Centre of international relations CATCHING UP AND DETERMINANTS OF SERVICES GROWTH IN NEW MEMBER STATES METKA STARE, ANDREJA JAKLIČ, ANŽE BURGER Univerza v Ljubljani Fakulteta za družbene vede Metka Stare, Andreja Jaklič, Anže Burger CATCHING UP AND DETERMINANTS OF SERVICES GROWTH IN NEW MEMBER STATES Electronic book series Analize CMO / CIR Analyses Editor: Sabina Kajnč Edtorial board: Sabina Kajnč, Andreja Jaklič, Matija Rojec, Ana Bojinovic Fenko Publisher: Faculty of Social Sciences, Založba FDV Reviewers: Agnes Ghibutiu and Matija Rojec Language editing: Nataša Zajec Hercog Book design: Luka Kaše Accessible to: http://www.mednarodni-odnosi.si/cmo/cir-analyses.htm Copyright © Faculty of Social Sciences, University of Ljubljana, 2010. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise without the prior permission of the copyright owner. CIP - Kataložni zapis o publikaciji Narodna in univerzitetna knjižnica, Ljubljana 338.46(4) 330.342.14/.15(4) STARE, Metka Catching up and determinants of services growth in new member states [Elektronski vir] / Metka Stare, Andreja Jaklič, Anže Burger. - El. knjiga. - Ljubljana : Faculty of Social Sciences, 2010. - (Electronic book series CIR analyses = Analize CMO) Način dostopa (URL): http://www.mednarodni-odnosi.si/cmo/cir-analys es.htm ISBN 978-961-235-396-4 1. Jaklič, Andreja 2. Burger, Anže, 1979251354880 METKA STARE,* ANDREJA JAKLIČ,* ANŽE BURGER* CATCHING UP AND DETERMINANTS OF SERVICES GROWTH IN NEW MEMBER STATES Ljubljana, 2010 Faculty of Social Sciences, University of Ljubljana, Ljubljana, Slovenia CONTENTS 1. INTRODUCTION..................................................................................................................7 2. TRANSFORMATION OF SERVICES DURING TRANSITION................................11 2.1. Defining service sub-sectors...............................................................................11 2.2. The features of the tertiarisation process in the CEECs.............................13 3. EVOLUTION OF PRIVATE AND PUBLIC SERVICES EMPLOYMENT DURING TRANSITION......................................................................................................................17 3.1. Stylised facts on services employment growth...........................................17 3.2. Catching up and convergence process..........................................................21 3.3. Benchmarking structural change in NMS services employment...........25 3.4. Assessing the performance of private and public services......................29 4. DETERMINANTS OF SERVICES employment GROwTH IN NMS................35 4.1. Explaining services growth - evidence from literature.............................35 4.2. Methodological approach and data................................................................38 4.2.1. The model.....................................................................................................38 4.2.2. Description of variables...........................................................................39 4.3. Evaluating services developement in NMS..................................................41 4.3.1. Understanding services heterogeneity.............................................41 4.3.2. Distinguishinig between public and private services..................43 4.3.3. Understanding the diference between public, private and mixed services............................................................................................. 45 4.3.4. Difficulties in explaining services growth in NMS..........................49 5. CHALLENGES OF SERVICES DEVELOPMENT AND SERVICES EMPLOYMENT IN NMS................................................................................................................................51 5.1. Variety of catching up patterens: top-down and bottom-up...............51 5.2. Transition matters and increases services heterogenity.........................53 5.3. Service outsourcing as a catalyst for productivity improvements and restructuring of public and private services...............................................54 5.4. Fostering quality and innovation in services..............................................55 5.5. Conclusions.............................................................................................................57 References...............................................................................................................................59 Index.........................................................................................................................................63 1. INTRODUCTION1 Since the launching of transition reforms, the Central and Eastern European Countries (CEECs) have seen a radical structural transformation and institutional changes that have been reflected in the shifts of output and employment in favour of services. Even though dynamics and patterns of these shifts vary among countries, depending on both, the different backgrounds and the speed and nature of transition reforms, the services sector has acquired a dominant position in all New Member States (NMS2) and has been catching up with the EU15. The discussion of the dimensions of the structural change in the NMS has mostly concentrated on the process of shedding jobs in manufacturing and the related increase in the share of services employment. Much less attention has been paid to the patterns of employment transformation in two major categories of services (private/market services and public/non-market services3) and to the determinants of their growth during the transition. In the past, most services in CEECs were provided by the public (state) sector, whereas private suppliers hardly existed prior to the transition. With the change of political system and particularly with the introduction of market mechanisms, private suppliers of services in the CEECs experienced a rapid growth driven by privatization, liberalization and deregulation of the economy. Private services were explored more intensively owing to their poor development record in the past and the catalytic role in overall competitiveness of the economy. These services were at the focus of research also because they attracted the bulk of FDI inflows. Foreign direct investors in CEECs targeted those services that were at the infant stage of development, yet critical for smooth functioning of the economy (e.g. financial services, telecommunications, and business services). On the contrary, the reform process was delayed in public services, as their relative share in total employment was already high at the outset of reforms; however, 1 Part of this analysis is based on an article (Burger, Stare, 2010) whose final and definitive form has been published in 'The Services Industries Journal' (c), Vol. 30, No. 3/4, 2010 Copyright Taylor & Francis; The Services Industries Journal is available online at http://www.informa-world.com/0264-2069 2 The term CEECs denotes ten former socialist economies that became the new EU members (NMS) in 2004 and 2007. Malta and Cyprus, who joined the EU in 2004, do not share the characteristics of the past socio-economic system with other NMS, but are included into the analysis in some chapters. Both acronyms, CEECs and NMS are used interchangeably throughout the text. 3 See chapter 2.1 for definitional issues. the quality and performance of those services is arguable. Nevertheless, public services were considered the building block of the socialist economies and perceived as a fundamental element of welfare and equality. During the transition, researchers and policy-makers seemed to be much less interested in exploring the role and the changed character of public services in a different socio-economic setting, although some early discussions on New Public Management could be observed in the CEECs, mostly as a reaction to public budget constraints. Only at a later stage of transition have the issues of public services transformation and adjustment to the new realities became more apparent, along with an increased demand for high-quality public services and improved efficiency in public services spending. Gradually, it became evident that both public and private services matter for a competitive economy, and that public services are not an isolated segment of market economy, but need to adjust to businesses, technology and society in terms of responsiveness, quality and efficiency. Moreover, some services are increasingly supplied by both private and public suppliers (mixed services). The use of various market mechanisms in the provision of public services and the need for collaboration with private sector is increasingly seen as instrumental for easing the pressure on public spending. Nevertheless, the growth patterns of private, public or mixed services were not examined in the NMS. The role of various determinants of services employment growth was studied extensively, based on the evidence of developed market economies where tertiarisation evolved as a gradual process reflecting the changes in demand (income growth) and supply (technological development). Much less is known, however, on the transformation of services sector and its drivers in the former socially-planned economies, where political, economic, social and institutional changes were very rapid in the last two decades and could have played a role. The analysis attempts to contribute to closing the gap in the research of private and public services transformation during the transition and to unveil the distinguishing features of the two categories based on solid empirical un-derpinning.4 The objectives of the analysis are twofold. Firstly, it examines the evolution of private and public services employment5 in the NMS against the benchmark of market economies, catching up process and the convergence with 4 Acknowledgement: The research was financed partly by the European Commission funds for FP7- ServPPIN project (The Contribution of public and private services to European growth and welfare and the role of public-private innovation networks) and partly by the Slovenian Research Agency (Research Programme on »Capabilities and Opportunities of Slovenia and its Actors within the Globalisation Processes», P5-0177). 5 In analysing structural changes over time and across countries, the use of employment indicators is preferred to value added indicators on account of poor availability of data sets at constant prices and volume indices. This is even more relevant for the transition economies characterised by high price distortions in the past and rapid changes in relative prices of services after the introduction of market reforms. the EU15 patterns. The empirical analysis is complemented by an examination of differences between the new and the old EU member states as regards private services efficiency and public services performance. Secondly, the analysis aims to shed some light on the underlying reasons for the shift in employment towards services in the NMS, building on the standard explanatory factors as well as transition related determinants. After the introductory remarks, the definitional issues of measuring private and public services employment are discussed. Further, we reflect upon the transformation of private and public services in the context of transition reforms in NMS. The third chapter of the analysis presents some stylized facts on services employment growth in the NMS during transition. This is followed by the convergence analysis of private and public services development in the NMS in the period 1995-2005; it is based on the EU15 benchmark and the GDP level benchmark for market economies with the aim to identify similarities and dissimilarities. Afterwards, we assess the structural transformation in the CEECs from a more qualitative perspective by exploring the progress in efficiency/performance of private and public services compared to the EU15 standards. Chapter four first briefly refers to the literature on the determinants of services employment growth, which sets the ground for an examination of factors contributing to the increase in services employment in NMS. After the description of methodological approach and variables, the regression analysis is applied on panel data for the period 1995-2007. We estimate the significance of different explanatory variables on employment growth for services in general and for service sub-groups. The results are discussed distinguishing between the impacts of standard determinants of services employment growth and transition-related factors referring to market reforms, institutional change and governance. Chapter five concludes by summarizing the main results with implications for policy shaping in the NMS, and by suggesting possible avenues of further research. 2. TRANSFORMATION OF SERVICES DURING TRANSITION 2.1 DEFINING SERVICE SUB-SECTORS Even though the division between private and public services is commonly used in services research to distinguish between the heterogeneous services along one key criterion, such as the type of ownership of the services provider,6 the dividing line is much less clear when it comes to the definition of private and public services for the purpose of empirical investigation.7 In addition, the borderline between private and public services changes across countries, time and institutional settings. A further inadequacy of definition refers to the fact that public services are often equated with the public sector. While this may be a problem when public sector expenditures are concerned, as they are not related solely to the provision of the public services, it is much less of a problem when public services employment is analysed, since the employment in the public sector mainly refers to the provision of public services. In defining the scope of the public sector, different approaches are recognized. Gem-mel claims that much of the theoretical and empirical literature has concentrated on the public expenditure approach, partly due to the ease of measurement and availability of data. He, however, suggests that it is perhaps most useful to define public sector with regard to economic activities, institutions or individuals which are controlled by the government. This does not refer only to ownership but to exercising legal control over areas of economic activity (e.g. regulation of education, consumer protection) without any legal ownership (Gemmel 1993: 3). Taking into account new trends in public management, the involvement of private sector in the provision of public services and the proliferation of public-private partnerships in services provision, it is getting increasingly difficult to properly capture and measure the volume of public or private services. In view of the above measurement problems, our research of the public and private services employment relies on the use of proxy categories for which data are more readily available. Private services are approximated with the data for market services employment 6 Other major categorizations distinguish between intermediate and final consumption services, producer and household services, basic and non-basic services, etc. Further, disaggregated groups of services, such as for example the division between distributive, producer, social and personal services, are used in empirical analyses (Illeris 1996: 25-27). 7 On account of better data availability, private and public services are most often approximated by data for market and non-market services. Throughout the text we use the terms public/ non-market services and private/market services interchangeably. and public services with the data for non-market services employment.8 It is assumed that market services are provided by private companies in a competitive environment with little public regulation and that the non-market services are produced by the state or by the public institutions, predominantly financed by public funds.9 The reality is, however, much more complex and there are numerous hybrid situations differing in the extent to which market/non-market services are subject to state regulation. An argument in favour of such an approximation is that even though a number of non-market services are provided by private suppliers (e.g. in health, education), they still involve a high level of public expenditure and regula-tion.10 This, however, does not imply that imperfections in measuring public and private services employment can be overlooked when interpreting the results. Beside the contentious character of the definition, public services are fraught with measurement problems of output, which may not reflect real changes but statistical artefacts11. In fact, we tend to measure outputs by inputs into public services provision (e.g. number of employees or hours worked) while output relates to quality-adjusted volume indicators (for example pupil hours in education) for which methodologically comparable data are not yet available.12 However, what is of ultimate importance in public services is the improvement of the outcome/performance for final users of public services, which is much more difficult to measure (e.g. health improvement of a patient). Nevertheless, a number of indicators could be used to approximate the performance of public services and then compared across countries (see chapter 3, subchapter 3). The following chapter aims to throw light on the context of structural change, the importance of inherited and evolving new institutions in the transformation of private and public services in the CEECs. 8 Using the NACE-Rev.1 classification of economic activities, market services include wholesale and retail trade (G), hotels and restaurants (H), transport, storage and communications (I), financial intermediation (J), real estate, renting and business activities (K), whereas non-market services include public administration and compulsory social security (L), education (M), health and social work (N), other community, social and personal services (O), private households with employed persons (P). 9 Non-market services cover the services provided to the community as a whole free of charge or at a fee which is well below 50 percent of the production costs (OECD 2008). 10 High level of regulation and substantial direct public involvement in the non-market services exist in all OECD countries. 11 The most important issues of the efficiency measurement in the production of public services are: how to define output (output versus outcome); how to define aggregate output over a range of different products; how to incorporate exogenous conditions in the valuation of efficiency; lack of information on the market value of the output as public services are often provided to the user at no cost or at subsidized price. Efforts are being made at national and international level to improve the measurement of public sector output, including the improvement of price and volume measures of non-market services (EC 2004: 23). 12 Eurostat - OECD Task Force on non-market services is investigating the use of output methods for comparisons of education and health services across countries. For details see draft OECD Handbook Measuring Education and Health Volume Output (OECD 2007). 2.2 THE FEATURES OF THE TERTIARISATION PROCESS IN THE CEECs When launching political changes and market-oriented reforms, the CEECs at first experienced a severe decline in output and employment, which was followed by a rapid recovery. Since 1995, economic restructuring has proceeded hand in hand with productivity and technological catching-up. The adjustment process initiated by the implementation of market reforms brought about the shifts in employment structures. The most significant was huge downsizing in manufacturing, and even though services became the major generator of new employment, this was not sufficient to yield overall positive employment growth in the CEECs. The employment levels did not reach those of the beginning of the transformation in most CEECs.13 Although the pace and scope of the tertiarisation varied considerably among the CEECs,14 services attained a dominant share in employment in all countries, except Bulgaria and Romania by the end of the 1990s. Enhanced terti-arisation was partly also a result of reclassifications and externalisation of services previously undertaken in-house by large industrial conglomerates.15 A number of scholars observe that the CEECs were characterised by excessive industrial employment compared to other countries at the similar income level, while the service sector was underdeveloped (Landesmann 2002; Mickiewicz 2001; Raiser, Schaffer and Schuchardt 2004; Rutkowski and Scarpetta 2005). This was particularly the case of market services and was related to the very essence of patterns of economic growth in the former socio-economic system and its underlying implementation and management mechanisms. The division of labour in this context did not favour market relations and the consequent comprehensive system of linkages which characterize market economy. In the absence of market forces and market relations in highly centralised economy, the lack of entrepreneurship and innovation, as well as isolation from the external markets, there was no room for development of service activities (Ghibutiu 2000: 307-308). Apart from the non-existent or poorly developed market-related services in the centrally-planned system, the character of services, such as for example distribution or financial services, was entirely different from the one in market economies 13 Based on the analysis of five countries - Czech Republic, Hungary, Poland, Slovakia and Slovenia - Havlik points out that Hungary was the only country recording employment growth throughout the period 1995-2004 (Havlik et al, 2008, 39). 14 The intensity of changes in the 1990s differed depending on the level of employment in services at the outset of the transition and on the trends in agriculture employment, which in some countries served as a buffer to shedding labour in manufacturing. 15 For example, a typical large conglomerate supplied auxiliary services such as kindergarten, doctor's office, residential construction, vacation facilities for employees, etc. and basically had to align with the central plan requirements.16 With the change of the socio-economic system, the CEECs were, on the one hand, faced with the need to adapt the existing services (e.g. distribution, financial services, communications) to market mechanisms, and on the other hand, to develop and assimilate new services that had not existed under the previous political and institutional system (e.g. asset valuation, auditing, management consultancy, marketing, advertising, public relations, etc.). Moreover, the adjustment of the business processes to technological modernisation (the introduction of ICT) required a variety of specialised services (e.g. computer-aided manufacturing, computer-aided design, intellectual property protection, and introduction of quality standards) (Stare 2007: 174). Foreign direct investment also played an important role in strengthening the service sector, particularly in some countries (e.g. in Hungary, Czech Republic, Estonia) and contributed to improved managerial and marketing skills, know-how, and service quality - the elements crucial for market services development. The complexity of changes undertaken by the CEECs since the early 1990s in developing market services was immense and has increased the share of market services in value added and employment. Nevertheless, the changes could not be absorbed rapidly enough in terms of efficiency or quality of market services, availability of advanced knowledge-intensive services or their innovation capacity. These are considered the missing links of CEECs' services development, as they depend rather on soft elements (knowledge and market skills, risk taking, attitude, and service culture) which need more time to accommodate to the new situation. They are deeply rooted in the mindset of entrepreneurs, policy makers and academic circles and accordingly change gradually over longer periods of time (Stare 2007: 182). Non-market services seem to have experienced a somewhat different trajectory during the transition period, thanks to relatively well developed public services that had been perceived as fundamental for supporting the political system. The term »relatively well developed« relates to a large share of employment in public services and their fairly good accessibility to all citizens; it is not related to their quality or the efficiency of public spending on services.17 The socialist socioeconomic system attributed high value to equality and economic security as an ideal (Tomer 2002: 422) and was better in providing certain »collective services«, such as health, education or security, while the »service gap« related to the provision of market-related services in a non-market economy remained (Kostecki 16 Nevertheless, the situation differed across the CEECs. Slovenia for example had a unique system of so-called social ownership and a planning system combined with elements of a market system that allowed for the provision of some services on a market basis. 17 The empirical evidence confirms that even at the beginning of the 21st century the NMS achieve only average performance of public services with too many inputs, which points to a much lower efficiency compared to the EU15 (Afonso et al. 2006). 1996: 4). Consequently, public services redesign received little attention and was delayed. It is not surprising though, that the share of public services employment was rather high at the beginning of transition compared to the situation in market economies at the similar level of per capita GDP (Raiser et al. 2004), with significant implications for public finances (Rutkowski and Scarpetta 2005: 130). It goes without saying that rather generous expenditures for public services that resulted in high employment in these services could not be sustained under a different socio-economic system and evolving market mechanisms. However, in the first phase of the systemic change up to 1995, the public sector in the CEECs lacked a strategic development concept and was to a considerable degree used as a buffer against unemployment and remained oversized, with adverse selection (Csaba 2005: 141). During the transition, some parts of public services were privatised and new private providers of public services appeared. Besides, some specialized welfare services were disentangled from the in-house provision in enterprises. Unlike in the EU15 where reforms in the public services started back in 1980s with the aim to create a management culture focusing on citizens and involving accountability for results, the process of improving the performance of the public sector had evolved very slowly in the CEECs until the late 1990s. Afterwards, these countries were faced with a need to cut public expenditures and downsize employment in public services. On the other hand, the demand for specialised high-skilled services, mainly in public administration, increased due to the establishment of new institutions and mechanisms needed for the implementation of market reforms and the accession to the European Union (e.g. legal services related to the harmonisation with the acquis communautaire, veterinary inspection services, autonomous regulatory agencies services, etc.). Additional employment in public administration was therefore not unusual; moreover, it is claimed that the magnitude of employment increase in the public administration was striking (Mickiewicz 2001: 15). By and large, the service sector in the CEECs has made a big leap forward since the beginning of the 1990s and has narrowed the gap in relation to the EU average. Nevertheless, individual CEECs experienced different dynamics in the transformation of the service sector and also in the transformation of public and private services. What follows is the empirical investigation of the transformation patterns of two major groups of services, private and public in the CEECs relative to those in the developed market economies, and in particular of the convergence with the EU15. 3. EVOLUTION OF PRIVATE AND PUBLIC SERVICES EMPLOYMENT DURING TRANSITION 3.1 STYLISED FACTS ON SERVICES EMPLOYMENT GROwTH Since the 1990s, overall growth of the service sector in the CEECs can be attributed to a set of factors, of which market-oriented reforms played a pivotal role. In addition to these factors, which have primarily accelerated growth of market services, institutional changes also provided an impetus to public administration services (Stare 2007: 174). Our research confirms such patterns, since the share of private services employment increased in all NMS in the period 1995-2005 (Table 1). Cyprus and Malta as non CEECs surpassed the EU15 average share in private services employment in 2005, while the Czech Republic and Slovakia came closest to the EU15 average among the CEECs. Public services employment expanded as a share of total employment in most NMS, except three countries, which saw a decline in public services employment share. The NMS still lag behind the EU15 in terms of employment shares in both private and public services, with the gap being on average larger in the former (See Figure 1 and Figure 2). Table 1: Share (%) of private and public services in total employment, 1995-2005 year BG CY CZ EE HU LT LV MT PL RO SK SI EU15 Private services 1995 24.2 43.0 33.8 31.2 30.6 26.8 28.4 36.5 26.9 14.8 28.0 29.3 37.2 2000 27.2 46.3 35.8 35.4 33.6 26.8 33.4 40.5 30.1 16.6 32.6 31.2 39.8 2005 27.4 44.8 37.1 34.6 35.7 29.7 35.2 43.3 31.6 20.6 37.5 32.8 41.1 Public services 1995 18.9 19.1 19.3 24.6 28.1 24.9 26.3 25.8 21.7 13.9 26.0 16.5 27.1 2000 18.3 21.2 20.3 24.3 26.1 27.9 26.2 27.0 21.7 14.4 26.7 19.3 26.9 2005 16.9 22.0 20.8 26.4 26.9 27.2 26.8 28.0 22.5 16.9 24.0 21.3 28.3 Note: BG (Bulgaria), CY (Cyprus), CZ (Czech Republic), EE (Estonia), HU (Hungary), LT (Latvia), LV Lithuania), MT (Malta), PL (Poland), RO (Romania), SK (Slovakia) , SI (Slovenia); for Bulgaria 1995 data is for the year 1996. Source: Own calculations, EU KLEMS Database; WIIW 2008 and ILO 2009 for Bulgaria and Romania. Figure 1: Deviation (in % points) of private services employment share in NMS from EU-15 average, 1995-2005 1995 2000 2005 year Note: BG (Bulgaria), CY (Cyprus), CZ (Czech Republic), EE (Estonia), HU (Hungary), LT (Latvia), LV Lithuania), MT (Malta), PL (Poland), RO (Romania), SK (Slovakia) , SI (Slovenia). Source: Own calculations, EU KLEMS Database; WIIW 2008 and ILO 2009 for Bulgaria and Romania. Figure 2: Deviation (in % points) of public services employment share in NMS from EU-15 average, 1995-2005 O BG CY CZ EE HU LT LV MT PL RO SI SK 1995 2000 2005 year Note: BG (Bulgaria), CY (Cyprus), CZ (Czech Republic), EE (Estonia), HU (Hungary), LT (Latvia), LV Lithuania), MT (Malta), PL (Poland), RO (Romania), SK (Slovakia) , SI (Slovenia). Source: Own calculations, EU KLEMS Database; WIIW 2008 and ILO 2009 for Bulgaria and Romania. A detailed examination of individual service industries reveals that NMS remain well below the EU15 average as regards employment share in financial services, health, real estate, renting and business services (see Table 2). Wholesale and retail trade, transport and communications, public administration, defence and education were the closest to the EU15 average in terms of employment share in 2005. However, particularly in distributive trades and education, a number of NMS have surpassed the EU15 average. Bulgaria and Romania appear to deviate most from the average employment structure of the EU15, while some NMS reveal exceptionally high shares of employment in some services (e.g. Slovakia and Latvia in wholesale and retail trade, Lithuania and Estonia in transport services, Latvia and Estonia in education). Table 2: Share (%) of services in total employment by sectors, 1995-2005 NACE year BG SI EU15 USA 12.3 14.7 17.6 12.3 14.7 17.2 12.3 14.7 17.0 4.8 7.0 7.0 7.0 13.0 8.9 8.6 8.0 8.7 8.4 9.0 7.0 Note: BG (Bulgaria), CY (Cyprus), CZ (Czech Republic), EE (Estonia), HU (Hungary), LT (Latvia), LV (Lithuania), MT (Malta), PL (Poland), RO (Romania), SK (Slovakia), SI (Slovenia); The 1995 figures for Bulgaria are for 1996. Source: Own calculations, EU KLEMS Database; WIIW 2008 and ILO 2009 for Bulgaria and Romania. Rubalcaba and Di Meglio (2009) analysed the changes in three categories of services for old and new member states in the period 1995-2005 to take account of the heterogeneity of services employment growth. Apart from public and private services, they introduced a third category of services based on ownership: some services have mixed ownership, meaning that they can be provided either by public or/and private suppliers18. Results presented in Table 3 indicate that in the period under observation the NMS experienced faster growth in market and in public services employment relative to the EU15. Employment in mixed services declined in NMS and increased slightly in the EU15. NMS recorded the largest gap in private/market services employment compared to the EU15, mainly due to low employment share in business services. Even though these services recorded the most rapid growth rates of all services in the period 1995-2005, there is still a substantial scope for catching up. There is almost no difference regarding the employment share in public services in a narrow sense of word (only public administration) between old and new member states. Mixed services display a gap in NMS employment share relative to the EU15, however the results vary across service activities. The lag of NMS behind the EU15 structure is the largest in health services, whereas in education NMS even have larger share of employment than the EU15 (Rubalcaba; Di Meglio, 2009). Table 3: Share of services in employment, 2005 and annual growth rate (AGR), 1995-2005 (%) Share in 2005, in % AGR 1995-2005 EU15 EU10* EU15 EU10* Market services 37.7 28.7 1.2 2.1 Distributive trades 15.1 15.3 -0.08 1.18 Hotels and restaurants 4.9 2.7 1.46 2.22 water & Air Transport 1.7 1.1 2.09 1.59 Financial services 2.9 2 -0.74 0.81 Real estate, renting and business services 13 7.5 3.15 4.75 Public services 6. 7 6.6 -1.12 1.36 Mixed services 25.3 22 0.5 -0.3 Education 6. 7 7.6 0.36 0.66 Health and social work 9.8 5.7 0.93 -1.35 Other community, social and pers. services 4.9 3.6 1.19 0.93 Post and telecommunications 1.5 1.4 -0.93 -1.16 Inland transport 2.5 3.7 -0.83 -1.15 * NMS that joined EU in 2004. Source: Rubalcaba, De Meglio, 2009, based on EU KLEMS Database. 18 Some services denoted as private services are supplied both by public and private providers (for example transport services). However, the share of publicly and privately supplied transport services is difficult to measure and it varies among modes of transport. While railway transport is mostly publicly provided, road transport is more extensively supplied by private sector. This broader perspective confirms that the NMS have narrowed the gap in services employment relative to the EU15 average19. The catching up process differed substantially across countries and across service activities, owing to different starting positions and the efficiency in implementing the reforms. 3.2. CATCHING UP AND CONVERGENCE PROCESS In order to test whether the observed progress in employment structure of NMS towards old EU15 was significant and deeper in less-developed economies, we turn to 6-convergence analysis. Since the NMS constitute a group of less-developed countries in the sample of EU members, the results at the same time test for the convergence of NMS employment structure towards developed EU members. The following regression equation is estimated: 1 ^log (s, r/5„ ) = a - [(l - e-pT )/ T[ log (5 „ )+ i/^ r , (l) where s is employment share of country i in service sub-sector j at time t, fi is the rate of convergence, and T is the last period in the observed time interval. If the term that multiplies the initial share turns out to be negative, we conclude that less-developed countries tend to increase relative employment in a given industry faster than developed economies20. 19 We do not enter into the discussion of differences between the old and new member states in other characteristics of employment, such as for example part-time work, which plays an important role in the old EU but is almost negligible in the NMS (Landesmann et al., 2004). 20 The form of estimation equation enables to test for absolute convergence, but not conditional convergence. If the initially advanced countries converge to a higher steady state level of employment share, the estimates offi will be biased towards zero (Barro and Sala-i-Martin 2004: 467). Insignificant coefficients thus rule out absolute convergence but are not able to reject conditional convergence. Table 4: Beta convergence statistics, EU27 (1995-2005) 1995-2005 1995-2000 2000-2005 Chow test B LR test Private Services -0.064*** -0.037* -0.092*** F(1, 50) = 3.43 0.101** X2(2) = 162.74 (0. 013) (0. 021) (0.020) Prob>F = 0.0700 (0.037) Prob > x2 = 0.000 Public Services -0.055** -0.076** -0.044 F(1, 50) = 0.37 0.080 X2(2) = 155.21 (0. 026) (0.033) (0.041) Prob>F = 0.5438 (0.058) Prob > x2 = 0.000 Notes: Standard errors in parentheses. *, **, *** denote the significance at 10%, 5% and 1%, respectively. Chow tests test whether the coefficients on initial employment share are the same for the two sub-periods. ft is calculated by nonlinear least squares estimation of equation (1). The likelihood ratio (LR) statistic refers to a test of the equality of the coefficients of the log of initial employment share over the two sub-periods (1995-2000 and 2000-2005). Number of observations is 27. Source: Own calculations, EU KLEMS. Table 4 reveals that convergence was faster in private than in public services. The two groups recorded opposite trends in the strength of convergence: private services of the NMS were catching up faster in the second half of the period, while the convergence within the public services was stronger in the 1990s. However, the difference in convergence intensity is not confirmed by the Chow tests due to large standard errors. On the other hand, nonlinear estimation and subsequent LR test statistics corroborates faster convergence in private services, as well as the opposite time trend of the convergence intensity. The evidence thus suggests that NMS converged faster towards more advanced EU economies in private services and the convergence speed intensified in time. The fastest convergence is observed in wholesale and retail trade, hotels and restaurants, public administration and defence and other community, social and personal services. Financial intermediation experienced no significant convergence across EU27, neither in the entire period nor in any of the two sub-periods, even though it attracted by far the largest inward FDI into the CEECs21 (UNCTAD 2004: 302). In order to further illustrate the changing landscape of public and private services in EU27, a cluster analysis was introduced to check the emergence of clusters of EU countries with respect to employment shares in private and public services.22 Cluster analysis applied reveals potential groupings of countries with similar employment shares in both private and public services. Figure 3 presents percentage point deviations from the EU15 average for the latest year available 21 This could be explained by a very low level of financial services development in the CEECs at the outset of reforms, which could not be significantly improved in the medium term. 22 Cluster analysis applied reveals potential groupings of countries with similar employment shares in both private and public services. for all EU member states23. The first cluster that can be identified is composed of the Netherlands and UK and can be aligned with the Anglo-Saxon economic system characterised by a predominant share of private services employment. Finland, Sweden, and Denmark can be ranked in a Nordic welfare-state category of countries with extensive provision of public services, whereas France, Belgium, Germany and Ireland conform to the Continental model, being close to the EU15 average employment share in both types of services. Figure 3: Deviations (in % points) from EU15 average employment shares in private and public services, 2005. Note: AT (Austria), BE (Belgium), BG (Bulgaria), CY (Cyprus), CZ (Czech Republic), DE (Germany), DK (Denmark), EE (Estonia), ES (Spain), FI (Finland), FR (France), GR (Greece), HU (Hungary), IE (Ireland), IT (Italy), LT (Latvia), LU (Luxembourg), LV (Lithuania), MT (Malta), NL (the Netherlands), PL (Poland), PT (Portugal), RO (Romania), SE (Sweden), SK (Slovakia), SI (Slovenia), UK (United Kingdom). Source: Own calculations, EU KLEMS Database (http://www.euklems.net); WIIW and ILO for Bulgaria and Romania. 23 Excluding Cyprus and Malta. The same refers to a cluster composed of Poland, Slovenia, and Portugal, which have roughly equally advanced private and public services, but a larger deviation from the EU average. The group comprising of Lithuania, Estonia, Latvia, and Hungary is closer to the EU15 average in terms of public services employment share, while the last group (Czech Republic, Spain, Slovakia, Italy, Greece, and Austria) is relatively better off in terms of private services. An obvious outlier among the more advanced EU15 is Luxembourg with an above-average share in private services and a relatively low share of public services employment. Romania and Bulgaria are the outliers with large discrepancies in employment shares in both types of services. Overall, the cluster analysis could not identify very homogenous groups (see Figure 4), although the patterns of some groups tend to resemble those proposed by Gadrey's models of service economy (Gadrey, 2007: 50-52). Figure 4: Dendrogram for hierarchical cluster analysis o _ "sl- IS O iS CN E C0 C/3 TD CN a n h AT GR IT SK ES CZ PT Si PL EE HU LV LT BE FR DK DE IE NL UK Fl SE LU BG RO Source: Own calculations, EU KLEMS. Another feature emerging from Figure 3 should be noted - the bottom right quadrant remains empty, except for Luxembourg, which is the only EU country with an above-average employment in private services and a large gap in employment in public services. The fact that no other EU country fits into a similar pattern seems to reflect the importance of public services employment. It is maintained further by the ageing population requiring more health care and increasing role of education services. By the manner of construction of Figure 3, no decisive conclusions can be made regarding the general level of public and private services employment shares. However, two conclusions can be drawn based on the distribution of countries' positions relative to the EU15 average. First, higher level of development results in a higher employment share in both groups of services on average. Second, according to the positive association in the scatter plot, there is no imminent trade-off between private and public services employment shares, as they obviously complement rather than substitute each other. Moreover, the integration of NMS into cluster analysis discloses additional characteristics of services evolution and suggests the existence of more varied service models in the enlarged Europe. This calls for a thorough examination of the determining economic and societal parameters that shape the service economy in different environments - a task that reaches beyond the scope of this analysis. 3.3 BENCHMARKING STRUCTURAL CHANGE IN NMS SERVICES EMPLOYMENT In order to acknowledge for different steady state levels of services employment towards which NMS converge, a complementary analysis was employed, which derives benchmarks for a structure of employment in service sectors from the established market economies. To this end, Chenery-type cross-country regressions were used (Chenery and Taylor 1968). This approach provides benchmarks for a given level of development based on the data from non-transition countries across the globe. Accordingly, the shift of NMS employment share relative to the corresponding benchmark shows the progress towards economic structure of the market economy at a certain income level. The approach is similar to Raiser et al. (2004), but with some important modifications. Our analysis focuses on private and public services, further dissected into narrower categories, while Raiser et al. (2004) analyzed the changes in the structure of employment across industry, agriculture and services. Further, we investigate a more recent time interval (1995-2005), when the initial transition-related recession in the first half of the 1990s was over. Most importantly, we introduce a series of dynamic reference points for employment shares in the selected market economies for each year, while the previous analysis referred to a single benchmark in 2000. Since technology and economic progress in benchmark countries change over time, we could rightly expect the shifts in benchmarks as well. So, the comparison of current employment shares on the past benchmark shares would be invalid. To determine the benchmark shares for each service sub-sector, the regresion analysis was employed, using data for approximately 50 market economies worldwide,24 24 Australia, Argentina, Bangladesh, Bolivia, Canada, Chile, Colombia, Costa Rica, Cyprus, Egypt, El Salvador, Iceland, Israel, Japan, Korea, Malta, Malaysia, Mauritius, Mexico, New Zealand, Norway, Panama, Peru, Philippines, Saudi Arabia, Singapore, Sri Lanka, Switzerland, Taiwan, Thailand, Turkey, Uganda, USA, Uruguay, and EU15 countries. regardless of the level of development. The latter is proxied by GDP per capita at PPP and is used to estimate the relationship between the level of development and the share of employment in a given service sector. We apply a 3-rd order polynomial in log GDP per capita in PPP to make the closest fit to the employment share.25 These reference employment shares are then compared to corresponding shares in NMS in order to analyze the dynamics of structural change in private and public services from 1995 to 2005. Figure 5: Distance between actual and benchmark employment shares in private and public services (in % points), 1995 and 2005 Note: Values represent % point deviations from Chenery-type benchmark employment shares. Arrows are informative and do not correspond to actual transition paths. AT (Austria), BE (Belgium), BG (Bulgaria), CY (Cyprus), CZ (Czech Republic), DE (Germany), DK (Denmark), EE (Estonia), ES (Spain), FI (Finland), FR (France), GR (Greece), HU (Hungary), IE (Ireland), IT (Italy), LT (Latvia), LU (Luxembourg), LV (Lithuania), MT (Malta), NL (the Netherlands), PL (Poland), PT (Portugal), RO (Romania), SE (Sweden), SK (Slovakia), SI (Slovenia), UK (United Kingdom). Source: Own calculations, EU KLEMS, WDI Online Database (World Bank 2009a), ILO 2009. 25 We experimented with lower order polynomials and linear instead of log-linear model as well, yet the chosen specification produced the best fit. Figure 5 shows the evolution of private and public services employment with respect to the benchmark values derived from consecutive regressions of sectoral share of employment of market economies. In half of the NMS, the employment shares in private services converged to the reference point by 2005. On the other hand, the development of private services in Bulgaria, Estonia, Lithuania, and Slovenia has not matched the economy-wide progress in terms of GDP per capita, either due to slow transformation from the agrarian to service economy (Bulgaria) or because of a slower implementation of reforms and still rather high employment in manufacturing. Malta and Cyprus as market economies clearly differ from the rest of the NMS in regard to private services employment, as it was above the benchmark level in both years. The most interesting feature of the structural change in public services is the top-down convergence to benchmark employment shares in many NMS, unlike the bottom-up convergence in the private services for some countries (Romania, Poland, Slovakia, Latvia and Hungary) and even divergence in others (Slovenia, Czech Republic, Cyprus, Bulgaria, Lithuania, and Estonia). In countries with the highest initial share of employment in public services, the number of employed decreased substantially over the 1995-2005 period, due to a vast over-employment and hard budget constraints that could not support overmanned public sector. Only Romania, Slovenia, Czech Republic and Cyprus started below the benchmarks of public services employment in 1995 and have been more or less rapidly converging upwards to the reference levels. Malta displays a different pattern also in public services, diverging further from the benchmark. Even though the actual employment share in public services was and remains below the EU15 average, adjusted for the level of development in the CEECs, some of these countries still maintain disproportionately large shares of employment in public services, reflecting over-employment in these services in the past. As pointed out above, the former socialist economies stressed equality and social protection as the ultimate goals of welfare society, regardless of the efficiency of public spending on public services. In contrast, Figure 6 compares the dynamics of employment shares in the two distinct classes of services with the average employment shares in the EU15. It reveals that four countries (Hungary, Latvia, Slovakia, and Poland) converged towards the EU15 employment share in private services but diverged from it in the public services. On the contrary, four other NMS (Slovenia, Czech Republic, Lithuania, and Estonia) moved towards the EU15 average in public services employment share but deviated along the private services dimension. Two contrasting cases are Romania and Bulgaria, where the former made progress in both service groups, while the latter retreated along both dimensions of services. As expected, Malta and Cyprus deviate from the CEECs patterns with the above-average employment share in private services relative to the EU15. The changes in public and private services employment in most NMS in the period 1995-2005 brought about more symmetric disparity in both dimensions (a move towards dashed symmetry line), reducing either relative weakness in private services (Poland, Slovakia, Hungary and Latvia) or relative shortcoming in public services (Slovenia and Czech Republic). Figure 6: Distance between actual and EU15-average employment shares in private and public services (in % points), 1995 and 2005 CO CD O O 0 CO o lt\tFee T ' Ik/IT S.2 Q. S-O to ■. & S W 00 lti un i- O CD u o h d n § 3 si rn eu Ž ^ d4 n k n a B lr roW d n D lr roW uE E L U E n e QC --A "D < ob £ S yu ti n g o s I e ■s Ii ° 5 O a) S (d kr d o 2 ^ 01 d on n n ae n p od col •Š ]p Em > ^ « g •H S3 m B f d e e Nw o ( rs gsr i % r di fe rc on ^ s s s 0 o 1 ° sre G ge 2 3 f iw ° -6 3 > clII aI v n, * 2 O id u0 o0 S2 avir r p y o e Q b baT K 0 0 2 k n a B d d n u b d n a g n D d n fi u E E L K U E n e ci iv vre es sf fo pu oa or g -d b u s e d n G of s n o is sc e r gn n o s d n ai alu o h n w O cur 4.3.4. Difficulties in explaining services growth in NMS The estimations of explanatory variables of employment growth in three differentiated categories of services reflect heterogeneity of services and bring some further insight into the investigation of explanatory variables of services employment share. In particular, they reveal that the transformation of mixed services is the most transition specific. It deviates from theoretical expectations and patterns in developed economies in regard to income growth impact. The indicators of transition reforms and respective institutional changes have a statistically significant effect on employment growth in public administration and in mixed services, while they are much less relevant for private services. In spite of some inconsistencies among various specifications of the model, the results robustly show that determinants for private services employment are more similar to those valid for developed economies. Lower explanatory power of specification for private services, however, calls for additional determinants that may provide more refined results. Differences in the role of GDP per capita in various subsets of services employment share require further research. One of the reasons of the poor explanatory power of macroeconomic determinants in general may lie in a shadow informal economy, which is estimated to have a much larger weight in transition economies than in developed economies (Schneider, 2002). In 2001/2002 the average size of the shadow economy amounted to 16.7% of "official" GDP in 21 OECD countries and to 38.0% in 22 transition countries. According to International Labour Organization (ILO, 2002) the bulk of new employment in recent years, particularly in transition economies and in services has been in informal economy. There has also been increasing flexibilization and informatization of production and employment relationships in the context of global competition and ICT, resulting in rapidly internationalized and diversified value chains and expansion of outsoutrcing. Rigid labour regulation and less flexible working places additionally increase informal employment.58 The average size of the shadow economy labour force (in percent of the working age population) in 1998/99 was estimated at 15.3% in seven OECD countries and 30.2% in 22 transition countries. An increasing burden of taxation and social security contributions coupled with rising state regulatory activities have been the driving forces of growth and size of informal economy labour force. While shadow/informal economy might be indirectly (and with a time lag) reflected in the value added and GDP, services employment data might be persistently underestimated (downward biased).59 58 ILO. Decent work and the informal economy. Report VI. International Labor Conference 2002. ILO http://www.ilo.org/wcmsp5/groups/public/---dgreports/---dcomm/---webdev/ documents/meetingdocument/wcms_069040.pdf 59 For example in retailing, cleaning, catering, etc. Overall, the empirical analysis and the discussion of results reveal an important influence of the transition process on the determinants of services employment in NMS in the period 1995-2007. Nevertheless, the findings on some variables should be understood as inconclusive and only preliminary due to various reasons, such as data deficiency, particularly short-time series and large amount of missing data. Further, the present study includes only 10 NMS while a broader set of transition economies could increase variations among countries and also enable better insight into the transition specific determinants. On the other hand, the comparison with EU27 or OECD countries in disaggregated service industries could improve the understanding of traditional macroeconomic variables in NMS. We may also assume that inconsistent results for the explanatory power of productivity differences in the case of private services are related to the problems in measuring services output, which are even more relevant in transition economies. Further, it has been pointed out that on average 30% of market services output is deflated by unacceptable or biased methods; the share is the biggest in business services and financial intermediation,60 while in hotels and restaurants it is rather limited. Variations across countries are even larger (Inklaar et al., 2008). The areas proposed for further analysis (e.g. shadow/informal economy, improvements in productivity measures, increased number of countries in the sample, longer data series) may in the future lead to more conclusive results concerning the determinants of services employment in NMS. In any case, the observed differences in determinants of services employment among sub-sets of services in NMS call for a disaggregated analysis. 60 Nearly a half of output, suggesting that measured trends could not accurately reflect the actual developments in prices and quantities. 5. CHALLENGES OF SERVICES DEVELOPMENT AND SERVICES EMPLOYMENT IN NMS Since the launching of market reforms, the transition economies have seen considerable structural changes. The service sector in particular underwent rapid transformation, as it used to be largely neglected under the previous socio-economic system when most services were provided by the public (state) sector or internalised in large manufacturing companies. With the change of the political system and particularly with the introduction of market mechanisms, private suppliers of services in the NMS experienced a rapid growth driven by privatization, liberalization and deregulation of the economy. New services, not available in the past, were introduced by local and foreign suppliers. Gradually, the supply and range of services provided to final and intermediate consumers increased. Public services grew much slower than private services, since the former accounted for a high share of total employment relative to income level already at the outset of the reforms. These stylized facts were confirmed by our empirical investigation, which shows that NMS have in the period 1995-2005 importantly narrowed the gap in services employment relative to the EU15. The catching up process differed substantially across countries and so did the trajectories of the two groups of service activities, owing to different starting positions and the efficiency in implementing the reforms. However, NMS still lag significantly behind the EU15 in private and public services development, with Romania and Bulgaria experiencing particularly large gaps. In the period 1995-2005, the NMS employment in private services accelerated faster than in public services with the legacy of huge over-employment at the outset of reforms. The convergence analysis supports more rapid catching-up of NMS towards advanced EU economies in private services than in public services. At the same time, the convergence in private services seems to have accelerated in the period 2000-2005, which is most likely related to the implementation of reforms prior to the accession. 5.1 VARIETY OF CATCHING UP PATTERENS: TOP-DOwN AND BOTTOM-UP The convergence analysis of the NMS was complemented with the Chenery-type benchmark analysis of employment shares in public and private services for a given level of development, based on the data for market economies. In the period 1995-2005, the NMS employment shares in public and private services converged to the reference level of market economies, however, the patterns of individual countries differ considerably. Only Slovakia, Latvia and Romania substantially narrowed the gap to the reference point in private services by 2005, while Estonia, Lithuania, and Slovenia slightly increased it. In contrast to the bottom-up trajectory of private services, the evolution of public services employment in the NMS towards the benchmark could largely be denoted as top-down. In 1995, all NMS with the exception of Romania, Slovenia, and Czech Republic recorded higher share of public services employment than would have been expected from their income level. Even though the actual employment shares in public services in all NMS were and remain below the EU15 average, adjusted for the level of development, Baltic States and Hungary maintained a disproportionately large share of employees in public services also in 2005. Finally, the comparison of the employment levels of public and private services in 2005 shows that the gap of the NMS behind the EU15 remains larger in private services than in public services, suggesting that a longer period of time is needed to adjust private services to market economies benchmarks and that structural transformation is far from complete. The preliminary examination of the qualitative elements of private and public services' transformation in the NMS indicates that despite notable improvement in productivity, value added per employee in private services remains below 50% of the average EU15 productivity, with Cyprus, Slovenia and Malta being the notable exceptions. However, these results should be interpreted with caution, because of methodological inconsistencies and data deficiency in measuring value added and comparisons across countries. The performance in public services based on the selected indicators converged to the EU15 average, mainly due to good starting conditions. In the period 1995-2005, the majority of the NMS improved the performance of public services relative to the EU15 average, while four NMS have deteriorated it. In view of the achieved progress in employment structure and a large gap in private services productivity, the challenge for NMS remains how to simultaneously boost employment and efficiency in private services, while curbing employment in public services without jeopardising their performance. It is becoming evident that more profound changes are needed to dismantle the legacy of over-employment in the public services. These refer to re-defining the role of public services in economy, increasing their responsiveness to the demands of other institutional sectors and citizens, introducing the accountability of public service providers for the quality of services and improving the efficiency of public services. Given the aggravation of budget positions with the ongoing crisis, the analysis of efficiency in public services spending appears to be an important issue of further research in most countries. 5.2. TRANSITION MATTERS AND INCREASES SERVICES HETEROGENITY In order to complement the convergence and benchmarking analysis of public and private services employment in NMS, the econometric analysis of the main determinants of services growth has also been made. Based on the findings of the studies for developed market economies, we expected that a similar set of major determinants could substantially explain the growing share of services employment in NMS. Besides, we attempted to test the validity of transition-related explanatory variables (FDI inflows, transition reforms indicator and governance indicator) resulting from the major changes in regulatory framework and institutional set up triggered off by market-oriented reforms. The estimation of determinants of services employment growth in NMS in the period 1995-2007 brings about somewhat surprising results. Basically, it turns out that GDP per capita and government expenditure fail to explain the increasing share of services employment in NMS, while only the productivity gap between services and manufacturing appears to be a statistically significant determinant. Besides, transition reforms exert statistically significant influence and slightly improve the explanatory power of previous specifications. Although some of the above results are unexpected in the framework of stylized facts on structural change, they earn more credibility when perceived through the lens of the convergence patterns in two major service groups. Here, specifically the over-employment in public services in these countries at the start of the transition needs to be mentioned and hence their disproportionate share in total employment relative to income level. Taking into account this feature, it becomes more plausible that income growth per se could not reveal positive impact on public services employment share. In fact, its impact was negative and statistically significant, which may have resulted in a poor explanatory power of GDP per capita for total services employment. However, this was the most important determinant of the increasing share of private services employment in NMS in the period 1995-2007. Distinction among public, private and mixed services provided some further insight into the explanatory determinants of services employment share. In particular, it reveals that the transformation of mixed services is most transition specific. These services show deviations from the theoretical expectations and patterns in developed economies in regard to income growth impact, as the coefficient is significantly negative. The indicators of transition reforms and institutional change are more relevant for public administration and mixed services than for private services. While transition reforms and EU accession speeded up convergence in public administration, they appear to be insufficient for catching up of employment in market services, where NMS experience the largest lag behind old member states. These findings do not suggest that standard explanations of services growth and stylized facts of structural changes are not valid for NMS. All in all, the results based on available data rather indicate that the transition process and related features disrupt the explanatory power of standard variables for services growth in NMS. Disaggregation of results for service sub-sectors indicates that past development could be the most probable explanation of such an outcome. However, a caveat applies to deficient data and short time period, which may have considerably influenced the results and should be interpreted with caution. Refinement of data, longer time series and introduction of additional factors that may have an influence is proposed as a potential extension of research in the future. 5.3. SERVICE OUTSOURCING AS A CATALYST FOR PRODUCTIVITY IMPROVEMENTS AND RESTRUCTURING OF PUBLIC AND PRIVATE SERVICES Possible explanation for relative over-employment in public services on one hand and underdeveloped private services on the other hand may be ascribed to public sector's insufficient use of contracting with outside service providers. Services like security and maintenance of buildings, legal services, sanitary inspection, computer maintenance, information system maintenance, accounting services, recruitment and skills management, are still performed in-house in many public sector firms, agencies and ministries, even though it would sometimes be more efficient to outsource them to an independent provider. In the process of such delegation of activities outside the public sector, there would also be a statistical reallocation of employment from public to private services. Provided that the services outsourced are paid competitive market price, the resulting decline of employment in the public sector could reduce overall operating costs and increase efficiency of public services. Larger employment due to privatization of service provision would increase competition in private services and employ resources more efficiently than before. In addition to the normative question of what role government should assume in providing services, it has also raised the positive question of what determines government privatization decisions in practice. There are at least two accounts of government privatization decisions. One view, which focuses on transaction costs, looks by analogy to the private sector "make or buy" decision (e.g. Williamson, 1985; Hart, Shleifer and Vishny, 1997). In this account, privatization is dictated by efficiency considerations. An alternative view, advanced by Boycko, Shleifer and Vishny (1996) among others, emphasizes the private benefits to politicians of keeping service provision inside the government. This view holds that privatization tends to occur only in response to external pressure, such as citizen discontent or tight budgets. Needless to say, service outsourcing is no clear-cut panacea for improvement in employment structure and productivity in NMS' service sector. There are important trade-offs in every make-or-buy decision that require thorough ex-ante comparison between potential benefits (due to cost savings, exploitation of economies of scale, and vendor's specialization) and costs (due to loss of control, higher organizational costs, potential hold-up problems, security issues). Services for which it is more difficult to write and administer performance contracts are less likely to be privatized. The relationship between potential benefits and costs is greater for larger and more urban regions, which presumably have a more readily available pool of external providers. Empirical evidence suggests that large cities make the greatest use of services provided by specialised private suppliers, and are the least likely to rely on in-house services (Levin and Tadelis 2007). Blank (2000) offers additional arguments conditions for beneficial transfer of services from public sector to private suppliers. The more that one believes measurable standards of quality can be observed in the social service field, the stronger the argument for government regulation of the private sector rather than government management or ownership. The more that one believes that nongovernmental agencies can provide credible signals of quality in a particular market, because of their long-term reputational concerns, or because of the nature of the market, the less the role for government ownership or management. But when standards are difficult to observe or when the recipient is not the agent who makes the decisions, government ownership may be preferable. The importance of the ability to measure and verify the correctness of supplier's services was proven vital in numerous cases.61 5.4. FOSTERING QUALITY AND INNOVATION IN SERVICES For consumers, quality of services delivered is of vital concern and this holds especially for public services. However, continuous quality improvements cannot be achieved in the absence of innovation and this is the area where NMS can progress farthest in their catching-up process. Le Grand (2007) lists four distinct models of delivering higher quality and more efficient public services: trust, where managers and employees in public services are trusted by general public to deliver a high quality service efficiently; command and control, where those workers are contractually demanded to perform specific tasks and deliver services; voice, where service consumers continuously convey their views and requirements about the quality of the service directly to service providers; and choice and 61 One example is British rail infrastructure provider Railtrack and its successor Network Rail that were unable to manage relations with independent suppliers to whom they outsourced maintenance activities through long-term contracts. After a series of railroad accidents, Network Rail decided to gradually insource maintenance activities. Public services providers in NMS as well face similar difficulties. competition, where users can select the service from the preferred among many service providers. Each of these models has its virtues and disadvantages and cannot be applied to all situations. The trust approach promotes morale, and since little monitoring is required, is relatively cost efficient. However, it has a major limitation: it assumes that service providers are exclusively motivated by the desire to provide exactly the services that customers need, and that they have no self-interest aspirations. The key benefit of command and control approach is that it is very effective in the short run. Despite its short run efficiency, in the long run targets and performance management suffer from the demoralisation and demotivation of those in charge of service delivery, the distortion of priorities, and the incentive for gaming behaviour of various kinds, ranging from straightforward fiddling of the figures to more subtle ways of meeting the target by changing behaviour in undesirable ways. The voice approach directly acknowledges users' wants and can be rich in useful information. The demerit of the voice approach is that the individual has to depend for a response on the goodwill of the person to whom they are complaining. This is also a delicate mechanism for quality enhancement since it offers little or no direct incentives for improvement to the fraudulent or self-interested professional or manager. Unlike the trust model, the choice and competition approach makes self-interest and altruism to serve the public good. If the money follows the choices of consumers, then the providers of better services will gain resources while those that provide inferior service in terms of quality or price will lose. Unlike the command and control model, it gives freedom and autonomy to professionals and managers, encouraging them to engage in innovation and creativity, and with no outside authority continuously telling them what to do. Unlike voice, in a world where choice and provider competition is the norm, users dissatisfied with the general quality of the service they can get from one provider have the opportunity to go to another who can provide them with a better service. The problem with the choice and competition model is that in some service sector, competition is hard to provide, customer-supplier relationships are fraught by information asymmetries and providers may choose only the users who are cheaper or easier to serve. Empirical evidence (see Le Grand 2007) shows that in most situations, the best performing services are those whose delivery systems incorporate substantial elements of choice and competition. Properly designed, such systems will deliver services that are of higher quality, more responsive, more efficient, and more innovative than ones that rely primarily upon trust, command-and-control, or voice. In general, services as compared to manufacturing suffer less from the traditional Schumpeterian concern that market power is a prerequisite for innovation. This implies that policy makers should strongly emphasise the general business framework and regulatory reform. NMS should also improve conditions which are conducive to effective entry into the service sector. Only then could consumers benefit fully from competitive forces and increased choice. Regulations need to be consistent and designed so as not to distort investment decisions and steps taken to remove excessive regulatory burden that gives rise to under-investment, low firm entry rates and inhibited competition. There are various ways in which support of innovation in services other than R&D could be conceived. These include policy measures to support clusters and innovation networks rather than trying to subsidise individual firms. Various institutions and incentive structures could also be put in place to foster continuous life-long learning in service sector firms. Last but not least, through public procurement, government can take on the role of a demanding customer with a view to developing competitiveness in services through innovation. Given the size of the government, including the many different public authorities, public procurement could represent a far more prominent driving force for renewal and innovation in services than is the case at present. Moreover, effective public procurement schemes could alleviate or postpone pressures in NMS to adopt rapid and politically sensitive modernization of public services by means of privatization, public-private partnerships and greater resort to outsourcing. 5.4. CONCLUSIONS Discussion of the patterns of public and private services transformation and growth in the NMS during the period 1995-2007 has revealed some new findings that are relevant also for policy shaping. It is safe to conclude that as much as further progress in private services in the NMS seems urgent from the perspective of catching up and building a globally competitive society, the transformation and modernisation of public sector services are of equal importance, particularly in view of its modest performance and trends towards increased social inequality in some countries. The NMS should move faster in incorporating innovation as a crucial element of strengthening the private services and improving governance of the public services. Exploiting the innovation potential in public services, which has been largely neglected so far, may also trigger off the innovation in private services and enhance the interfaces between the two, to the benefit of both groups of services (e.g. formal and informal networks of public and private services providers). New technologies bring enhanced dynamism into private and public services and enable the introduction of new and more efficient services (e-banking, e-commerce, e-government, e-health, and e-education). Nevertheless, there are other ways in which innovation can contribute to better performance. Policy learning accumulated in the use of innovative undertakings is of major importance and should be incorporated into pro-innovation policy actions (Mar-roto and Rubalcaba 2008). In doing so and in policy shaping, the NMS should not only imitate more developed EU economies but introduce innovative mechanisms that best suit the absorption capacity of public and private actors and the broad cultural context of their economy and society. To this end, more emphasis and resources should be assigned to the research of services-related issues, an area hitherto fairly ignored in NMS. REFERENCES D'Agostino, A.; Serafini, R. and Ward, M. (2006): Sectoral Explanations of Employment in Europe: The Role of Services. Working paper series 625. European Central Bank. Afonso, A., Shucknecht, L. and Tanzi, V. (2005) Public Sector Efficiency: An International Comparison. Public Choice, 123(3-4): 321-347. Barro, R.J. and Sala-i-Martin, X. (2004) Economic Growth. Cambridge (MA): The MIT Press. Baumol, W. (1967): Macroeconomics of unbalanced growth.The anatomy of urban crisis. American Economic Review, 57 (3), pp. 416-426. Baumol, W. J. (2001), "Paradox of the Services: exploding costs, persistent demand", Chapter 1 in "The growth of Service Industries - The Paradox of Exploding Costs and Persistent Demand" edited by T. ten Raa and R. Schettkat, Edward Elgar Publishing Limited. Blank, R. M. (2000) When Can Public Policy Makers Rely on Private Markets? The Effective Provision of Social Services, Economic Journal 110(462), str. C34-49. Burger, A.; Stare, M. (2010): Public and private services transformation in the CEECs, The Service Industries Journal, 30(4), pp.479-496. Clark, C. (1957): The Conditions of Economic Progress. London: MacMillan. Csaba, L. (2005) Regulation and Public Sector Development: A Post-Transition Perspective. Post-Communist Economies, 17(2): 137-152. EC (2004) European Competitiveness Report (Commission staff working document, SEC(2004)1397). Brussels: European Commission. Elfring T. (1989): The Main Features and Underlying Causes of the Shift to Services. Service Industries Journal, Vol. 9, No. 3, pp. 337-356. Eschenbach, F.; Hoekman, B. (2006): Services Policy Reform and Economic Growth in Transition economies 1990-2004, Review of World Economics , 142(4), pp.746-765. Francois, J.F.; Reinert, K. (1996): The Role of Services in the Structure of Production and Trade: Stylized facts from a Cross-Countries Asia Pacific Economic Review 2, pp.35-43. Fuchs, V. R. (1968), The service economy. New York: National Bureau of Economic Research. Gadrey, J. (2007): National Economies and the Service Society: the Diversity of Models', in Bryson, J.; Daniels, P. (eds): The Handbook of Service Industries. Cheltenham: Edward Elgar, pp. 45-61. Gemmel, N. (1982): Economic Development and Structural Change: The Role of the Service Sector. The Journal of Development Studies, Vol. 19, No. 1, pp. 37-66. Gemmel, N., ed. (1993): The Growth of the Public Sector. Aldershot: Edward Elgar. Gershuny J., Miles, I. (1983): The New Service Economy: The Transformation of Employment in Industrial Societies. London: Frances Pinter. Ghibutiu, A. (2000) Serviciile si dezvoltarea: De la prejudecati la noi orizonturi (Services and Development: From Prejudices towards New Insights). Bucharest: Editura Expert. Havlik, P.; Leitner S. and Stehrer R. (2008): Growth Resurgence, Productivity Catching-up and Labour Demand in CEECs. WIIW Statistical Reports 3. Vienna: WIIW. Hunya G. (2000): Integration through foreign direct investment: Making Central European Countries competitive. Cheltenhamn, (UK), Northampton (MA): Edward Elgar. ILO (2009) LABORSTA - database of labour statistics. Retrieved October 21, 2008, from the http://laborsta.ilo.org/. Inklaar, Robert, Timmer P. Marcel and Bart van Ark, 2008, Data for Productivity Measurement in Market Services: An International Comparison, International Productivity Monitor, No.16. http://www.csls.ca/ipm/16/IPM-16-inklaar-timmer-vanark-e.pdf Kostecki, M. (1996) Business Options in the Service Sector of the Transition Economies: A Framework for Inquiry. In M. Kostecki and A. Fehervary (Eds.) Services in the Transition Economies, 3-28. Oxford: Elsevier Science Ltd. Landesmann, M. (2002) Structural Change in the Transition Economies, 1989-1999. In United Nations - Economic Commission for Europe (Ed.) Econom ic Survey of Europe, 2000, 95-117. Geneva: United Nations. Landesmann, M., Vidovic, H. (2006): Employment Developments in Central and Eastern Europe: Trends and Explanations. WIIW Research Reports, No. 332, Vienna: WIIW. Landesmann, M.; Vidovic, H.; Ward, T. (2004): Economic Restructuring and Labour Market Developments in the New EU Member States. WIIW Research Reports, No. 312, Vienna: WIIW. Le Grand, J. (2007) The Other Invisible Hand: Delivering Public Services through Choice and Competition. Princeton, NJ: Princeton University Press. Levin, J. and Tadelis, S. (2007) Contracting for Government Services: Theory and Evidence from U.S. Cities," NBER Working Paper 13350, National Bureau of Economic Research. Maroto, A. (2009): Relationships between Services, Growth, Productivity and Welfare: The State of the Art. ServPPIN Project, WP 1, http://www.servppin.com/ Marroto, A. and Rubalcaba, L. (2008) Structure, size and reform in the public sector in Europe. In P. Windrum and P. Koch (Eds.) Innovation in Public Sector Services. Cheltenham: Edward Elgar. Messina, J. (2004): Institutions and Service Employment: A Panel Study for OECD Countries. Working paper series 320. European Central Bank http://www.ecb.int/pub/pdf/ scpwps/ecbwp320.pdf Mickiewicz, T; Zalewska, A. (2002): Deindustrialisation: Lessons from the Structural Outcomes of Post-Communist Transition. William Davidson Working Paper, No. 463. http:// www.bus.umich.edu/KresgeLibrary/Collections/Workingpapers/wdi/wp463.pdf Nusbaumer, J. (1987), The service economy: lever to growth. Boston: Kluwer. OECD (2000): Employment in the Service Economy: A Reassessment. Employment Outlook 2000, Chapter 3. Paris: OECD. OECD (2007) OECD Handbook Measuring Education and Health Volume Output (Chapter 1). Unpublished manuscript. Retrieved on September 16, 2008 from http://www. oecd.org/dataoecd/1/21/38597417.pdf OECD (2008) OECD Glossary of Statistical Terms. Retrieved on September 16, 2008 from http://stats.oecd.org/glossary/detail.asp?ID=1812 Peneder, M. (2003): Industrial Structure and Aggregate Growth, Structural Change and Economic Dynamics 14, pp. 427-448. Raiser, M., Schaffer, M. and Schuchardt, J. (2004) Benchmarking Structural Change in Transition. Structural Change and Economic Dynamics, 15(1): 47-81 Riddle I. D. (1986): Service-Led Growth: The Role of the Service Sector in World Development. New York : Praeger. Rojec, M. (1998) Foreign Direct Investment's Effect On Restructuring And Efficiency Upgrading In Slovenia's Manufacturing Sector, Journal of International Relations and Development, Vol. 1, p. 46-64. Rowthorn, E.R.; Wells, R.J. (1987): Deindustrialization and Foreign Trade. Cambridge : Cambridge University Press. Rubalcaba, L.; Di Meglio, G. (2009): The new service economy in Europe: New evidence and challenges. Deliverable 1.1 - WP 1: New service developments and the contribution of private and public services to growth and welfare. ServPPIN project. http:// www.servppin.com/ Rutkowski, J.J. and Scarpetta, S. (2005) Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union. Washington: The World Bank. Schettkat, R. and Yocarini, L. (2003): The shift to services: A review of the literature, IZA Discussion Papers, 964, Institute for the Study of Labor (IZA). Schlicht, E. (2004): Estimating the Smoothing Parameter in the so-called Hodrick-Pres-cott Filter, Discussion Paper 2004-2, Dept. Economics, University of Munich. Schneider F. (2002) The Size and Development of the Shadow Economies of 22 Transition and 21 OECD Countries. Discussion Paper No. 514. Institute for the Study of Labor. Stare, M. (2007) Service Development in Transition Economies. In J. Bryson and P. Daniels (Eds.) The Handbook of Service Industries (168-185). Cheltenham: Edward Elgar. UNCTAD (2004) World Investment Report 2004: The Shift Towards Services. Geneva: United Nations Conference on Trade and Development. WIIW (2008) WIIW Handbook of Statistics 2008 (Edition 127, November 2008). Vienna: The Vienna Institute for International Economic Studies. Wolfl, A. (2005): The service economy in OECD countries, STI Working Paper, 2005/2, Paris: OECD World Bank (2009a) World Development Indicators Online Database. Retrieved on October 15, 2008 from http://ddp-ext.worldbank.org/ext/DDPOQ/member.do?method=g etMembers&userid=1&queryId=135. World Bank (2009b) Worldwide governance indicators, 1996-2007. Retrieved on October 16, 2008 from http://info.worldbank.org/governance/wgi/resources.htm. index Accession process ■ 15, 33, 45, 51, 53 Afonso, A. ■ 14, 30 Annual growth rate (AGR) ■ 20 Barro, R. J. ■ 21 Baumol, W. J. ■ 36, 45 Baumol's 'cost disease' ■ 35 Benchmark ■ 8, 9, 25, 26, 29, 32, 44, 53 Blank, R.M. ■ 55 Bulgaria (BG) ■ 13, 19, 23, 24, 26, 27, 31, 32, 33, 38, 51 Catching up ■ 7, 8, 13, 20, 21, 22, 34, 51, 53 CEECs, Central and East European Countries ■ 7, 8, 9, 12, 13, 14, 15, 17, 22, 27, 30 Chenery-type benchmark analysis ■ 26, 51 Chenery-type cross-country regression ■ 25 Clark, C. ■ 35 Cluster analysis ■ 22, 24, 25 Control of corruption ■ 30, 40 Convergence analysis ■ 51 bottom-up ■ 27 patterns ■ 53 process ■ 8, 15, 21, 22, 28, 31, 44, 52 6-convergence analysis ■ 21, 22 top-down ■ 29 Csaba, L. ■ 15 Cyprus (CY) ■ 17, 19, 26, 27, 28, 29, 31, 32, 52 Czech Republic (CZ) ■ 14, 17, 19, 23, 24, 26, 27, 29, 31, 32, 52 D'Agostino, A. ■ 36, 37, 41 Deregulation ■ 7, 46, 51 Di Meglio, G. ■ 20 E EBRD transition index ■ 37, 38, 41, 42, 43, 44, 45 Econometrics ■ 36, 37, 38, 53 Education (M) ■ 12, 30 Efficiency ■ 8, 9, 14, 21, 27, 29, 30, 37, 46, 51, 52 Efficient services ■ 8, 57 e-banking ■ 57 e-commerce ■ 57 e-education ■ 57 e-government ■ 57 e-health ■ 57 Elfring, T. ■ 36 Employment ■ 7, 8, 9, 11, 12, 13, 14, 15, 17, 19, 20, 23, 24, 25, 27, 29, 34, 35, 36, 37, 38, 41, 42, 43, 44, 45, 49, 50, 51, 53 female ■ 36 growth ■ 8, 9, 13, 17, 20, 35, 36, 37, 38, 41, 42, 43, 45, 46, 49, 53 over-employment ■ 27, 46, 51, 52, 53 rate ■ 39, 40 services employment share ■ 17, 24, 25, 27, 29, 39, 40, 41, 43, 46, 49, 53 structure ■ 13, 19, 21, 25, 37, 52 unemployment rate ■ 39, 40, 41 Engel's law ■ 35 Enlargement ■ 33 Eschenbach, F. ■ 41 Estonia (EE) ■ 14, 17, 19, 23, 24, 26, 27, 31, 32, 33, 52 EU KLEMS Database ■ 17, 19, 20, 22, 23, 26, 28, 31, 32, 40, 42, 43 EU15 ■ 7, 9, 15, 17, 19, 20, 21, 22, 23, 25, 27, 28, 29, 31, 33, 36, 52 EU27 ■ 22, 50 Euclidian distance ■ 28 European Commission (EC) ■ 12 European union (EU) ■ 9, 15, 21, 22, 23, 24, 33, 38, 45, 51, 53, 57 Eurostat ■ 31, 32, 40, 42, 43 Financial intermediation (J) ■ 45 Foreign direct investment (FDI) ■ 7, 22, 39, 40, 41, 42, 43, 53 Francois, J.F. ■ 35 Fuchs, V.R. ■ 36 Gadrey, J. ■ 24 Gemmel, N. ■ 11, 36 Gershuny, J. ■ 36 Ghibutiu, A. ■ 13 Government effectiveness ■ 30, 40 Government expenditure ■ 39, 41, 45, 53 Havlik, P. ■ 13 Health and social work (N) ■ 12, 31, 33 Hodrick-Prescott Filter (HP Filter) ■ 39, 40, 41 Hoekman, B. ■ 41 Hotels and restaurants (H) ■ 12 Hungary (HU) ■ 14, 17, 19, 23, 24, 26, 27, 31, 32, 52 Hunya, G. ■ 42 I Income distribution ■ 31, 33 Income elasticity of demand ■ 35 Income level ■ 13, 25, 28, 51, 52, 53 Infant mortality rate ■ 31, 33 Information and communication technologies (ICT) ■ 14, 36, 42 Inklaar, R. ■ 30, 50 Innovation ■ 13, 14, 57 Institutional setting ■ 11, 36 Institutions ■ 11, 15, 36, 37, 38, 40 International Labor Organization (ILO) ■ 17, 19, 23, 26, 31, 32, 40, 42, 43 K Kostecki, M. ■ 15 L Landesmann, M. ■ 13, 21 Latvia (LT) ■ 17, 19, 23, 24, 26, 27, 31, 32, 33, 52 Le Grand, J. ■ 55, 56 Leitner, S. ■ 13 Levin, J. ■ 55 Liberalization ■ 40, 46 Life expectancy at birth ■ 31, 33 Lithuania (LV) ■ 17, 19, 23, 24, 26, 27, 31, 32, 33, 52 M Malta (MT) ■ 17, 19, 26, 27, 31, 32, 52 Market economy ■ 8, 13, 14, 25, 34 Market reforms ■ 9, 13, 15, 37, 51, 53 Maroto, A. ■ 35, 36 Marroto, A. ■ 57 Measurement problems ■ 11, 12, 50 Messina, J. ■ 35, 36, 41 Mickiewicz, T. ■ 13, 15, 37 Multicollinearity ■ 41 N New member states (NMS) ■ 7, 8, 9, 17, 19, 20, 21, 22, 25, 26, 27, 28, 29, 31, 33, 37, 38, 41, 44, 46, 49, 50, 51, 52, 53 Nusbaumer, J. ■ 36 O Organization for Economic Co-operation and Development (OECD) ■ 36, 49, 50 Other community, social and personal services (O) ■ 12, 31, 33 Outcome ■ 12, 34, 54 Output ■ 7, 12, 13, 30 Outsource ■ 54, 57 P Peneder, M. ■ 39 Performance indicators ■ 30 Planning system ■ 14, 34 Poland (PL) ■ 17, 19, 23, 24, 26, 27, 31, 32, 33 Private households with employed persons (P)■12 Privatization ■ 37, 40, 46 Productivity gap ■ 41, 44, 45, 53 Productivity growth ■ 35, 37 Proxy indicators ■ 30, 33 Public expenditures ■ 11, 12, 15, 38, 41, 44 management ■ 8, 11 Public administration and compulsory social security (L) ■ 12, 30, 33 R Raiser, M. ■ 13, 15, 25 Real estate, renting and business activities (K) ■ 12 Regression model ■ 38 Regulatory quality ■ 30, 40 Reinert, K. ■ 35 Riddle, I.D. ■ 36 Rojec, M. ■ 42 Romania (RO) ■ 13, 17, 19, 23, 24, 26, 27, 31, 32, 33, 51, 52 Rowthorn, E.R. ■ 36 Rubalcaba, L. ■ 20, 57 Rule of law ■ 30, 40 Rutkowski, J.J. ■ 13, 15 S Sala-i-Martin, X. ■ 21 Scarpetta, S. ■ 13, 15 Schaffer, M. ■ 13, 15, 25 Schettkat, R. ■ 41 Schlicht, E. ■ 39 Schneider, F. ■ 49 Schuchardt, J. ■ 13, 15, 25 Serafini, R. ■ 36, 37, 41 Service output ■ 29 Services growth ■ 35, 36, 49, 53, 54 mixed ■ 8, 20, 38, 39, 40, 43, 45, 49, 53 non-market ■ 7, 12, 14 private ■ 7, 8, 9, 11, 12, 15, 17, 20, 22, 24, 25, 26, 27, 28, 29, 31, 38, 39, 40, 41, 42, 43, 45, 49, 50, 51, 52, 53, 57 public ■ 7, 8, 9, 11, 12, 14, 15, 17, 20, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 36, 38, 41, 42, 43, 44, 45, 51, 52, 53, 57 Shadow economy ■ 49, 50 Shucknecht, L. ■ 14, 30 Slovakia (SK) ■ 17, 19, 23, 24, 26, 27, 31, 32, 52 Slovenia (SI) ■ 17, 19, 23, 24, 26, 27, 28, 29, 31, 32, 52 Socialist economies ■ 8, 27, 37 Standard variables ■ 54 Stare, M. ■ 14, 17 Stehrer, R. ■ 13 Summary indicator ■ 30 T Tadelis, S. ■ 55 Tanzi, V. ■ 14, 30 Tertiarisation ■ 8, 13, 35, 37 Tertiary educated ■ 30 Timmer, P.M. ■ 30, 50 Transformation ■ 7, 8, 9, 11, 12, 13, 15, 27, 28, 29, 34, 49, 51, 52, 57 Transition process ■ 7, 8, 9, 11, 15, 17, 33, 34, 37, 41, 46, 50, 54 reforms ■ 7, 9, 38, 45, 46, 53 Transport, storage and communications (I) ■ 12 U Union density ■ 37 United Nations Conference on Trade and Development (UNCTAD) ■ 22 Urbanization ■ 36 V Value added per employee ■ 14, 29, 30, 31, 40, 52 van Ark, B. ■ 30, 50 Vidovic, H. ■ 21, 37 Vienna Institute for International Economic Studies (WIIW) ■ 17, 19, 23, 31, 32, 42, 43 Volume measures ■ 12 W Wages ■ 35, 37 Ward, M. ■ 36, 37, 41 Ward, T. ■ 21 WB governance indicator ■ 38, 39, 40, 41, 42, 43, 44, 45 Welfare-state ■ 23, 36 Wells, R.J. ■ 36 Wholesale and retail trade (G) ■ 12, 45 Wolfl, A. ■ 29, 36 World Bank ■ 26, 31, 32, 40, 42, 43 World Bank's Worldwide Governance Indicators database ■ 30 World Development Indicators ■ 31, 32, 42, 43 Y Yocarini, L. ■ 41 Z Zalewska, A. ■ 13, 15, 37