalent UK and Czech accountants might approach the subject of accounting for contracts.
In the Czech Republic, there is an expectation that there will be full guidance in the law or accounting standard
concerning the detail of the management information necessary. However such guidance would not be expected
in the UK.
26
Article 37
27
Article 37(1c) and 37 (2b)
28
Which would seem to be practical, given the likely problems of ensuring that tax inspectors are trained in
IFRS, in the short term. A representative at the Union of Accountants confirmed that it was currently unclear
how the Tax Inspectorate will check the IFRS accounts.
29
An outline of the capital market in the Czech Republic, p26, accessed http://www.sec.cz, 28th
January 2004
30
Usually this consists of taking the client‟s trial balance and then converting it to IFRS and presenting all the
disclosure notes.
31
Source, Chamber of Auditors
32
Source, Ernst and Young documentation.
33
Source, ACCA in Prague, January 2004.
34
This sample was not supposed to include subsidiaries of overseas companies. It had not been clear from the
information held on the RM system that this company was a subsidiary of an overseas company.
35
See „approach to analysis‟ section above
36
Though this research is now slightly out of date, as it was based on a review of the 1999 IFRS financial
statements, which were prepared by enterprises listed on the PSE before it became obligatory to prepare IFRS
financial statements.
37
Though this is not an area we investigated, except for the one interview with a subsidiary of an overseas
company.
38
Though one firm was fined, in 2003, in relation to its audit of a leading Czech bank in 1998-1999.
(Lesenarova, 2003)
123
Část 5
Response of regulatory bodies to financial crises:
role of auditors and international comparison
[Odpověd' regulačních orgánů na finanční skandály:
role auditora a mezinárodní srovnání]
Irena Jindrichovska
In Rozvoj české společnosti v Evropské unii 2 dil.,
ed. V. Kotabova, I. Prazova & O. Schneider. Prague:
MatfyzPress. 2004. 193-198.
124
Response of regulatory bodies to financial crises:
role of auditors and international comparison
(Odpověd' regulačních orgánů na finanční skandály: role auditora
a mezinárodní srovnání)
Irena Jindrichovska
Institute of Economic Studies, Charles University Prague, CR
In the light of recent crises capital markets are re-evaluating value relevance of
financial information for price formation on the stock market. In this paper we
investigate this issue through a comparison of recent developments in the US market
as compared to Czech market.
The paper identifies specific and significant contrasts between the US and Czech
economies and draws conclusions on regulatory structures in developing as compared
to advanced economies.
Keywords: Capital Markets, enforcement, regulation
1. Introduction
In this paper we seek to utilise some of the work of Ball et al. (2003) to review the
interaction between accounting standards, preparer incentives and economic income in the
Czech Republic and the USA. In particular we focus on the regulatory and market reactions
to particular financial crises: Enron in the USA and Investicni a Postovni Banka in the Czech
Republic. (Jindrichovska et al, 2004)
2. Investicni a Postovni Banka (IPB)
IPB was established in 1990 as a then new Czechoslovak bank and it rapidly gained
market share. The bank management had two strategies from its origin – a strategy of
business development and a strategy of ownership by its own management. The first
strategy led to success in retail banking. At the same time, however, the credit expansion in
the risky environment of a transforming economy and the conflict of interest within IPB led to
a crisis. The bank acted at the same time as a creditor as well as an owner (through its
subsidiaries) of a vast industrial empire.
In 1996 the IPB had problems as an increasing proportion of bad debts in its portfolio
began to appear. IPB was under the pressure of capital inadequacy and it faced a potential
125
insolvency problem. The bank itself did not perceive any problems as it felt it had sufficient
deposits from individuals. It was alleged in the press (PBJ, 2000a) that Coopers & Lybrand,
their auditors, were requesting additional provisions to cover bad debts for the 1996 yearend.
The bank sacked Coopers & Lybrand and bought in Ernst and Young who carried out a
fresh audit for the 1996 year-end. The subsequent accounts for IPB for 1996 showed the
bank earning a profit. KPMG were also asked to tender for the same audit, but according to
one of its partners, “in the time that we were given, we refused to do it”, (PBJ, 2001).
Though the new auditor in 1997 approved the 1996 annual report, it required the bank to
increase its share capital by 11 bill. CZK. Also, one of the extraordinary audits for the year
1997 showed that the value of the bank was negative. IPB managed to hide these problems
and postpone the day when they would become apparent. However it was clear that the
majority of these problems started already in 1996. (Kudrna et al., 2002)
The Czech state sold its minority share in IPB to Nomura in 1998, hoping to have
found a strategic partner who would help to improve the capital adequacy situation of IPB by
injecting 6 bill CZK. Initially, Nomura‟s entry was perceived as a positive signal and the
bank‟s deposits increased. Nomura, however, did not act as a strategic partner. Instead, it
concentrated on selling off significant stakes in Czech industrial companies held in portfolios
of investment funds owned and managed by IPB.
The situation of the bank did not improve and it gradually started loosing credibility.
IPB got into even more dramatic problems in February 2000 when many creditors started
withdrawing short-term deposits. CNB had to impose forced administration. The situation
was finally resolved by selling IPB to CSOB (Ceskoslovenska Obchodni Banka). (CNB,
2000c).
2.1 Circular ownership
In general, investment funds that were established by banks during the voucher
privatization could not include banks in their portfolio to prevent circular ownership. The way
around it was to establish an investment management company as a “go between”. Then the
fund could own the bank, which became its “grandmother”. This was formally legal.
Incestuous ownership was particularly well developed in the IPB Group. The bank
management officially owned negligible percentage of the bank shares, but through a chain
of informal structures, relationships and option agreements it governed a substantial stake of
the bank. And gradually it became an owner de facto (beneficial owner) even if there were
different owners de iure (nominal owners). „There was evident management effort to rule the
bank through subsidiaries and other firms mutually owned by bank. The management made
126
various steps (1) to manage the ownership stakes and (2) to prevent entry of strong strategic
investor.‟ (Kudrna et al., 2002; 29)
The Czechoslovak voucher privatization was a big opportunity to increase
management‟s influence. IPB was privatized in the first wave of voucher privatization and
various investment privatization funds gained 43% of its shares. Out of this 20% was directly
owned by PIAS (Prvni investicni akciova spolecnost) controlled and managed by IPB
management. Part of the remaining 23% was controlled either by direct agreement with
other investment funds or by the purchase of shares by friendly companies. Furthermore, the
position of the management was strengthened by a gradual reduction in the state share of
ownership. The state minority stake was finally sold to a Japanese investor, who acted in
concert with IPB management. Another example of cross ownership stemmed from the fact
that, the bank acted as both creditor and owner of many industrial companies. This led to
significant conflicts of interest. The original justification for this behavior was that the bank
wanted to facilitate the reform of Czech industry and merge companies together to establish
viable groups in different sectors.
3. Reaction of regulatory bodies (CNB and the Chamber of Auditors)
Bank supervision incorporates substantial regulatory measures over the banking
sector. However, these measures are largely connected with administrative hearings
conducted through the courts of law. When the central bank wants to try a bank or its
shareholders the CNB has to present sufficiently conclusive proofs. On the other hand, the
Bank Supervision department has rather limited possibilities to react in a situation where the
bank management systematically distorts the information about economic results and its
auditor signs such distorted reports. (CNB, 2000c) During its final days the IPB obstructed
the control procedure. This did not allow the CNB to identify early and conclusively the real
state of the bank. Furthermore, the CNB had to act very cautiously to avoid damage to IPB,
as any decision could lead to irrevocable changes on the inter-bank market. Moreover, with
regard to the position of IPB, the bank‟s failure would lead not only to instability in the
banking sector but it would also impact on the economy as a whole.
A Parliamentary Committee criticised the slow reaction of CNB to internal
developments in IPB disregarding the fact that IPB continually misinformed the CNB by
providing information that did not provide a true and fair view on its financial position. CNB
held the opinion that IPB had failed to fulfil its reporting obligations and that it was the fault of
IPB representatives who were responsible for deferring any effective solution. Even the
127
auditors did not signal the true situation of the bank. For this reason the CNB started legal
proceedings against IPB management. To cope with a problem of such size and complexity
was not within CNB capabilities, as it also concerned many industrial companies through
inter-connected ownership. The CNB therefore asked the state for support. (CNB, 2000b)
After the collapse of IPB, the Chamber of Auditors also investigated the role of the
auditors in their audit of IPB. „It took them almost three years…and finally..the Czech
Chamber‟s investigation resulted in a $25,000 fine for Ernst and Young, which has since
replaced its management and drawn closer to the international auditing firm.‟ (PBJ, 2003)
4. The Enron scandal
In October 2001, Enron revealed that there were some problems with the figures it
had reported in its financial statements. Over the next few months these problems
unravelled to reveal one of the largest cases of financial misreporting for some years. By
July 2002, when the USA had enacted new laws to deal with the perceived inadequacies of
the preparation and enforcement of financial reports, Enron‟s auditors, Arthur Andersen had
also imploded. Though it is difficult to know as yet the full extent to which the auditors of
Enron up to 2001 had misled the investing public the damage that was caused to the
reputation of Arthur Andersen, though admissions that they had „shredded documents‟
amongst other issues, led to their demise as an international audit firm. (Michaels, 2002)
Public opinion polls in the USA showed rising concern about the quality of financial
information. In March 2002 a third of respondents thought the problem was confined to a few
isolated cases. By June, at least two thirds of them thought most companies might hide
shadow operations. However corporate ethics hardly figures in the mid term elections yet.
The American Enterprise Institute argued there had been no break in the overall pattern of
attitudes to business as a whole, although the numbers are showing a decline in confidence
but no backlash against business as such. The bubble is deflating rather than bursting. The
scandals at Enron, WorldCom and the rest may simply be too recent to have had their full
effect. (The Economist, July 6, 2002)
Most of the misbehaviour seems to be related in some way to the huge incentives
Wall Street provided in the past decade for reporting rising earnings quarter after quarter, or
rather, earnings that exceeded ever rising expectations. The half dozen firms, which
responded to this incentive “improving” their accounting results, cannot be the only ones.
The likelihood is that many more will emerge.
128
5. The US regulatory response
The US regulatory response to the Enron crisis1
was to enact the Sarbanes Oxley
(SBA) legislation in July 2002. The SBA provided for the setting up of a Public Company
Accounting Oversight Board (board). All accounting firms that prepare audit reports for
issuers in the USA are to register with this board. The board also has responsibility for
establishing rules on auditor independence and any other standards relating to audit work,
inspections of all audit firms and conducting investigations and disciplinary proceedings
where appropriate against accounting firms (Hermsen et al., 2002). There are also
requirements concerning listed company audit committees, the provision of non-audit
services by auditors (detailed prohibitions on the provision by auditors of particular services
such as valuation services and financial systems design), declarations by directors and
executive officers and expanded information in the filing form for companies (the form 20F).
For example, the chief executive officer and chief financial officer must certify that the form
20-F is materially accurate and complete and, amongst other aspects, that the information in
the report „fairly presents, in all material respects, the financial conditions and results of the
operations of the issuer‟ (Hermsen et al., 2002:111). There are also increased penalties for
accounting irregularities and financial fraud. As stated, „the legislation...creates a new
oversight board for the accounting industry, until now a largely self-regulated profession.‟
After a „succession of scandals from Enron to WorldCom Accountants now face a tougher
regime. Maximum jail time for executives who commit fraud is quadrupled to 20 years and a
new crime of securities fraud will have a maximum sentence of 25 years.‟ (Guardian
Newspapers, 2002). Over the last 18 months, these parts of the SBA have come into force.
The crises with Enron, and subsequently WorldCom, in the USA led to a market
response, with a decline in share prices and the demise of Arthur Andersen, and a
regulatory response, the SBA. These were all on a much larger scale than the responses in
the Czech Republic.
6. Monitoring and sanctions in the Czech Republic
Currently, monitoring of financial statements is carried out by three organisations: the
Czech Trade registry, the Czech Securities Commission and the Czech National Bank.
There is a requirement to file audited financial statements with the Czech Trade
Registry. However, as has been noted in previous research and reports (Sucher and
1
And the issues that arose with financial reporting at WorldCom, where it was revealed that worthless
WorldCom had “improved” its financial results by 3.9 billion dollars through misrepresenting revenue and
assets (Economist, 2002, p.45).
129
Zelenka, 1998; ROSC, 2003) though this is a requirement in many cases it is not observed.
In those financial statements, which were filed, some substantial non-compliance with Czech
regulations has also been observed (ROSC, 2003).
The current enforcement of compliance with financial reporting standards, noted in
the Accounting Act, 2001 article 37, was dealt with by the tax inspectorate. In the Act, 2003,
it is now stated that penalties will be levied on accounting units, which are obliged to prepare
IFRS financial reports, and do not do so,2
and the Tax Inspectorate should collect these
penalties. The penalties for non-compliance with financial reporting regulations had been
increased to a maximum of 6% of balance sheet gross assets.
The Czech Securities Commission also reviews the financial statements of listed
companies for completeness and obvious errors. On its website it carries all the latest
publications from the Centre for European Securities Regulation (CESR) which suggest how
IFRS compliance should be enforced. On the website, there is a comment that,
„The Commission, together with the Ministry of Finance and the Czech National Bank, is
creating conditions for the performance of supervisory activities over the so-called
consolidated financial groups.‟
As noted in the FEE survey referred to above, „The Czech National Bank enforces
the rules for banks and provides a specific review of certain bank operations in addition to
the common mechanism applied in the case of other enterprises‟ (FEE, 2001). However, it is
not yet clear what these conditions (referred to on the website above) might be. The
Commission has certainly not had the technical resources to enforce accounting standards
until now (ROSC, 2003). Therefore a large burden of the enforcement of compliance falls on
the auditors.
7. Conclusion and directions for further research
The regulatory and market responses to the crisis with IPB contrast sharply with
those to the crisis in the USA with Enron. In the Czech Republic, there has been no major
overhaul of either the financial reporting or audit enforcement regime, except in so far as it
related to the conditions for joining the European Union. The audit firm concerned has not
collapsed and disappeared. Indeed, ironically, it has continued to grow and has locally
absorbed the audit firm, Arthur Andersen that did collapse following the Enron crisis in the
USA. One could argue that this is a „bad thing‟, however, on the other hand, it could indicate
that expectations for high quality financial reporting and enforcement through audit, were
2
article 37(1c) and 37 (2b)
130
never that high in the Czech Republic, compared to the USA, and investors had other means
to assess the worth (or lack of it) of enterprises. The lack of a properly functioning legal
system, capable of interpreting and enforcing the law is particularly important here.
In the Czech Republic, this latter presumption has particular problems. As our
research on the movement in share prices and the reaction of accounting income to changes
in economic income indicate, there are some odd findings which may indicate that
movements in share prices may not reflect those that would arise in a semi-strong efficient
stock market.
This in part might flow from the particular institutional context in the Czech Republic –
which reflects some of the circularity in the arguments of Ball et al., (2003). Where there may
be extensive cross holdings of companies, and it is perceived that enforcement of
compliance with financial reporting standards does not work (Sucher and Kosmala
MacLullich, forthcoming) then investors may use alternative, informal, insider information to
establish what is actually happening with particular enterprises. The response to the crisis
over IPB is an example of this. There was ample coverage in the local business press of
developments at IPB, yet, unlike with Enron, there was no immediate regulatory response for
reasons outlined above. It is likely that responses were informal and insider oriented.
References:
Ball, R. Robin, A. and Wu, J.S.: Incentives versus standards: properties of accounting income in four
East Asian countries, Journal of Accounting and Economics, Vol. 36, 2003, pp 235-270.
The Backlash Against Business, The Economist, July 6
th
2002, p 45-46.
Bush Signs Corporate Fraud Bill, Guardian Newspapers, 30.7.2002, accessed via the web
http://www.guardiannewsngr.com/
CNB: Report for the negotiation of the Czech Parliament on the development and solution of the
situation of IPB, www.cnb.cz/pdf/zprava_ipb.pdf [in Czech], 2000a.
CNB: Opinion on the Report of the Investigative Committee of the Czech Parliament on issues of IPB,
www.cnb.cz/pdf/ipb_reakcecnb.pdf [in Czech], 2000b.
CNB: Enclosure to the Report for the negotiation of the Czech Parliament on the development and
solution of the IPB situation, www.cnb.cz/pdf/zprava_ipb-otazky.pdf [in Czech], 2002c.
FEE: Enforcement mechanisms in Europe, FEE, April 2001.
Hermsen, M.L., Niehoff, P. and Uhrynuk, M.R.: An extraordinary expansion, Accountancy, UK,
October 2002, pp 110-112.
Holthausen, R.W.: Testing the relative power of accounting standards versus incentives and other
institutional features to influence the outcome of financial reporting in an international setting, Journal
of Accounting and Economics, Vol. 36, 2003, pp 271-283.
Jindrichovska I., Rhys, H. and Sucher, P.: Capital Markets, Accounting and Audit: Implication of
Recent Reporting Crises. Conference paper for 27 EAA Congress, Prague, 2004.
Kudrna, Z. et al.: Rise and fall of Investicni a Postovni Banka, Case Study, IES FSV UK [in Czech],
2002.
Michaels, A.: Andersen faces Enron documents claim, Financial Times, January 14
th
, 2002, p 22.
131
Prague Business Journal: Ernst & Young‟s local operations are folded into regional audit efforts, PBJ,
May 22 2000.
Prague Business Journal, Ernst & Young fine too small by billions, Mlynar‟s joy too unreserved, PBJ,
June 9, 2003.
ROSC: Report on the Observance of Standards and Codes: Czech Republic, World Bank, July 17,
2003.
Sucher, P. and Zelenka, I.: The development of auditing in the Czech Republic, European Accounting
Review, Vol.7 No.4, 1998, pp 723-741.
Sucher, P. and Kosmala MacLullich, K. (forthcoming): A construction of auditor independence in the
Czech Republic: Local Insights, Accounting, Auditing and Accountability Journal.
132
Odpověd' regulačních orgánů na finanční skandály: role auditora a
mezinárodní srovnání
Irena Jindřichovská
Institut ekonomických studií, Karlova Universita Praha, ČR
Ve světle finančních krizí poslední doby přehodnocují kapitálové trhy relevantnost
finančních informací pro tvorbu cen na burzách. V tomto článku se zabýváme
uvedenou otázkou prostřednictnvím porovnání nedávného vývoje ve Spojených
státech a v České republice. Článek identifikuje specifické rozdíly mezi způsobem
regulace ve Spojených státech a v České republice a srovnává přístup finančních
autorit k finančním skandálům v rozvynutých ekonomikách a v rozvíjející se České
republice.
Klíčová slova: kapitálové trhy, prosazování pravidel, regulace
Shrnutí a závěr
Odpovědi trhu a reguačních orgánů na krizi v české IPB jsou v ostrém kontrastu s
reakcemi na krizi Enronu v USA. V České republice nedošlo k žádné zásadní změně
finančního reportingu nebo auditorských postupů mimo těch, které se týkají přistoupení
země k Evropské unii. Auditorské firmy, kterých se český skandál týkal nezbankrotovaly a
nezmizely. Naopak rostly a lokálně absorbovaly auditorskou firmu Arthur Andersen, která
zbankrotovala po krizi Enronu v USA. Pokud je cílem České republiky vysoká kvalita
reportingu je to rozhodně nepříznivý vývoj. Ale na druhou stranu se v České republice nikdy
neočekávalo, že audit a finanční reporting budou pro investory tak důležité.
V Čechách a jiných podobných ekonomikách používají investoři jiné prostředky a jiné
cesty k informacím, které jim dovolí ohodnotit hodnotu firem. To je při porovnání se
Spojenými státy v naprostém kontrastu. Nedostatky ve správně fungujícím právním systému,
který je schopen interpretovat a posilovat právo je zde velmi významný.
Fungující právní a institucionální systém, který by garantoval, že akcionáři dostanou
všechny potřebné informace, je v České republice zvláště závažným problémem. Jak náš
výzkum pohybu cen akcií a reakce účetních zisků naznačil, můžeme narazit na některé
zvláštnosti, které ukazují, že pohyb cen akcií nemusí nutně odrážet to, co bychom očekávali
podle hypotézy středně-efektivního kapitálového trhu.
Tento jev nejspíše vyplývá z konkrétního institucionálního kontextu v České
republice, který umožňuje existenci kruhového vlastnictví podle definice Balla et al. (2003).
Zdá se, že v ekonomikách, kde jsou četné křížové vlastnické vazby mezi společnosti nelze
jednoduše prosazovat dodržování pravidel finančního vykazování (Sucher and Kosmala
MacLullich, forthcoming) investoři potom používají alternativní neformální vnitřní informace,
aby si udělali obrázek o tom, co se s konkrétní společností děje. Odpověď na krizi IPB je
toho příkladem. V tisku byly tyto události hojně konmentovány, avšak na rozdíl od případu
Enron, nedošlo k okamžité regulatorní odezvě. Je pravděpodobné, že odpovědi byly
neformální a orientované dovnitř společnosti.
133