172  Marketization in a federal System
Linder, W. (2004) Schweizerische Demokratie. Institutionen - Prozesse - Perspektiven (Bern: Haupt).
Magnin, C. (2004) 'Ein für den aktivierenden Sozialstaat typisches Handlungsdilemma', Swiss Journal ofSociology, 30/3, 339-61.
Obinger, H. (1998) Politische Institutionen und Sozialpolitik in der Schweiz. Der Einfluss von Nebenregierungen auf die Struktur und Entwicklungsdynamik des schweizerischen Sozialstaats (Frankfurt am Main: Feter Lang).
Perret, V., (). Giraud, M. I lelbling and M. Battaglini (2007) Les cantons Suisses face au chömage. Federalisme etpolitiques de l'emploi (Paris: L'Harmattan).
Pfister, N. (2007) 'Sie springen ein und fangen auf, ZeSo Zeitschrift for Sozialhilfe, 104/2, 12-3.
IMister, N. (2009) Integrationsauftrag der Sozialhilfe in der Praxis (Bern: SKOS). Piller, (). (2006) Die soziale Schweiz. Die schweizerischen Sozialwerke im Überblick (Bern: Haupt).
Piotet, G. (2006) 'Regionnlisierung im Kanton Waadt. Fine Anlaufstelle, viele Dienstleistungen', ZeSo Zeitschrift for Sozialhilfe, 103/1, 12.
Ragni, f. (2007) Die Wirksamkeit der öffentlichen Arbeitsvermittlung in der Schwei/.. Übersicht über die Hauptergebnisse des 'Follow up' der Evaluation der aktiven Abeitsmarktpolitik und Beurteilung der sechs Abschlussberichte, Discussion paper of che Department for Economic Policy of SECO, http://www.seco.admin.ch (accessed 25 April 2008).
Sager, F. and C. Rüefli (2005) 'Die Evaluation öffentlicher Politiken mit föderalistischen Vollzugsarrangements. Eine konzeptionelle Erweiterung des Stufenmodells und eine praktische Anwendung', Swiss Political Science Review, 11/2,101-29.
SFGO (2003) Hauptsache Arbeit, die schweizerische Arbeitsmarktbehörde, http://
www.trelfpunkt-arbeit.ch (accessed 15 April 2008). SECO (2008a) Die Lage auf dem Arbeitsmarkt im März 2008, http://www.seco.
admin.ch (accessed 10 Mai 2008). SECO (2008b) Über uns, http://www.treffpunkt-arbeit.ch (accessed 10 Mai 2008). SKOS (2005a) Richtlinien für die Ausgestaltung und Bemessung der Sozialhilfe (Bern:
Schweizerische Konferenz für Sozialhilfe).' SKOS (2005b) SKOS-Richtlinien-Revision 2005, Grundzüge, http://www.skos.ch/
Dokumente (accessed 20 April 2008). SKOS (2006) Regionalisierung und Professionalisierung sozialer Dienste. Auf dem
Weg zu einer tragfähigen und wirksamen Sozialhilfe, http://www.skos.ch (accessed
20 April 2008).
SKOS (2007) Häufig gestellte Fragen zur Sozialhilfe, http://www.skos.ch (accessed 20 April 2008).
Stremlow, J. (2008) 'Wie die Behörden organisiert sind: Ein Überblick', ZeSo Zeitschrift für Sozialhilfe, 105/2, 4-5.
Zürcher, E. (2002) 'Bildung einer Nationalen Koordinationsgruppe zur interinstitutionellen Kooperation (KG HZ)', Soziale Sicherheit CHSS, 4, 210-1.
9
Governance of Activation Policies in the Czech Republic: Uncoordinated Transformation
Tomáš Sirovátka and Jiří Winkler
Introduction
Activation policies represent one of the key common trends in the development of the welfare states in Europe (compare Van Berkel and M0ller, 2002; Serrano Pascual, 2004; Zeitlin and Trubek, 2005), no matter that individual national activation strategies are far from being uniform. Policies aiming to increase labour-market participation seem to be appropriate for achieving both the goal of sustainable public finance and that of eradication of poverty and social exclusion. In post-communist countries the policies of activation are being implemented in a specific societal and institutional context: certain failures of governance and implementation conditions have already been identified (Winkler and Zizlavsky, 2004; Sirovatka, 2007, 2008; Sirovatka et al., 2007). In this chapter we examine the question as to what specific kind of governance of activation policies have emerged in the Czech Republic and, secondly, the various effects of the specific mode of governance of activation policies. The chapter is structured as follows: after the introductory part we characterize the institutional framework of activation. In the third section we explain the key trends of governance of activation policies since 2004 in the Czech Republic. In the fourth part we assess the effects of governance of activation policies, and in the last and concluding section we discuss the main findings.
Institutional framework of activation
The formation of labour-market and employment policy institutions in the Czech Republic goes back to the beginning of the 1990s when the
173
174   Governance of Activation Policies in the Czech Republic
governance structure of the Public Employment Services (PES) was being shaped. This structure survived with no important changes until 2004. During this period, the establishment of local (district) Employment Offices (EOs), with relatively tough powers to implement active policies, took place within a model of a two-tier PES: this included the PES section at the Ministry of Labour and Social Affairs (that is, Employment Services Administration (ESA)) in the centre (strategic and coordination level) and 77 EOs in the districts, originally District Employment Offices, at the second level (executive level).
The ESA as an integral section of the Ministry of Labour and Social Affairs is responsible for the coordination and delivery of active employment policy and unemployment benefits - these tasks are performed by local Employment Offices. After the split of the Czech/Slovak federation in 199;-?, the power to create employment policy was conferred on the national ministry (more specifically on the ESA), but without transfer of the necessary personnel and financial capacity that previously existed at the federal ministry. Since its establishment in 1991 and during its whole existence, the PES has thus been understaffed and underfinanced when compared with other European Union (EU) countries. This fact is mirrored first of all in the poor managerial capacity of the centre (ESA), as well as in the work overload at local/district EOs. The above-mentioned deficits have translated themselves into a limited scope and quality of active labour-market policies (ALMPs) and job mediation. For example, if we compare the International Labour Organization (1LO) figures (2003) for the number of the unemployed with the total number of PES staff, we see that the workload (in numbers) is about three times as high in the Czech Republic when compared with Germany and Sweden, and even higher when compared with the UK.
The EOs at the local level are responsible for delivery of unemployment benefits as well as for the implementation of ALMPs. They are guided by legislation (Employment Act) and by decisions of the PES governing body - ESA. This central body outlines the strategy of labour-market policy (embedded since 1999 in the National Action Plans of Employment (NAPE)). The decisions of ESA are directed mainly towards the distribution of financial resources allocated for ALMP measures to the EOs, based on specified criteria (such as the unemployment rate in the region, the proportion of long-term unemployment) and towards recommendations (methodical notes) on ALMP measures delivery. Secondly, ESA exerts influence on the formation of the Employment Act. Notwithstanding the conditions given by legislation and the financial resources allocated by the Ministry of Labour and Social Affairs
TowአSiravátka and Jiří Winkler   1 75
listrict EOs are largely independent in making decisions ;,;mi: the SCope and kind of measures to be realized, about the groups of n,P unemployed to be involved in the measures, depending on the local abour-market conditions, as well as about the specific approach to the unemployed and the job-mediation procedures.
TP high degree of autonomy enjoyed by the EOs is not, however, a i* rtf a ourooseful delegation of power but rather an unintentional result of a I institutional frame. While the central ESA pos-
outcome ^ [mte6 capacity for conceptual decision making, the EOs sesses o  y i(Jerable knowledge of the local labour market. The
usually nav ^^.^ functi0ns of the PES centre (ESA) are therefore control an ft does not poSsess much authority and influ-
significan y ^ ^ planning and formation of activation policy ence, eit ie nitoring an(j overseeing implementation. On the other or during ^ nQt influence the formation of national employment hand loca  ^ contrast to Germany, Austria and other countries,
policy m any    Z'onaj self-Government Units were established in the When the   eg undeI lhe Public Administration Reform initi-
Czech ^public m       »       ssion to the EU, 14 of the local (district)
j finp to tne coiii11'/ a ted out: i*.       ■ extended competences as partners to the new
kOs were e"tr"jmil^stration Units. Nevertheless, these entrusted EOs Regional Sell- c i deciSion-making powers: their task is mainly
were grantee ve*7 tive proCedures necessary for implementation of to ensure a<^^al Fund (ESP) projects/measures. They also fulfil some the European o ■ coordination activities; however, they are not pro-methodologica a making discretion related to the national ALMPs. vided with deciS 0• - ^ ^ National Tripartite Council, the Parlia-Other actors, _ ^ Health Care and Social Issues or the inter-mentary CoEaS°^^aD for the NAPE (or later, the National Reform ministerial comm ^ adviSory role in the field of ALMP. Since Programme)/ »u directors of local EOs have consulted with the the early 1990s, Coinmittees (advisory bodies established at local so-called Counse & partners, NGOs, municipalities and educational level) where loca ^ ThesG; however, are not endowed with any decision-maKm» f , t;o
t the local level, EOs represent the principals in the held of acti Municipalities may be considered passive actors: they del v Vatl0n- Tl assfstance benefits that are fully refunded tr0lll \^
cen-
*.ut social a»oio,,i-w , ■•"in rne
Government budget. They do not receive any ftmds for rral S°yen       b dv possessing resources and competences rv, ' since the only^fJnlemenMon of ALMP is the PES ELS?0** ing activatioi
, and implementation ot aliw ,s the PES. The EOs can
176   Governance of Activation Policies in the Czech Republic
provide from their budget wage subsidies for public works organized by municipalities, but municipalities themselves have no economic incentives available for any activation measures. Cross-sectional coordination between EOs (subordinated to the ESA as a section of the Ministry of Labour and Social Affairs) and municipalities (subordinated to the Ministry of Regional Development) in the field of activation is limited to the organization of public works and unsystematic exchange of information.
In spite of the increase in unemployment and in the numbers of welfare state clients during the late 1990s, the government did not allow an increase in public administration staff, with no exemption for PES. Instead, the government established following elections in 2006 promised to decrease public administration staff by 9 per cent, and this objective was also applied to PES that are considered to be part of public administration. A similar practice was adopted following the crisis in 2008-9, when further reductions in public servant numbers were announced in spite of the fact that unemployment exploded during 2009 and that Individual Action Plans (lAPs) that had been used to a limited extent (see below) since 2004 became obligatory, which significantly increased the workload for EOs. Employment Offices had to reduce staff, and evidence shows that there are now only around 1800 staff working directly with the clients as mediators or counsellors (Kaluzna, 2008), while there were 540,000 unemployed people at the end of 2009 (that is, a ratio of about 300 clients to one frontline staff).
Trends in governance of activation since accession to the EU
The Czech labour market was not hit by unemployment increases until the end of the 1990s (registered unemployment level came close to 10 per cent), which were, however, marked by a high proportion of long-term unemployment (more than 50 per cent of unemployment in 2004) and sharp differences in specific unemployment rates according to education, health status, age and gender. Activation became a challenge. This period overlaps with the country's accession to the EU: minor institutional reorganization took place in order that the ESF projects ! could be implemented. In principle, the governance framework did not change. The lack of top-down governance reforms complicated - as ,we will argue - effective implementation of activation policies in several respects. Nevertheless, this fact does not imply that no changes in governance appeared. We have evidence of several transformations that
TomdS Sirovdtka and Jifi Winkler 177
occurred as spontaneous, uncoordinated adaptations to the demands raised by a more intensive implementation of activation measures within the regime of the ESF schemes/projects. The co-existence of weak coordination by the centre and spontaneous adaptations of governance at the local and regional levels may be considered to be a specific feature of Czech governance of activation policies.
When assessing the trends in governance of activation policies in the Czech Republic since accession to the EU (2004) we focus primarily on the key dimensions of governance of activation as outlined in Chapter 1: decentralization, marketization, inter-agency cooperation, new public management and coordination by objectives, with the most important aspects being decentralization and marketization (for their characteristics, see Table 9.1).
Decentralization and inter-agency cooperation
Decentralization represents the first and most significant (spontaneous) trend of the governance changes, although the system has remained highly centralized in terms of financing the ALMP measures. While administrative reforms of PES in many European countries (Austria, Germany, Denmark and Ireland) involved a formally governed functional and territorial decentralization and devolution of some services, decentralization in the system of Czech employment services takes place as an unexpected consequence of governance failure. We may label the trend as a spontaneous transformation of governance.
The transformation of governance was necessitated by the appearance of new measures financed through the ESF. However, this reform was little coordinated by the centre (ESA) as a strategic change, except for formal steps such as the establishment of an administrative unit at the central level responsible for administration of the implementation of the ESF projects (see below). The spontaneous transformation at the local and regional levels followed as a consequence of the complex process of social learning, consisting of several parallel processes, such as rational problem solving, normative (legislative) transformation, imitation and ideational transfer (March and Olsen, 1989; Obinger et al., 2005; Taylor-Gooby and Zinn, 2005).
At the same time, the availability of new resources of the ESF and opportunities for participation of various actors confronted central authorities of PES/ESA with a serious challenge concerning their coordinating ability, and made their managerial and capacity deficits more apparent. This was a catalyst to further decentralization (and devolution) rather than centralization.
178  Governance of Activation Policies in the Ciech Republic
Table 9.1   Trends in governance of activation after accession to the EU (2004)
Decentralization     Spontaneous decentralization
Political-administrative
Non-competitive
Internal
De-concentration
Marketi/.ation Purchasers
Providers
Contracts
Poor capacities of the centre to coordinate the policies (legal instruments and central budget being the main tools)
Employment offices continue to maintain a high level of discretion (political decision-making discretion) while being restricted by central budget constraints
Municipalities continue to be responsible for administration of social assistance (administrative decentralization)
Purchase of training, individual diagnostics, counselling, private mediation - increasing in scope due to availability of ESI: resources and implementation of agency work in 2004 (new Employment Act), low numbers of com petitors.
Decision-making power given to local EOs (implicitly due to outsourcing in conditions of weak governance, most decision power flows to the providers) - devolution.
Entrusted EOs are responsible for coordination of methodical functions, control and assessment of activation and, more recently, for implementation of regional projects under the ESF
Cooperation among local EOs through informal networks Unregulated market failures
Public EOs (with no use of vouchers for services, no role played by the unemployed themselves)
Public (educational institutions), private (increasing role), NCOs
Classic but weak regulation: soft conditions - lack of incentives in terms of performance targets due to the weak role of service purchasers
No duty to outsource (however, due to a lack of PES capacities, outsourcing implicitly becomes a dominant strategy)
Due to spontaneous decentralization even prior to 2004, EOs had spontaneously developed an adaptation mechanism for the governance of problems emerging from the deficits on the part of ESA (Sirovatka, 2003; Winkler and Zizlavsky, 2004). In conditions of weak central coordination, directors of EOs had begun to create informal regional support networks among themselves, coordinating procedures to resolve new
TomdS Sirovdtka and )ui Winkler  179
problems, issues concerning competence or vaguely defined tasks. These clusters of directors and clusters of other staff of EOs occupying similar positions now cooperate during preparation of calls under the ESF project schemes. They also offer consultations to potential recipients/coordinators of the ESF projects concerning preparation of their individual project proposals.
In a number of regions, EOs became informal coordinators of collaboration between EOs, employers, NGOs and municipalities during the formation of local programmes of social and economic development. This bottom-up activity was later institutionalized by the Employment Act from 2004 as a specific labour-market programme within what PES called Purpose-built Programmes in the Local Labour Markets (Winkler et al., 2005). Consequently, and especially after 2004 (in connection with implementation of the ESF projects), public-private partnerships (PPPs) in the regional labour markets have developed, emerging in two forms: first, partnerships associated with local and regional programmes, and secondly, those associated with national programmes and serving to stabilize the national strategy in regional conditions. The central role of local EOs (and the entrusted local EOs) as principals is typical of the latter type of PPPs. Their role is to organize independent tenders announced ad hoc under specific calls for projects, and subsequently to contract the services providers.
The PPPs later merged with the first type of the ESF projects, called the Regional Integrated Projects, which are created from the bottom up through cooperation among local EOs coordinated by the EOs entrusted with methodological guidance. Due to this spontaneous and institutionally supported decentralization and networking, the performance of PES is characterized by a high degree of decentralization, even fragmentation. It is often the case that the EOs do not follow the government policy guidelines given by the ESA, with the centre thus losing control over implementation of activation programmes and their outcomes.
Secondly, the formation of the state employment policy and the labour-market policy is strongly politicized. In such a situation the ESA faces difficulties in defining any widely acceptable government policy objectives. Lastly, the incorporation of other private participants (each having their own specific interests) in the implementation structure increases an already high dispersal and heterogeneity of decision making and causes chaos in policy making at the local level, and misunderstandings in communication between the centre and the regions/districts. Similarly, the cooperation between state and non-state
180   Governance of Activation Policies in the Czech Republic
participants does not proceed on the basis of consensus over activation policy objectives in the labour market.
The fulfilment of the principal role of the EOs in this process of implementation of activation is an uneasy task. The staff of EOs often settle for less than completely satisfactory results in cooperation with external service providers who do not provide better measures/services (regarding their scope and quality) than the EOs, although the cost of the outsourced services is considerably higher. At the same time, they emphasize the limited capacity of local EOs to coordinate and control implementation of activation by private agencies and NGOs, as well as the poor communication with the responsible coordination units for the ESF projects at the entrusted EOs (Winkler et al., 2005; Winkler, 2009).
On the other hand, innovative bottom-up initiatives and cooperation sometimes emerge to substitute national programmes, the capacity of which appears insufficient to solve local problems. These are usually ad hoc regional networks of actors with a principal role played by the local EO. Subjects involved in such networks have a strong social capital (trust, in the first place) built during cooperation on projects at the regional level. Their cooperation is associated with the possibility of identifying a common regional interest that inspires joint action by different subjects. Cooperation among EOs and municipalities on public works projects in underdeveloped regions (border regions, inner peripheries) is one example. Existing research shows a high degree of satisfaction of the participants in such cooperation with their ability to find a common platform. This finding is even more relevant as the participants report that controversies were more common ten years earlier -when they started cooperation on these projects (Winkler, 2009).
Marketization
Operation of private employment agencies has been possible in the Czech Republic since the early 1990s. Employment Offices have also contracted out retraining programmes to educational institutions, including private ones. In addition, the new Employment Act (2004) made agency employment possible, and subsequently the number of private actors increased rapidly. Their activities further expanded as a result of the ESF projects enabling EOs to subcontract a range of various activities, such as job mediation, individual diagnostics, counselling, vocational training, job placement of vulnerable groups of the unemployed, and individual assistance. The ESA as a central body has been outsourcing activities such as analytical and conceptual work, special educational programmes for PES staff, marketing, tenders and
Tomáš Sirovatka and Jiří Winkler 181
so on. The range and scope of activities being outsourced has increased dramatically.
There are two kinds of the ESF project schemes: (1) 'national' projects, where the entrusted EOs play the role of project coordinator and services purchaser, and private and public agencies as well as NGOs as services providers; and (2) 'individual grant' projects, where the entrusted EOs take on the role of tender managers, and private and public agencies and NGOs act as project coordinators and may purchase those services that they are unable to provide by themselves from other providers. Between 2004 and 2008 there occurred a marked increase in the numbers of non-governmental (profit and non-profit-making) organizations providing educational, counselling and mediation services in the labour market. Strong, long-standing agencies operating in this market have gained a dominant position in several economically significant regions. This is well documented in the attitudes of EOs' managers who evaluate best those agencies with which they have cooperated for a long time (Winkler and Zelenkova, 2010). Surprisingly, such a development has not caused any increase in competition between service providers. The given circumstances prevent the emergence of competition. According to EOs' staff (interviews led by the authors), the established leading agencies misuse their market position by working together in order to divide the market amongst themselves and do not allow competition to develop with other educational and mediating agencies. In the case of the ESF projects, the performance of service providers is supervised by PES on the basis of obligatory financial and output indicators. However, these output indicators seem to be rather soft and vague and thus do provide the services providers with a lot of space for manoeuvre (see below). For example, indicators of quality are missing from evaluations of programmes provided to vulnerable groups of the unemployed. Similarly, the selection criteria for evaluating project proposals do not include indicators of quality.
Failed attempts at new public management and lack of coordination
The Czecfi Republic has witnessed two processes in connection with new methods of governance since 2004.
First, an attempt to move towards a management-by-objectives approach in ALMP has taken place, as expressed in the continued specification of the NAPE guidelines and targets. In 2004, the ESA formulated 'A plan of institutional, merit and time-related preconditions for the realization of the NAPE measures in 2005-2006'. This plan introduced
182   Governance of Activation Policies in the C/.ech Republic
quantified targets: it specified the number of Individual Action Plans to be realized (for both young and long-term unemployed people) and the number of long-term unemployed and other disadvantaged people qualified to participate in the ALMP measures. Additionally, since 2005 PES have obliged EOs to meet certain quantified targets in the form of ALMP targeting, and have also specified the targets concerning the application of selected tools (vocational training and work rehabilitation for disabled people). However, the targets set for 2006 and 2007 lacked ambition, since the required proportion of participants from the target groups did not reach their designated proportion in the unemployment stock: for example, participation of 30 per cent of people unemployed for more than six months in ALMP measures was required (while there were twice as many of those in the registers), as well as 20 per cent of unemployed people above 50 years of age (30 per cent in the registers) and 25 per cent of unemployed people under the age of 25 years of age (which slightly exceeded their proportion among the unemployed). So it comes as no surprise that these goals had already been fulfilled at the majority of the EOs. Another issue is that EOs have never been evaluated with regard to the achievement of the set objectives (see MPSV, 2008). We can characterize these developments as unfinished, incremental reforms of the governance of activation. The intended shift towards the managerial approach has not, as yet, significantly affected the prevailing bureaucratic-administrative style of policy making, given the weak conceptual and control capacity of the centre: consequently, the set goals have been neither ambitious nor strictly required and evaluated.
Secondly, the ESF projects also induced organizational changes within PES. In a relatively short time a plural implementation structure was created. The administrative agenda of the ESF schemes was handed over to the newly established ministry sections. A multi-track administration emerged, one track being the 'ESF line' and another being the 'national ALMP line', with little coordination between them. This lack of coordination is apparent in particular at the level of local EOs that are responsible for the 'national' ALMP measures and are also involved in the roles of tender managers, purchasers or project coordinators in the ESF projects.
The current stage of development of Czech activation policy and its governance is seen as a result of historically developed national rules and routine procedures: a low level of internal feedback/self-retlection and an inability of the system of employment policy to reform itself on the one hand, and the influence of the European Employment Strategy,
TomäS Sirovdtka and fin Winkler  183
mostly implemented by means of the Open Method of Coordination (OMC), as well as programming and funding from the ESF, on the other hand. Owing to mutual interaction of the factors mentioned above, the present governance system shows a number of deficiencies and compromises (institutional and organizational gaps) that may affect the efficiency of Czech activation and labour market policy (March, 1999). Given this situation, the central authorities have attempted to reform this malfunctioning system, although they prefer bureaucratic methods to a formal centralization of powers.
The emerging mode of governance
Considirie (2001) distinguished between procedural, market, corporate and network types of governance. Activation processes in practice always includes a mixture of various procedures and means of organizing activation policy represented by these models (see Chapter 1). The modes of governance are rather the ideal type, that is, tools enabling empirical measurement and descriptions. Hence it is no surprise that the current Czech mode of activation policies delivery has hardly been converging towards any of the ideal types. It is not fully compliant with the rules of either the market or corporate governance mode, since the ESA has not been successful in coordinating employment services by using operationalized objectives. In addition, as yet there exists no competitive market of non-state services, and rational client demand for such services is still lacking. Neither is the mode of coordination close to the rules of network governance, which would be underpinned by mutual dialogue among the various interested parties (entrepreneurs, regional political representatives, employees and jobseekers, and others) and by the citizens' and service providers' trust of the coordinative role of the state - for example, in the case of cross-sectional partnerships. Such a role of the principle is not appropriately fulfilled by FES. Finally, the current situation is far from the procedural governance mode where providers of PES perform under a hierarchical control by the policy makers.
We assess Czech governance of activation as a hybrid mode, positioned somewhere between the procedural mode (which is strongly embedded due to path dependency, but is characterized by the insufficient managerial and implementation capacity of the ESA, and therefore spontaneously decentralized) and the market mode of governance (based on widespread contraclualization and outsourcing, and characterized by weak regulation of the market and a lack of incentives - vague contracts are typical). This mode of governance is, however, enriched
184   Governance of Activation Policies in the Czech Republic
with fragmented, unfinished and inconsistent innovations evoked primarily by external stimuli after the accession to the EU (the European Employment Strategy (EES) and EST projects). It is also characterized by a low level of professional capacity, a strong influence of political actors on the process of policy making, and a high degree of decentralization in real policy making/implementation. Considering this specific institutional set-up of activation policy, the question follows as to how effective activation policies can be.
The effects of the governance of activation
ft is difficult to assess the effects of the governance changes in the case of the Czech Republic. First, hardly any governance reforms were implemented from the top down, except for new activation measures such as Individual Action Plans, increased benefit conditionality, making-work-pay measures and an increased scope for ALMP measures. In spite of this lack of top-down reforms, spontaneous processes of marketization/privatization and decentralization emerged, in particular in connection with the spread of the ESF schemes/projects.
The absence of top-down governance reforms hampered effective implementation of activation measures. Nevertheless, we observed that the spontaneous reforms from the bottom (local level) up enabled, to some extent, implementation of activation policies at the local level. Although they could not effectively countervail all the governance deficits, they have facilitated timely implementation of the ESF projects.
High expectations were associated with the implementation of measures under the ESF, concerning an increased scope of activation measures and greater involvement of the vulnerable groups (Horáková et al., 2010). More individualized treatment and more intensive job search facilities were expected from the lAPs. For these reasons we will attempt to compare here the effects of the new measures financed through the ESF and lAPs (where the spontaneous changes of governance such as privatization/marketization and decentralization were most significant) with the other (traditional, national) ALMP measures. We are interested to discover whether and how the spontaneous changes of governance enable more effective implementation of activation - where these are more strongly associated with the newly implemented ESF projects rather than with the old ALMP measures.
In line with the classification of the effects of governance of activation policies as outlined in Chapter 1, we will focus here on the
Tomáš Sirovátka and Jiří Winkler   185
outcomes (in our case, indicated by the indirect measure of employment effects) and the outputs of activation measures (in our case, the scope and targeting of the policies/measures and their content), and finally on procedural effects (the degree of individualization of activation).
Outcomes: employment effects of activation measures
Using administrative data, measurement was made of how often the participants in ALMPs and the ESF projects appeared in the registers during the 12 months following completion of their participation in the national ALMP and ESF projects in 2006 (that is, an indirect measure of the employment effect) (Kulhavý and Sirovátka, 2008). We found evidence that there were differences in the employment effects of various national ALMP programmes: in the case of training and public jobs the occurrence of unemployment immediately after participation in the programme was over 60 per cent, dropping to approximately SO per cent after 12 months. The participants in more selective and less scaled national measures of job creation in the private sector exhibited lower rates of unemployment when compared with the participants in training activities and public jobs (only about 10 per cent of the former appear in the registers shortly after completion of the programme, and also after 12 months), and even lower rates in the case of self-employment support schemes.
As regards the employment effects of the ESF measures, these were apparently better than those of the national requalification programmes in 2007 (after 12 months only about 10 per cent of the participants appear in the registers). These encouraging results, however, show only gross employment effects and include a considerable dead weight, mainly due to creaming-off effects. The findings indicate higher employment effects for the ESF projects when compared with labour-market training and public works under national ALMPs. Nevertheless, analysis of the implemented measures one year later, in 2007 (Hora et al., 2009), documented a converging trend in effectiveness: after 12 months 27 per cent of the participants in the ESF measures were on the registers, as compared with 25 per cent in national requalification programmes and only 16 per cent in the national public works programme. The decreasing relative effectiveness of the ESF measures and the increasing effectiveness of national ALMP measures are due to the increasing scope of the ESF measures (see below), which have reduced creaming-ofl in selecting participants for these measures. To summarize, it seems that the new ESF measures only temporarily improved the placement effects, which lasted only until the scope of these measures was increased.
186   Governance of Activation Policies in the Czech Republic
The outputs: the scope, creaming-off/targeting and the contents of activation measures
The scope
Basically, labour-market measures are among the least developed types of public policies in the Czech Republic. The expenditure on these measures amounted to only about 0.12 per cent of GDP in recent years, which is five or even seven times less than in countries with a comparable unemployment rate (see OECD, 2007). Nevertheless, since 2004 the ESF has made it possible to finance specific projects of ALMP. This simple fact has brought new resources and more opportunities for participation that have been used by actors from both the profit and non-profit sectors (as services providers, see above).
Thanks to financing from the ESI- the scope of ALMP increased, with the proportion of active labour-market participants in the unemployment stock rising from 19 per cent to 39 per cent between 2005 and 2007. In 2008, the absolute number of participants in ALMP measures (85,000) dropped below the levels of 2006 and 2007 (141,000 and 137,000, respectively), and the proportion of active labour-market participants in the unemployment stock was 24 per cent, with participants in the ESF projects representing 50 per cent of all participants. Data from 2009 show that the scope of ALMPs had not increased much in spite of the crisis (102,000 participants, although unemployment increased from 352,000 to 539,000); the share of all labour-market measures participants in the unemployment stock was only 19 per cent, with two-thirds of these being financed through the ESF projects (MPSV, 2010).
With regard to the last five years, the ESF measures have clearly replaced a substantial part of previously extant national ALMP measures without any added value in terms of the scope of measures. This must be attributed primarily to a lack of managerial capacity at the EOs in administering an increased number of ALMP measures, even where these are being outsourced to other agents. The government shows no commitment to changing the situation, since it does not consider the modest scope of ALMPs to be a policy deficit. Instead, it prefers to improve incentives to work (see NRP, 2005, 2008) and continues its restrictive staffing policy.
Targeting
It is interesting to note that during the development of activation policies no positive change has appeared in relation to targeting and
Tomáš Skovátka and Jiří Winkler 187
eliminating the creaming-off effects. The national ALMPs have been analysed (Kulhavý and Sirovátka, 2008; Hora et al., 2009) as being somewhat biased towards selection of those unemployed people at lower risk of exclusion (that is, unemployed persons with an intermediate-higher level of education are more likely to participate in some kind of ALMP than those with only elementary education, even though the unemployment risks are reversed between the two groups). The reason for this is that the EOs select those unemployed people whom they expect to represent the least demanding workload. For example, while in 2007 around 32 per cent of the registered unemployed were people with only basic education and 31 per cent were over 50 years, their proportion among requalif'ication/training participants represented 15 per cent and 17 per cent, respectively (Kulhavý and Sirovátka, 2008).
Although the implementation of the ESF projects is based on management by indicators and objectives, which should improve targeting of the unemployed at higher risk of unemployment (in the Czech case, these are poorly educated, the disabled, the unemployed over 50 years of age and the long-term unemployed), the creaming-off effects are most evident in the programmes of vocational training for the unemployed with basic education, the unemployed over 50 years of age and the unemployed with (partial) invalidity, who are under-represented both in the national ALMP measures and in the ESF projects (Kulhavý and Sirovátka, 2008; Hora et al., 2009). For example, a logistic regression analysis shows that in 2007 the probability of participation in the national ALMP measures was twice as high for those having completed secondary education when compared with those with only a basic/elementary education. In the case of requalilication/training it was three times as high, while in the case of the ESF programmes it was 2.3 times as high. The other above-mentioned target groups are also somewhat under-represented, in both the national ALMP measures and the ESF schemes.
We can see, when looking at the actual specifications of the target groups in the ESF projects, that these definitions are often rather broad, allowing easy selection between or within groups (the latter not being apparent in data, but still representing a typical strategy to circumvent the demands set by the conditions of the ESF projects), in favour of the better-equipped unemployed. In this creaming-off process, a crucial role is played by specific employers (for whom the unemployed are typically being retrained) and their negotiations concerning the final selection of participants with the private agencies (services providers). Despite all this, targeting is still slightly better under the ESF
188  Governance of Activation Policies in the Czech Republic
programmes in comparison with the national schemes, due to the introduction of management by indicators. Even though the target groups are rather broadly/generally denned and the target indicators within the ESF projects are rather soft, the objectives to be achieved are still somewhat more explicit and more binding than those that regulate the targeting of measures of the national ALMP.
Contents of activation
The content of activation policies has, to a large degree, been influenced by the institutional framework and governance of the policies. As we have already explained, the institutional framework has not been open to much innovation, especially in terms of more demanding forms of activation and labour-market policies like vocational training and individual counselling. The Employment Act from 2004 and the National Reform Programme 2005-8 showed a strong preference for measures improving work incentives, both positive incentives (such as improvements in the minimum wage and tax bonuses, back-to-work benefits and in-work benefits/tax credits) and - especially since 2006 - negative incentives (such as a delayed revaluation of the living minimum, implementation of the existence minimum - set at a lower level than the living minimum), with the latter being provided to the unemployed assessed as passive or non-cooperative, and since 2009 to all social assistance claimants who (after six months) do not participate in public service jobs. Furthermore, greater conditionality of benefit entitlements, a stricter definition of suitable job and, finally, to some extent individual casework with the unemployed are principally aimed at increasing the administrative pressures on the unemployed.
Procedural effects
The lack of top-down governance reforms and the deficits in institutional capacity imply a rather narrow approach to activation. Individual action plans, which represented a policy innovation, have not been effectively implemented. In most countries, the process of individualizing services is a significant feature of the current governance of activation. Individualization is associated with voice being given to the unemployed (their voice is strong when they have rights in terms of co-determining the content of their individual contract and when they are provided with a choice of instruments/providers, for example in the form of vouchers). Such empowering forms have not been used in the Czech Republic. Since individual contracts, which had been strongly recommended in guideline 1 of the EES (prevention and early
TomdS Sirovdtka and Jifi Winkler  189
activation), appeared to the ESA to be a convenient and inexpensive instrument for eliminating passivity and benefit dependency or abuse, they succeeded in embedding this instrument in the new Employment Act No. 435 from September 2004, thus obliging EOs to offer the IAP to every unemployed person under 25 years after a certain period of unemployment (participation was voluntary). While the original intention was to invite large numbers of the unemployed to participate in lAPs, EOs in locations with high unemployment, where this goal was beyond the capacity of available staff, immediately deviated from (his target. Not only did the planned impacts of the programme not materialize due to the limited numbers of participants who agreed to contract the IAP, but also a formal approach in most EOs narrowed the measure to a mere signing of the contract and a rather standard job-mediation process, while simultaneously focusing the IAPs on more cooperative (more motivated) clients. While the National Reform Programme 2005-8 promised 25,000 individual contracts to be concluded in 2006, the realized number was only 7000 (MPSV, 2008).
Nevertheless, since 2009, the centre-right government has established IAPs as obligatory for all unemployed people after five months of unemployment, in spite of the lack of personnel capacity at EOs. Thus the IAPs have largely become a formal step, possibly increasing the administrative pressure on the unemployed. Since the start of 2009 the ESA has been negotiating with the government regarding the possible cancellation of the IAP as an obligatory measure, on the grounds of increased numbers of unemployed and the lack of capacity for this task (this statement is based on the authors' interview with a representative of the ESA). These ambiguous effects have resulted from the development of the IAPs: at present, IAPs are understood to be a specific tool of enforcement policy and as stimulation for the unemployed to become active in the labour market (Sirovatka et al., 2007). This basic activation tool is simply a contract that exerts a specific form of pressure on the applicant: breaking the contract may be a reason for removal of the individual from the register of unemployed people and for withdrawal of unemployment benefits.
There are also other reasons for ineffectiveness in terms of individual treatment. Privatization and outsourcing of employment services creates an expectation of a more individual approach and a higher quality of service provision due to the expected increase in interest to satisfy individual demands under conditions of competition among various service providers. However, studies from other countries show that implementation of purely quasi-market relationships (formulation
190   Governance of Activation Policies in the Czech Republic
of impersonal supply and demand, contracting services) as the decisive coordination mechanism is not a sufficient precondition for improving the quality of public services. Competition among providers under approximately identical conditions is not the only important factor (Hood, 1991) - services providers' responsibility is also significant (Goodin, 20(B). Some authors emphasize the necessity of the coexistence of various coordinative mechanisms in order that services improve. According to Gray and Jenkins (2007), in the case of educational services, for instance, the combination of a strictly defined legislative framework, privatization and the contracting of services from private agencies, as well as the development of professional communities, are all essential.
In the case of the Czech activation policy, the contracting-out of services has so far been accompanied by neither fair competition among service providers nor the development of professional education and the formation of professional communities as a coordinative mechanism for cross-sector collaboration. The present state of unfinished privatization (which lacks the principles of fair competition) and the ungoverned level of professionalism in human resources reduce any effort at individualization of services to a rather formal contract concluded between clients, service providers and the state. To summarize, the governance of activation has not brought any significant procedural effects.
With regard to the overall effects, it seems that insufficient personnel resources, low professional competence (implying a rather procedural governance in the implementation phase of activation), unfinished and inconsistent innovations, as well as mostly instrumental forms of cooperation between PES on the one hand and employers and NGOs on the other, have made it possible to use the ESF resources without paying attention to the establishment of an adequate governance framework. This has now led to a situation where governance changes have only unimportant and ambiguous effects on activation.
Conclusions: the role of governance and implementation conditions
We have shown that in the Czech Republic the governance of activation policies is traditionally a weak element of public policy making characterized by deficits in the management of activation, unfinished reforms, spontaneously emerging mechanisms and informal support networks filling the gaps in governance, and a neglected system of PES in terms of their managerial and personnel capacity.
TomdS Sirovdtka and Jifl Winkler   191
In 2004, external factors of a governance change gained more strength with the country's accession to the EU: the programming of the ESF projects associated with new formal procedures in governance, an increase in resources available for activation measures and the transmission of new ideas from the EES conveyed through the NAPE. Nevertheless, no important top-down governance reforms have been implemented despite the fact that the focus on activation has increased. In these circumstances, the policy makers deliberately focus largely on simple measures of activation that do not require a very advanced coordination of policies: these are typically work incentives implemented into the benefit schemes. The introduction of the ESF projects into activation policy has brought some change: at least an initial extension of the scope of active labour-market measures, marketization and outsourcing, and a pluralization of actors.
The coordination and managerial capacity of the centre is still weak: coordination by objectives and performance measures at the national and local levels as well as the rules set to govern the involvement of other actors are insufficient in terms of promotion of the strategic objectives of the centre or EOs, such as better targeting (at specific groups) and better quality of certain aspects of measures. Similarly, the implementation capacity of EOs needed to meet more demanding tasks, such as a more individualized and comprehensive service to clients, is also inadequate.
Consequently, an important factor is the governmental and managerial deficit concerning the preparation of an institutional environment and adaptation of the ESF programmes and facilitating their implementation in the Czech context. The key government objectives have been the making of savings in this area and a reduction in public administration staff. For this reason the ESF project measures have been aimed simply at substituting the national ALMP measures. At the same time, some functioning mechanisms of local/regional cooperation have been weakened since the National Integrated Projects that replaced the national policies have been prioritized.
Despite the foregoing, the governance of activation policies has become more complex with the implementation of new measures under the ESF. This has happened without any institutional reform pursuing the objective of better coordination, and so the previously existing institutional constraints have worsened. On the one hand, the spontaneous adaptation mechanisms of the system of PES make the functioning of the new and more complex governance structure possible, while on the other the effectiveness of the ESF projects is not very good in
192   Governance of Activation Policies in the C/.ech Republic
several respects. Initially, the ESF projects brought an increased scope of labour-market policy measures, but this effect did not last due to the insufficient managerial capacity of the EOs. Besides, the added value of these measures is doubtful since they have barely improved the targeting of activation policies towards vulnerable groups, and are sometimes associated with increased selection and creaming-off processes during implementation. The development of effective implementation of activation measures (with regards to their scope, targeting, contents and quality) is blocked by a weak coordination capacity, a poorly coordinated process of outsourcing that does not promote competition among private agencies, and deficient professional competences among PES staff. Although local actors (directors of EOs) try to respond to some of these deficits by forming local networks of mutual support, this effort cannot eliminate the substantial system deficits and the fragmented governance of the PES.
The spontaneous decentralization evoked by the new challenges represents a specific feature of the Czech governance of activation, making it possible to implement the ESF projects and adapt them to the existing mode of governance. Nevertheless, neither decentralization nor marketization, which are the key trends, have brought, by themselves (that is, without effective management by the centre), any significant effects on activation policies.
Acknowledgements
This study was written with the support of the Ministry of Education of the Czech Republic (NPV112D06009).
References
Considine, M. (2001) Enterprising States: The Public Management of Wei'fare-to-Work
(Cambridge: Cambridge University Press). Goodin, R.E. (2003) 'Democratic accountability: Distinctiveness of the Third
Sector', European Journal of Sociology, 44, 359-96. Gray, A. and B. Jenkins (2007) 'Profession and bureaucracy', in J. Baldock (ed.)
Social Policy (Oxford: Oxford University Press), flood, Ch. (1991) 'A public management for all seasons?', Public Administration,
69/1, 3-19.
Hora, O., T. Sirovátka, II. Tomešová Bartáková and J, Vyhlídal (2009) Hodnocení programu aktivní politiky zaměstnanosti realizovaných v roce 2007 se zaměřením na rekvalifikace (Prague: VÚPSV).
Horáková, M., O. Hora and J. Vyhlídal (2010) Hodnocení přínosu projektů Operačního programu Rozvoj lidských zdrojů (Prague: VÚPSV).
Tomáš Sirovátka and Jiří Winkler   193
ILO (2003) Summary of International Public Employment Services (PES) Administrative Statistics (Updated September 2003), http://www.sedi.oas.org/ddse/english/ documentos/cancun/()12()()31,ESAdiiiinStats-%20ElIeii%2()IIansen-ILC).doc (accessed 28 September 2010).
Kalužná, I). (2008) Main Features of the Public Employment Services in the Czech Republic (Paris: OECD).
Kulhavý, V. and T. Sirovátka (2008) Hodnocení efektivity programu APZ a doprovodných nástrojů a projektů politiky zaměstnanosti v roce 2007 (Prague: VÚPSV).
March, J.G. (1999) The Pursuit of Organizational Intelligence (London: Blackwell).
Marci!, J.G. and P. Olsen (1989) Rediscovering Institutions: The Organizational Basis of Politics (New York: Free Press).
MPSV (Ministry of Labour and Social Affairs) (2008) Analýza situace na trhu práce v roce 2007 (Prague: MESA).
MPSV (Ministry of Labour and Social Affairs (2010) Data available through web portal http://portal.mpsv.c/./sz/politikazainest/trh_prace/, own computations (accessed 5 October 2010).
NRP  (2005)  National Lisbon Programme 2005-2008 (National Reform Programme of the Czech Republic).  Available on http://ec.europa.eu/archives/ giowthandiobs/national-di]nension/meniber-states-2005-2008-reports/index_ en.htm (accessed 30 October 2010).
NRP (2008) National Reform Programme of the Czech Republic 2008-2010. Available on http://ec.europa.eu/archives/growthandlobs_2009/documentation/index_ en.htm#national (accessed 30 October 2010).
Obinger, IL, S. Leibfried and E.G. Castles (2005) 'Bypasses to a social Europe? Lessons from federal experience', Journal of European Public Policy, 12/3, 1-27.
OECD (2007) Employment Outlook (Paris: OECD).
Serrano Pascual, A. (ed.) (2004) Are Activation Policies Converging in Europe? The European Employment Strategy for Young People (Brussels: ETUI).
Sirovátka, T. (2003) Problémy trhu práce a politiky zaměstnanosti, Research report (Brno: Masaryk University).
Sirovátka, T. (2007) 'The individual approach in activation policy in the Czech Republic', in R. van Berkel and B. Valkenburg (eds) Making It Personal. Individualising Activation Services in the EU (Bristol: The Policy Press).
Sirovátka, T. (2008) 'Activation policies under conditions of weak governance: Czech and Slovak cases compared', Central European Journal of Public Policy, 2/1, 4-29.
Sirovátka, T., P. Horák and M. Horáková (2007) 'Emergence of new modes of governance in activation policies: Czech experience', International Journal of Sociology and Social Policy, 27/7-8, 311-23.
Taylor Gooby, P. and J. Zinn (eds) (2005) Risk in Social Science (Oxford: Oxford University Press).
Van Berkel, R. and EEL Moller (eds) (2002) Active Social Policies in the EU. Inclusion
through Participation? (Bristol: The Policy Press). Winkler, J. M. Žižlavský and P. Horák (2003) Institucionální předpoklady české
politiky zaměstnanosti, Research report (Brno: Masaryk University). Winkler, J. and M. Žižlavský (eds) (2004) český trh práce a Evropská strategie
zaměstnanosti (Brno: Masaryk University). Winkler, J., L. Klimplová and M. Žižlavský (2005) Účelové programy na lokálním
trhu práce (Brno: Masaryk University).
194  Govemance ofActivation Policies in tne Czech Republk
Winkler, J. (2009) 'Institucionální a organizační limity individualizace české politiky zaměstnanosti v podmínkách duálního trhu práce', in T. Sirovátka, |. Winkler and M. Žižlavský (eds) Nejistoty na trhu práce (Urno: Masaryk University).
Winkler, }. and I. Zelenková (2010) 'Služby zaměstnanosti a formy spolupráce s občanským sektorem na regionálním trhu práce', in J. Winkler and Ľ K! im plová (eds) Nová sociální rizika na trhu práce a potřeby reformy české veřejné politiky (Urno: Masaryk University).
Zeitlin, j. and 1). Trubek (eds) (2005) Governing Work and Welfare in a New Economy (Oxford: Oxford University Press).
10
Decentralization and Back to Centralization: the Swedish Case
Renate Minas
Governing activation in Sweden: features and trends
Labour-market policy reform has a long tradition in Sweden. With Sweden as a pioneer, Nordic governments introduced active labour-market policies early in the twentieth century and relied on these to facilitate structural changes in the economy. A significant feature of Swedish labour-market policy since then has been extensive state intervention. With the deep economic recession and massive increase in unemployment in the early 1990s, the conditions for labour-market policy were, however, fundamentally altered. To counter fears of a long-term increase in working-age people dependent on benefit payments, as well as growing costs for social assistance and unemployment compensation, the government introduced a wide array of reforms aimed at creating and strengthening the link between work and welfare systems.
To understand recent reform processes concerning the governance of activation policies in Sweden, several specific conditions have to be taken into account. One is the fact that local government and local responsibility for welfare provision have a long history, and municipal self-governance is enshrined in the constitution. Nevertheless, Sweden is a traditional unitary state where legislative power is held entirely at the national level, and the constitution also gives parliament the right to define the municipal scope of authority and to establish the range of local political influence. In this respect, national government has the ultimate power to decide how far decentralization should be carried through (Bergmark and Minas, 2007). In practice, this means that those parts of the Swedish welfare sector that are administered by the municipalities are under national government control to varying degrees. This control is exercised by means of a number of steering instruments that have a more or less imperative character. The other
195