To catch up with more advanced economies, Serbia urgently needs to improve the functioning of its labour market. This report reviews labour market trends and the principle challenges to labour market policy, making a series of recommendations.
Despite many reforms, new business growth until now has been far too slow to compensate for job losses elsewhere. Recent reforms of labour law should be followed up by further efforts to improve the climate for business and productive work. Labour regulations must be flexible, but they should also be enforced more consistently. For all this to happen, it is essential that an effective social dialogue can take place and that it encompasses expanding and declining segments of the labour market.
OECD Reviews of Labour Market and Social Policies SERBIA A LABOUR Y IVIT MARKET IN TRANSITION UCT PR O D R L AW ENT L A BOU N LOYM NSITIO N NSITIO N EMP YMEN T TR A Y TR A R L AW TR A N S ITIO E M PLO CTIVIT A BOU OUR L AW P ROD U YM ENT L R L AW EMPLO B IT Y L A IT Y ENT PR O D UCTIV L A BOU DUCTIV LOYM ITION IT Y Y MENT NSITIO N PR O N EMP SITION TR A N S PR O UCTIV EMPLO ENT T RA NSITIO ENT T R AN TIVIT Y ANS ITION PR O D EM PLOYM TIVIT Y TR A MP LOYM W PR ODUC PL OYME NT TR BOU R L AW PL OYME NT T L AW E L A BOU R L AW PR ODUC IT Y L A BOUR A BO UR L A OUR L AW EM Y ME NT L A OUR L AW EM ION E R L AW UCTIV ENT L CTIVIT Y L AB EMPLO CTIVIT Y L AB R ANSIT L A BOU NS ITION PR O D EM PLOYM NSIT ION P RODU ANS ITION IT ION P RODU UCT IVIT Y T ION P RO T TR A ITION T TR A IT Y TR TR A NS PR O D R ANSIT E M PLO YMEN IT Y TR A N S E M PLO YMEN DUCTIV UR L AW EMPLOY MENT R L AW PLOYM ENT T ENT UCTIV L AW PR O A BOU L AW E M EMPLO YM PR O D IT Y L A BOUR R L AW IT Y L A BO ENT L BOUR ITION DUCTIV A BOU DUCTIV LOYM IT Y L A N PR O YM ENT L N PR O IO N EMP PR O D UCTIV IT Y TR A N S ANSIT ITIO NSITIO EMPLO ION TR A NS T TR A IT Y TR R ANSIT UCTIV ITION YMEN ENT T PR O D TR A N S R L AW E M PLO PR O DUCTIV L AW EM PLOYM OU R L AW IT Y L A BOU R L AW IT Y L A BOUR T L AB UCTIV A BOU UCTIV Y MEN PR O D YM ENT L ITION PR O D ION EMPLO ION EMPLO ENT T R ANS ANSIT ENT T R ANSIT PLOYM IT Y TR PLOYM L AB OUR L AW EM PR O DUCTIV Y L AB O UR L A W EM R L AW CTIVIT L A BOU T R ANSIT ION P RODU T YMEN E M PLO IT Y L A ROD UCTIV ION P R L AW IT TR A NS M PLO Y MENT T L A BOU L AW E Y MEN R L AW EM IT Y L A BOUR N EMPLO Y L AB OU UCTIV ITIO CTIVIT IT Y PR O D TR A N S UCTIV U PR O D TIVIT Y R AN SITION PR O D W PR ODUC LOYM ENT T R L AW LOYM ENT UR L A MP A BOU MP A BO BOUR L AW E YM ENT L BOUR L AW E ITION ENT L UCTIV IT Y L A EMPLO UCTIV IT Y L A TR A N S N EM PLOYM ITION PR O D AN SITION ITION PR O D DUC TIVIT Y N PR IT Y TR NSITIO NS TR A NS NSITIO YMEN T TR A DUCTIV UR L AW EMPLOY MENT W PR O T TR A IT Y TR A E M PLO PR O A BO UR L A M PLO YMEN LOY MEN UCTIV AW L AW E PR O D Y L AB OUR L R L AW Y L AB O ENT L BOUR N EMP CTIVIT A BOU CTIVIT LOYM IT Y L A NSITIO R AN ION P RODU ENT L ION P RODU N EMP PR O D UCTIV Y TR A PLOYM ENT T T R ANSIT EMP LOYM ENT T R ANSIT TR A NSITIO T TR A NSITIO N ODU CTIVIT OUR L AW E M ANSIT ITION PLOYM TIVIT Y YMEN AW PR Y L AB IT Y TR TR A N S UR L A W EM PR ODUC L AW E M PLO AB OUR L ION P RODU CTIVIT PR O DUCTIV IT Y L A BO BOUR ENT L BO IT Y L A R L AW IT Y L A R ANSIT R L AW UCTIV PR O D UCTIV T L A BOU PR O D UCTIV N EM PLOYM LOYM ENT T TL A BOU SITION PR O D PR ODU Y MEN EMP N Y MEN T TR A ION EMPLO R ANSIT NSITIO R L AW YMEN R L AW PLOYM ENT T Y TR A L A BOU EMPLO E M PLO ENT L A BOU T T R L AW EM ODU CTIVIT PR ODUC TIVIT Y R ANS ITION Y L A BOU R L AW LOYM E M PLO YMEN R L AW L A BOU RL AW PR T TR A NS ITION UCT IVIT Y T P RODU CTIVIT ITIO N EMP IT Y L A BOU YM ENT L A L A BOU E M PLO YMEN PR O D TR A NS ITION TR A N S PR O D UCTIV EMPLO L OECD Reviews of Labour Market and Social Policies Serbia A LABOUR MARKET IN TRANSITION ORGANISATION FOR ECONOMIC CO-OPERATION AND DEVELOPMENT The OECD is a unique forum where the governments of 30 democracies work together to address the economic, social and environmental challenges of globalisation. The OECD is also at the forefront of efforts to understand and to help governments respond to new developments and concerns, such as corporate governance, the information economy and the challenges of an ageing population. The Organisation provides a setting where governments can compare policy experiences, seek answers to common problems, identify good practice and work to co-ordinate domestic and international policies. The OECD member countries are: Australia, Austria, Belgium, Canada, the Czech Republic, Denmark, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Japan, Korea, Luxembourg, Mexico, the Netherlands, New Zealand, Norway, Poland, Portugal, the Slovak Republic, Spain, Sweden, Switzerland, Turkey, the United Kingdom and the United States. The Commission of the European Communities takes part in the work of the OECD. OECD Publishing disseminates widely the results of the Organisation’s statistics gathering and research on economic, social and environmental issues, as well as the conventions, guidelines and standards agreed by its members. This work is published on the responsibility of the Secretary-General of the OECD. The opinions expressed and arguments employed herein do not necessarily reflect the official views of the Organisation or of the governments of its member countries. Corrigenda to OECD publications may be found on line at: www.oecd.org/publishing/corrigenda. © OECD 2008 OECD freely authorises the use, including the photocopy, of this material for private, non-commercial purposes. Permission to photocopy portions of this material for any public use or commercial purpose may be obtained from the Copyright Clearance Center (CCC) at email@example.com or the Centre français d'exploitation du droit de copie (CFC) firstname.lastname@example.org. All copies must retain the copyright and other proprietary notices in their original forms. All requests for other public or commercial uses of this material or for translation rights should be submitted to email@example.com. FOREWORD – 3 FOREWORD This report was prepared by the Directorate for Employment, Labour and Social Affairs under the auspices of the OECD’s Centre for Co-operation with Non-Members (CCNM). It is part of a series of similar policy reviews devoted to central and eastern European countries, designed to permit comparison with other countries in the region as well as with OECD member countries. Previous reviews in the series concerned Poland (1993), the Czech Republic and Hungary (1995) and the Slovak Republic (1996) – before their accession to the OECD – and subsequently Slovenia (1997), Bulgaria (1998), Romania (2000), the Russian Federation (2001) and the Baltic States (2003). The main report was written by Anders Reuterswärd, while Vladimir Gligorov, acting as a consultant to the Secretariat, wrote the annex. Delegates from the OECD’s Employment, Labour and Social Affairs Committee discussed the review with representatives of the Ministry of Labour and Social Policy, the Ministry of Employment and Regional Development, Serbian labour market associations and other Serbian experts at a meeting held in Belgrade on the 12 December 2007. John P. Martin Eric Burgeat Director Director Directorate for Employment, Centre for Co-operation Labour and Social Affairs with Non-Members SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 TABLE OF CONTENTS – 5 TABLE OF CONTENTS Executive Summary ............................................................................................. 9 CHAPTER 1. SERBIA’S OVERDUE TRANSITION ...................................... 15 Introduction ........................................................................................................ 15 A problematic legacy ......................................................................................... 21 From under-employment to labour shortage ...................................................... 23 Population ageing and international migration ....................................... 24 High unemployment and declining employment .................................... 27 The employment structure needs to change faster ............................................. 31 Private-sector growth and its impact on job conditions .......................... 34 Self-employment: from casual own-account work to small enterprises .................................................................................. 37 Concluding remarks ........................................................................................... 39 CHAPTER 2. THE INSTITUTIONAL FRAMEWORK FOR EMPLOYMENT ................................................................................ 41 Introduction ........................................................................................................ 41 The industrial relations system is under pressure .............................................. 41 Labour taxes, compliance and unreported work ................................................ 45 Unreported work – and under-reported wages in formal jobs ................. 46 The labour inspectorate should be strengthened ................................................ 49 The problem of unreformed enterprises ............................................................. 50 Phasing out social ownership .................................................................. 51 Employers often disregard the Labour Law – but seldom use the flexibility it allows ........................................................ 56 Working time .......................................................................................... 56 Employment protection legislation (EPL) ............................................... 56 Concluding remarks ........................................................................................... 60 SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 6 – TABLE OF CONTENTS CHAPTER 3. THE NATIONAL EMPLOYMENT SERVICE AND ITS PROGRAMMES ......................................................................... 61 Introduction ........................................................................................................ 61 Who are the registered unemployed? .................................................................. 64 New principles for registration from 2007 ......................................................... 71 A difficult challenge for job counsellors ............................................................ 72 Profiling and selection of priority groups .......................................................... 73 Active labour market programmes (ALMPs) .................................................... 75 Concluding remarks ........................................................................................... 79 Annex: Rights and Risks: Labour Market Challenges in a Post-self-managed Economy ................................................................ 81 References .......................................................................................................... 99 List of boxes Box 1.1. Sources of statistics about the labour force and employment ............. 18 Box 2.1. The labour inspectorate ....................................................................... 49 Box 3.1. Unemployment insurance .................................................................... 70 List of figures Figure 1.1. Serbia’s social product (SP) and GDP ............................................. 16 Figure 1.2. GNI per capita in transition countries .............................................. 16 Figure 1.3. Employed persons in 2001-2007 ..................................................... 19 Figure 1.4. Serbia’s population in 2007 by age and labour force status ............ 25 Figure 3.1. Registered job seekers: stock data ................................................... 65 Figure 3.2. The unemployed by duration of unemployment .............................. 65 Figure 3.3. Women as percent of the employed and the unemployed ............... 66 Figure 3.4. Employed and unemployed men: age distributions ......................... 67 Figure 3.5. Employed and unemployed women: age distributions .................... 67 Figure 3.6. Registered unemployed persons in August 2007: distribution by gender, age and education attainment ............................. 68 Figure 3.7. Employed and unemployed persons in October 2005: distributions by gender and education attainment ................................... 69 Figure 3.8. Registered job seekers: monthly flow data ...................................... 71 Figure 3.9. Registered vacancies: monthly flow data ........................................ 72 SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 TABLE OF CONTENTS – 7 List of tables Table 1.1. Household budgets in 2006 by income decile .................................. 20 Table 1.2. Unemployment: duration and individual backgrounds ..................... 28 Table 1.3. Employment/population ratios in 2006 for the working-age population and selected age and gender groups. OECD and selected Balkan countries ...................................................................................... 29 Table 1.4. Labour force status of the population in 2004 to 2007 ..................... 30 Table 1.5. Employment by economic sector and ownership ............................. 33 Table 1.6. Relative wages and employment by economic sector, 2005-2006 ............................................................................................... 34 Table 1.7. Employed persons by tenure in the current main job ........................ 35 Table 1.8. Weekly working time ........................................................................ 37 Table 1.9. Employed working age and older persons by job status and main sectors ...................................................................................... 38 Table 1.10. Self-employed persons aged 15-64 without employees by place of work .................................................................................................... 39 Table 2.1. Employed persons and contributors to pension funds ...................... 47 Table 2.2. Number of employees and value added (VA) per employee in registered enterprises by ownership as reported in company statements for 2005 ................................................................................. 52 Table 2.3. Incidence of unpaid wages ................................................................ 53 Table 2.4. Relative wages by sector and ownership .......................................... 55 Table 2.5. The strictness of the employment protection legislation (EPL) ........ 57 Table 2.6. Reported job separations ................................................................... 60 Table 3.1. Active labour market programmes: estimated spending in 2006 ...... 76 Table 3.2. Participants in active labour market programmes ............................. 77 SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 EXECUTIVE SUMMARY – 9 EXECUTIVE SUMMARY Serbia has achieved high rates of economic growth since 2000, and its transition to a competitive market economy is well underway. But precious time was lost in the 1990s, with the result that GDP per capita is now among the lowest in Europe. The country suffers from a large labour surplus, and the restructuring of the economy is still in a phase when many jobs are bound to disappear at the same time as new ones are created. In this situation, it is crucial to establish the best possible conditions for creating more new jobs on market conditions. Key policy objectives must be to foster greater flexibility in employment, a consistent and equitable application of labour law and a transparent job market. The on-going transformation of the National Employment Service should be completed with a further shift of emphasis from administrative register functions towards counselling and job-search assistance to the unemployed. This report first reviews the labour market trends in the years after 2000 and the principal challenges to labour market policy. It then considers the available policy instruments, beginning with the institutional framework including industrial relations, labour law and issues about compliance and enforcement. This is followed by an assessment of the role of the public employment service and its different programmes. In all considered policy areas, Serbian policy makers have drawn significant lessons from experience in OECD economies and other transition countries. However, it is argued below that Serbia, in order to catch up economically, should aim to make its labour market more flexible than those of most EU countries. An annex takes a longer historical perspective. It considers the experience of the former Yugoslavia’s peculiar labour market model and draws some policy lessons from the ways in which the different successor states have dealt with this legacy. Major employment reductions have occurred Chapter 1 observes that employment declined in every year from 2001 through 2006, followed by a small recovery in 2007. Over the period as a SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 10 – EXECUTIVE SUMMARY whole, major employment reductions occurred in big enterprises and peasant farming, while net job creation was observed mainly in small non-farm family businesses. The survey-based unemployment rate peaked at almost 22% in 2005 and 2006. But it fell to just under 19% by October 2007, reflecting the combination of economic recovery and a shrinking labour force. The working-age population is set to decline further as a result of population ageing and net emigration. However, Serbia still has large potential labour supplies because only about 50% of the working-age (15-64) population is employed. Employment rates are low by OECD standards for most demographic groups, especially youths, women and the elderly, although the situation is comparable in several countries in south- eastern Europe. Moreover, a significant but declining part of employment concerns subsistence farming and other low-productive forms of self-employment, which often are informal. In some part, the recent decline in total employment appears to reflect a growing reluctance among parts the population to accept the most low-paying types of work. Disappointing employment growth in small firms The economic transition has brought profound changes in the structure of the job market. Almost two-thirds of employment is now in the private sector, where labour turnover is on average much higher than in the public sector, and the average size of enterprises has declined significantly. However, a worrying weakness in Serbia’s recent labour market performance has been the anaemic growth of employment in new small firms. Non-farm self-employment still plays a modest role by international standards. The authorities have sought to facilitate business start-ups by streamlining administrative procedures, but international comparisons show that these are still relatively cumbersome. It should be a high priority for the authorities to remove these barriers and make the public administration more helpful to small firms. Chapter 2 finds that Serbia’s formal labour market regulations are broadly appropriate and in accordance with international standards. But this observation has limited consequence as long as implementation is weak. For the same reason, it can be difficult to determine to what extent various provisions in the law might in fact be too rigid if they were to be rigorously enforced in future. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 EXECUTIVE SUMMARY – 11 Serbia needs more flexible EPL than most EU countries have Numerous rules in Serbia’s Labour Law have been aligned with EU practice, which is relatively stringent on certain points such as employment protection legislation (EPL). As a result, EPL is now less flexible in Serbia than in most of the other transition economies, and much less flexible than in, for example, the English-speaking OECD countries. However, international experience shows that low and medium-income countries need more flexible EPL rules than those typically found in continental western Europe. Serbia should simplify its rules about dismissing workers and remove unnecessary restrictions on the use of temporary job contracts. The Labour Law cites collective bargaining as the preferred method for wage-setting, and it stipulates strict rules for recognition of trade unions and employer associations. But in reality, trade unions and collective agreements have little impact outside the public sector and large firms. Where unions do not exist, the law permits employers to decide on wages and working conditions. Social dialogue should involve potential winners and losers Given Serbia’s history and the difficulty of its economic and labour market reforms, policy making must involve an element of social dialogue. For this purpose, tripartite Councils have been established at the national and regional levels, in which the existing labour market associations are represented. However, in policy matters of great importance for economic development, the social dialogue remains insufficient if it does not encompass both growing and declining segments of the labour market. In order to achieve this, it appears important to consider possible ways of giving voice to groups that are not well represented in the tripartite councils, as for example the workers of small firms. The range of participants in such discussions should be flexible and not necessarily linked with a status as collective-bargaining partner. Strengthen tax enforcement and the labour inspectorate Better enforcement is needed in several policy areas of relevance to employment, including labour law, social insurance, business and labour taxation, and occupational health and safety. The report identifies three types of compliance problems: i) many jobs are informal; ii) some formal enterprises are so low-productive that they cannot afford to fulfil their SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 12 – EXECUTIVE SUMMARY legal obligations; and iii) some rules, such as those on working time, are widely disregarded even in profitable formal enterprises. According to OECD experience, the most important instrument for formalising informal employment is usually the tax administration and its ability to promote good accounting standards. Enhanced tax enforcement should be coordinated with the on-going development of Serbia’s official registers of enterprises and entrepreneurs, as well as with that of the social insurance administration. Enterprises should not be allowed to stay in business if they cannot fulfil their obligations. The remaining socially- owned firms, which on average have the worst record, should be privatised or liquidated as planned. In order to fulfil its functions, the labour inspectorate needs more resources and stronger legal powers, for example in terms of its right to inspect various types of business and to decide about sanctions. Client-oriented employment services Chapter 3 is devoted to the modernisation of the National Employment Service (NES) and its programmes. These reforms are guided by national strategy documents that take account of international experience, reflecting Serbia’s cooperation with the European Union and various countries. Several strategy papers emphasise the need to develop client- oriented services to job seekers and employers. In Serbia as elsewhere, varying views have been expressed about the need to increase spending on the more expensive types of active labour market programmes, especially training. With about 0.1% of GDP devoted to such programmes in 2007, their annual participant intake corresponded to barely 5% of the stock of registered unemployed persons. To develop its main service functions, the NES needs to modernise the job-seeker and vacancy registers. These should concentrate on the actual information needs for job search and filling vacancies. An important step in this direction was taken in 2007, when the principles for registering clients as unemployed were changed and numerous persons not seeking jobs were eliminated from the register. Similarly, vacancy registers should focus on the job openings for which employers are actually seeking candidates. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 EXECUTIVE SUMMARY – 13 Limited capacity for counselling justifies a focus on benefit recipients and motivated clients Given a large stock of registered unemployed clients, and the likelihood of continued large inflows in the near future, the NES must prioritise. It recently introduced a system for profiling of clients, and individual employment plans are developed for those deemed most employable. Following current practice in many countries, unemployment benefit recipients are asked to sign mutual-obligations agreements with the employment service. However, only about 10% of the clients are eligible for such benefits, and even for them, the mutual obligations have questionable legal consequence because the benefit payments are in arrears (by four months in 2007). According to the profiling outcomes in 2007, a majority of the registered unemployed were considered as employable only after participation in substantial active programmes (training or job subsidies), if at all. But despite a recent increase in programme spending, budgetary and efficiency considerations make it unrealistic to expect such expensive programmes to play more than a very marginal role in resolving Serbia’s unemployment problem. In this situation, the best use the NES can make of its limited budget is to concentrate on the basic employment service functions, especially filling vacancies, job-related individual counselling and related group activities. Existing self-service systems for job information, which are accessible to anyone, should be developed, while available resources for counselling and monitoring of individual job-search efforts should target benefit recipients and the most motivated among the other registered clients. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 1. SERBIA’S OVERDUE TRANSITION – 15 CHAPTER 1 SERBIA’S OVERDUE TRANSITION Introduction After a decade of economic and political distress, Serbia achieved a significant turnaround in the years that followed its democratic breakthrough in 2000. Successive governments in office since then have sought to restore lost momentum in the long-delayed transition to a market economy. A process of privatisation and economic restructuring has resumed, supported by numerous pieces of new legislation including a labour law from 2001, modified in 2005, and measures to facilitate enterprise start-ups. The social safety-net has been partly modernised within the narrow budget limits that can be afforded. As a framework for many of these reforms, the country has adopted a National Employment Strategy and a Poverty Reduction Strategy that take account of international experiences. This report reviews the key issues of labour market policy from an OECD perspective. Serbia is currently one of the poorest countries in Europe, notwithstanding over 40% real GDP growth between 1999 and 2006, corresponding to over 5% per year. Given a slightly negative population trend, the increase in GDP per capita was nearly 50% over the same period, or 6% per year. But this strong growth performance must be considered against the background of an extremely depressed starting point in 1999, the year of the Kosovo war when much of the formal economy had ceased to function, comparable only to a similar low point in 1993 (Figure 1.1).1 1. GDP estimates are available only from 1997. Previous measures, called “social product” or “gross material product”, excluded several service sectors as “non-economic”. It would also be relevant to compare GDP or GNI trends in purchasing power parities (PPPs), but such data do not appear to be available over the considered period. In a preliminary estimate for 2007, the World Economic Outlook (IMF, October 2007) estimated Serbia’s GDP per capita at USD 7 265 in PPPs and USD 5 397 in nominal terms. This was higher than in Albania, Bosnia, Bulgaria and Macedonia in nominal terms; but in PPPs, only Albania among the neighbouring countries had a lower estimated per capita GDP than Serbia. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 16 – CHAPTER 1. SERBIA’S OVERDUE TRANSITION Figure 1.1. Serbia’s social product (SP) and GDP Alternative trend estimates for 1985-2006 at constant prices 3,000 2,500 2,000 Billion 2006 dinars 1,500 1,000 SP, old estimates 500 SP, recent estimates GDP - Source: Secretariat calculations based on the NBS Statistical Bulletin, various issues; www.nbs.yu accessed 8 May 2007. Figure 1.2. GNI per capita in transition countries EMU = 100 60 50 40 1990 30 1995 2000 20 2005 10 0 * For 1990: UN data, not fully comparable. This also applies to 1995 in the case of Serbia and Montenegro. Source: WDI, Atlas method, comparing GNP at current US dollars with the European Monetary Union average = 100. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 1. SERBIA’S OVERDUE TRANSITION – 17 The necessary restructuring and modernisation of the economy is still at an early stage. Serbia lags several years behind other transition economies in central and eastern Europe, of which a majority have experienced relatively steady growth since the early or mid-1990s. Economies like Hungary, the Slovak Republic and the Baltic States, which in 1990 appeared about as productive as Serbia, are now generating much higher per-capita incomes. Serbia has also fallen behind the previously less developed economies of Poland, Romania and Bulgaria (Figure 1.2).2 Such comparisons must be treated with caution because of definitional problems and different starting points in the transition, but they suggest that Serbia has a potential to catch up on EU countries if it can adapt and make better use of its resources. The country’s experience of previous industrial development, as reflected in its human capital, infrastructure and institutions, is now largely out-dated, but elements of this legacy could still be advantageous if used in an innovative manner. For example, there is no doubt that the inherited former Yugoslav institutions of self-management and social ownership were economically unsustainable (see Annex). The still-existing remnants of social ownership must be phased-out, as planned, but this defunct system leaves in its wake an acquaintance with decentralised management. In the emerging new institutional framework for entrepreneurship and corporate governance, which must respect the distinct roles of capital owners and managers, relevant parts of this experience may well be useful. Aggregate employment has probably declined in every year from 2001 through 2006. Figure 1.3 shows official summary estimates using different data sources (cf. Box 1.1), which suggest an average employment reduction by about 2.5% per year over the mentioned period. It is unclear to what extent the overall trend would change if it were possible to take better account of the most informal and casual types of employment. But there is no doubt that major employment reductions have occurred in large and medium-sized enterprises and in peasant farming. These retrenchments have been only partly compensated for by job creation elsewhere, principally in small non- agricultural family businesses. Until 2004, some marginal net employment growth was also recorded in registered enterprises with under 50 employees, but this enterprise category has recently been stagnant in employment terms. As discussed further below, much of the recent employment reduction concerned the most low-productive 2. Figure 1.2 compares GNI, which differs from GDP by including income received from abroad (e.g. remittances) while excluding income sent abroad (e.g. profits which foreign investors take home). SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 18 – CHAPTER 1. SERBIA’S OVERDUE TRANSITION segments of the economy. Combined with the GDP growth observed since 2000, these employment trends point to a significant improvement of productivity. But while this has facilitated some export growth in recent years, it has not yet been sufficient to prevent a continued widening of Serbia’s trade deficit. Judging from the 2007 labour force survey – published in February 2008, when this report was at a final editing stage – the long-lasting employment decline may have come to an end. Between October 2006 and October 2007, employment reportedly increased by 1.0%. Box 1.1. Sources of statistics about the labour force and employment Labour force surveys (LFS) are conducted during one week in October each year. Methods and definitions have essentially followed the recommendations of the ILO and Eurostat since 2004. The 2006 survey covered 6 500 households and 17 000 persons (0.3% of the population aged 15 or more). The surveys conducted before 2004 were not comparable. For registered enterprises, the Statistical Office of Serbia publishes administrative statistics on employees and average wages. Data are collected in March and September each year from all registered enterprises with 50 or more employees and from a sample of smaller registered enterprises. Annual averages are calculated from the March and September data, while monthly updates refer to reduced samples. These statistics have been available for several decades. Statistics about registered “entrepreneurs” (self-employed persons) and their employees have until now been collected from the health-insurance authority. A new register of entrepreneurs, set up in 2006 (see Chapter 2), will probably be used for such statistics in the future. Concerning peasants – self-employed persons in agriculture – administrative data can be obtained from the farmers’ pension fund (see Chapter 2), but only the LFS gives information about unpaid family workers. The Statistical Office of Serbia regularly compiles an “employment balance” (see Figure 1.3), designed to give adequate coverage of all labour market segments while permitting longer time series than is possible with the LFS alone. It combines the above-mentioned administrative statistics about registered enterprises with health-insurance data about entrepreneurs and their employees and LFS data about peasants. Unreported employment is not covered in administrative statistics, and it is not known to what extent it may go undetected in the LFS as well. Chapter 2 makes partial estimates of informal employment by comparing LFS-employment with enrolment in the principal pension funds. The National Employment Service produces monthly statistics about registered unemployed persons, other job seekers, vacancies, hirings and job separations. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 1. SERBIA’S OVERDUE TRANSITION – 19 Figure 1.3. Employed persons in 2001-2007 3,000,000 2,500,000 2,000,000 1,500,000 1,000,000 Peasants Non-farm self-employed persons Employees of the non-farm self-employed 500,000 Employees in enterprises with up to 50 workers Employees in enterprises with over 50 workers - Source: Data about employees from semiannual surveys of enterprises, institutions and organisations in all ownership categories, with supplementary sample surveys for enterprises with up to 50 employees. The non-farm self-employed and their employees: Health insurance data. Peasant farming: Labour force surveys in October. The income distribution has remained moderately unequal, implying that a large majority of the population now have low incomes by European standards. In 2006, the average household budget per month was estimated at EUR 137 per capita (or EUR 178 per adult-equivalent consumption unit; see Table 1.1). The median value per capita was about EUR 120, with about 20% earning over EUR 200 and nearly 20% earning under EUR 60 (EUR 2 per day).3 Work income accounted for at least about half of the total in all income deciles; but in the poorest decile, almost half of the work income was earned in-kind, essentially from farms. The existence of numerous low-income households whose heads are employed – the “working poor” – often reflects a combination of low productivity and uneven working time, especially in agriculture but also among other self-employed groups and in enterprises that impose unpaid leaves or fail to pay wages regularly. 3. Source: Household budget surveys of the Statistical Office of Serbia. The concept of “household budget” corresponds broadly to the conventional concept of “disposable income”, but it includes some additional items such as consumer credits. Based on the same surveys, the Office has estimated that the Gini coefficient for household consumption declined from 0.29 in 2003 to 0.25 in 2005. As measured for income, it was estimated at 0.33 in 2002 (World Bank, 2003, p. 54). SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 20 – CHAPTER 1. SERBIA’S OVERDUE TRANSITION Table 1.1. Household budgets in 2006 by income decile Deciles Average 1 2 3 4 5 6 7 8 9 10 A. Available budget, Euros per month Per capita 33 59 78 94 109 128 150 177 217 399 137 Per adult-equivalent consumption unit* 44 78 102 123 143 166 193 226 277 502 178 By source, percent distribution Work income 51 57 56 59 62 63 60 61 64 49 58 Of which: Wages 20 32 37 40 48 48 50 50 56 41 45 Other non-farm work income 6 5 3 4 4 3 4 2 2 3 3 Money income from agriculture 3 4 2 3 2 5 2 4 3 3 3 In-kind income (mainly agriculture) 22 17 13 11 7 8 5 5 3 2 6 Pensions (old-age, disability and other) 27 28 32 29 25 25 29 26 24 18 24 Other social benefits 10 5 4 2 3 2 2 1 1 2 2 Capital income, gifts, transfers, credits and other 11 9 9 10 10 10 8 11 11 32 16 B. Consumption expenditure (Jan-Sept.) Amount per capita, Euros per month 43 69 86 95 110 120 143 160 191 307 128 Food share in consumption, percent (Engel coefficient) 54 48 45 44 43 42 40 37 35 27 38 Each decile represents 10% of the households. * The number of adult-equivalent units per household = 1 + (number of adults - 1) * 0.7 + (number of children under 15) * 0.5. Source: Data from household budget surveys submitted by the Statistical Office of Serbia. Another indication of modest living standards is that the share of food in household consumption expenditure is over 40% for most deciles, and over 25% even for the top decile (Table 1.1, Panel B). In western European countries, food typically represents around 15% of the average household’s consumption. Such household budget data must be used with caution, however, in view of the risk of under-reporting of incomes and spending, not least in the top deciles. However, the apparent absence in Serbia of a large upper-middle class with rising relative incomes can also be taken as a sign of slow economic restructuring. Judging from the experience of other transition countries, the type of wage differentiation that is based on individual qualifications can be expected to increase as the transition proceeds, which, in turn, may lead to a rise in income inequality. The rest of this chapter seeks to identify and illustrate the main challenges. After some general observations about the conditions inherited from the past, the chapter reviews available labour force statistics and seeks to illuminate how labour market and social conditions depend on specific structural problems in the economy and how they are affected by its on-going restructuring. A final section summarises the policy conclusions that will be considered in subsequent chapters. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 1. SERBIA’S OVERDUE TRANSITION – 21 A problematic legacy Serbia’s industrial structure was shaped largely between the early 1950s and the late 1970s, a period of rapid industrialisation and rural-urban labour mobility. High rates of growth were then driven by investments and job creation in socially-owned enterprises. As discussed in the Annex, studies have found that this peculiar form of ownership and the related “worker self-management” – both typical of socialist Yugoslavia – often had a distorting impact on resource allocations. On the whole, however, the current restructuring needs do not appear exceptionally difficult in the former Yugoslav economies compared with many other countries of the former socialist bloc. The transition policies followed by Slovenia and Croatia have been relatively successful. The comparisons made in the Annex between Yugoslav successor states suggest that alternative approaches to privatisation and institutional reform can be successful, but they must focus on the need for good management in enterprises. The OECD has previously observed that while Serbia’s move to embrace economic reforms appeared irreversible, the country was facing a major challenge in implementing such reforms (OECD, 2003). The present report, which focuses on labour market issues, finds evidence of further progress in institutional reform. But in terms of the restructuring of the economy and the job market, the achievements until now are uneven and somewhat disappointing. A considerable number of enterprises have not been privatised, often due to concerns about their poor performance – a fact that should make it all the more urgent to speed up the process. Such uneconomic enterprises continue to represent a financial cost to the government and a cause of market distortions that can discourage job creation in other firms. A 2003 survey of Serbian enterprises found that privatised firms were 90% more productive than socially-owned ones, while new private firms were over three times as productive.4 The analysis below of employment and wage trends up to 2006 (see Chapter 2) finds signs of continued malfunctioning in a number of public and socially-owned enterprises. The Reassessed OECD Jobs Strategy of 2006 identifies product-market competition as a key factor behind international differences in labour market performance. Anti-competitive product-market rules tend to generate rent- seeking, to reduce employment and output and to slow down restructuring. Compared with leading OECD economies, these product-market distortions and their negative employment effects were generally greater in transition countries. This conclusion also seems to hold for Serbia, judging from a study in 2002 (Center for Liberal-Democratic Studies, 2002). Competition has 4. World Bank (2004, p. 7). SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 22 – CHAPTER 1. SERBIA’S OVERDUE TRANSITION undoubtedly increased since then as a result of foreign trade and domestic adjustments, but constraints in transport and distribution systems continue to restrict local markets, as does the weight of monopolies in public utilities. Perhaps the most worrying weakness of the Serbian economy – in terms of its capacity to generate new jobs – concerns its small-business sector, which is growing at an anaemic pace. Without much more vigorous growth in new and innovative enterprises, aggregate employment will continue to stagnate or decline, which would make the necessary downsizing of uneconomic firms even more painful for the population. An OECD survey in 2002, covering a sample of small registered enterprises in Serbia, found that only about 20% of them were then under five years old.5 An encouraging sign, however, was that many managers gave positive assessments of the institutional conditions for business as reformed by 2002. The most often-mentioned obstacle to business growth was the local market’s low purchasing power. However, as shown below, job creation in small enterprises has not picked up after 2002 despite rising household incomes. In Serbia, as elsewhere in the Balkans, the business climate suffers from administrative complications inherited from the past. An important step was taken in 2004 with the creation of a new Business Register Agency, which has begun to coordinate several administrative functions that previously required contact with different authorities. From 2006, this Agency also keeps records of “entrepreneurs”, i.e. self-employed own-account workers with or without employees, and handles their enrolment in social insurance. In recognition of this reform, Doing Business in 2006 (World Bank, 2005) mentioned Serbia as the “leading reformer” amongst 155 countries. However, further reforms are needed. Notwithstanding the Business Register Agency’s goal of becoming a “one-stop shop”, Doing Business in 2008 (World Bank, 2007) found that the setting-up of a new business in Serbia still requires 11 administrative procedures and takes 23 days. The situation is similar in many neighbouring countries, but Serbia’s global ranking is now only 86th on the summary indicator “ease of doing business” and 90th on “starting a business” (this time amongst 178 countries). Even less favourable rankings were obtained for “dealing with licenses”, “registering property” and “paying taxes” (despite a low rate of business tax, see Chapter 2).6 5. OECD (2003, Annex II). 6. Following the praise for Serbia in Doing Business in 2006, the subsequent Doing Business in 2007 reported further progress, notably with respect to contract enforcement and bankruptcy procedures. But Doing Business in 2008 found no SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 1. SERBIA’S OVERDUE TRANSITION – 23 The administrative simplifications that began in 2005 should facilitate business growth when they are fully implemented, but such effects are hardly visible in available statistics for 2006. Non-agricultural self-employment in any form – registered as enterprises, “entrepreneurs” or not at all – continues to play a moderate role by international standards, accounting for about 10% of all employed persons in 2006. A closer look at data from labour force surveys (see below) suggests that this figure includes an element of irregular or casual activities, often with short working time and no fixed place of work. Some of those concerned are probably working informally, and so represent a challenge to the Business Registers Agency, social insurance and tax authorities.7 The “formalisation” of informal employment must be a priority goal. OECD experience suggests that an efficient collection of taxes and social insurance contributions is an important instrument to achieve this objective. It should be accompanied by recurrent revisions of employment and product market regulations in order to make them less onerous for formal-sector employers. Studies in many countries have found that formal businesses are more likely than informal ones to grow and to enhance productivity. Informal employers tend to have few legitimate business contacts and limited access to credits and legal protection, while informal jobs are typically insecure and low-productive.8 The following sections seek to quantify the principal challenges with the help of statistics, especially from labour force surveys, considering first the conditions of labour supply and subsequently the employment structure and some indicators of qualitative job characteristics. From under-employment to labour shortage Like many European countries, Serbia is facing at the same time a labour surplus and a prospect of future labour shortages. The current surplus includes not only the 585 000 persons who were unemployed in October 2007 – about 19% of the labour force, down from about 700 000 or 22% a year earlier – but comparable reforms in the twelve months up to June 2006, with the result that Serbia fell back significantly in ranking terms. 7. The reported numbers of self-employed persons are similar according to both the LFS and health-insurance records, but it is not known to what extent the two sources cover the same individuals. 8. See OECD (2004), Chapter 5. Negative effects of informality have been observed for example in central and eastern Europe and Mexico (Belev, 2003; Winkler, 1997). Informality has been found to contribute to low productivity growth and uneven economic development in Latin American countries (e.g. Gonzaga, 2004). SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 24 – CHAPTER 1. SERBIA’S OVERDUE TRANSITION also various forms of under-employment and “hidden” unemployment. Depending on definitions, the latter can be considered to include employed persons whose jobs are “inadequate” in one way or the other (e.g. they may involve few hours, be part-year or not utilise the individuals’ skills and qualifications), as well as inactive persons who might be able to work but are not seeking jobs. It is notable that the inactive, as estimated in LFS, are more numerous than the unemployed among women in almost all age classes and among men aged 50 and above (Figure 1.4). To sustain economic growth in a period of population ageing, Serbia needs to reverse a downward trend in labour force participation (see below). Both demand and supply-side factors are important. On the side of labour demand, economic growth should permit a gradual recovery in aggregate terms, once the lion’s share of the necessary downsizing has been accomplished in the many enterprises now slated for restructuring, privatisation or liquidation. A subsequent section probes into these developments with the help of data about the changing composition of employment. On the supply side, it is relevant to consider the main demographic parameters and the possible role of various obstacles to labour force participation, especially among youths, women with children, elderly persons and the disabled. Population ageing and international migration The latest population census in 2002 recorded 7 498 000 inhabitants in Serbia excluding Kosovo, with a declining trend since a historical peak at 7.8 million was reached during the 1990s. The actual peak number including refugees was possibly even higher. For the period after 2002, official estimates based mainly on data on births and deaths indicate a further population decline of 25 000 to 30 000 persons per year, suggesting a total population of a little less than 7.4 million in 2007. Serbia’s population is ageing. Judging from population estimates used as a frame for the LFS, the biggest population cohorts in 2007 were already 50-59 years old – an age when numerous women are about to retire while most men have just a few more years to work (Figure 1.4). About one-fifth of the inhabitants were aged 65 or more. By European standards, Serbia’s demographic imbalance is not extreme as measured by its crude birth rate of nearly 11 per 1 000 inhabitants in 2006 (after a recent increase) or by its total fertility rate of about 1.5 children per woman, close to the EU average. However, in contrast to western Europe, net migration to Serbia has turned negative and emigrants are probably younger on average than most of the recent immigrants, of whom many were refugees. The elderly population is set to grow further as a result of a projected rise in average life expectancy, which in 2005 was only 70 years for men and 75 for women. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 1. SERBIA’S OVERDUE TRANSITION – 25 Figure 1.4. Serbia’s population in 2007 by age and labour force status Males 350 Thousands of persons per 5-year age class 300 250 200 150 Inactive Unemployed 100 Employed 50 0 0-4 5-9 10-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75+ Females 350 Thousands of persons per 5-year age class 300 250 200 Inactive 150 Unemployed Employed 100 50 0 0-4 5-9 10-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75+ Source: Labour force survey; for children, estimates based on birth data. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 26 – CHAPTER 1. SERBIA’S OVERDUE TRANSITION No reliable statistics are available about international migration after 2002. By way of comparison, about 350 000 persons appear to have emigrated between the two census years of 1991 and 2002. The outflow was then compensated for by immigration, principally of refugees from Croatia (about 240 000 persons) and Bosnia and Herzegovina (about 140 000) and displaced persons from Kosovo (around 250 000), altogether 600 000 to 700 000 individuals.9 Some of them have now returned to their countries of origin or moved elsewhere, but over 400 000 including many elderly persons appear to have settled permanently in central Serbia or Vojvodina. Assuming no further large refugee movements, future migration will depend on employment and education opportunities in Serbia compared with prospects in the main receiving countries. Net emigration will probably persist for some time, even under a hypothesis that foreign countries take no further steps to facilitate immigration from Serbia. A progressive (re)integration of Serbia’s labour market in its regional and wider international context appears both desirable and inevitable in the medium term, notwithstanding many obstacles. More open borders would create new opportunities for job seekers and employers, and this would put additional pressure on enterprises to make their job openings attractive. For individuals, it would also give stronger incentives to acquire market-relevant skills, e.g. via higher education. The effects on unemployment are less certain: emigration reduces domestic labour supply, but the resulting upward pressure on wages can also reduce labour demand. Another effect of emigration consists of remittances and other foreign transfers to Serbian households, a source of income that has been important for a considerable time, amounting to around 10% of GDP.10 International experience suggests that many migrants would probably return to Serbia if its economy continues to catch up, bringing with them valuable financial and human capital acquired abroad. But it is seldom possible to predict the extent of such returns or when they might occur. The 2002 census identified 415 000 Serbian citizens with work or residence permits abroad, most often in Germany, Austria and Switzerland, while the actual number is probably higher.11 Vukovic (2005) estimated the Serbian 9. Estimates based on UNHCR data (www.unhcr.org/statistics/ accessed 30 April 2007). 10. In the national accounts for 2006, current transfers to and from Serbia corresponded to, respectively, 14% and 5% of GDP. In addition, some smaller amounts counted elsewhere, e.g. as income and capital transfers, may be due partly to emigration (Alfieri et al., 2005). These data do not include remittances that bypass formal channels through the financial sector. 11. In Germany alone, the official foreigner register identified 317 000 non-naturalised residents with Serbian or Montenenegrin citizenship at the end of 2006 (www.destatis.de accessed 6 May 2007). SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 1. SERBIA’S OVERDUE TRANSITION – 27 and Montenegrin “diaspora” (regardless of citizenship) at almost 4 million persons, of whom 1.2 million were in Bosnia and Herzegovina, over 300 000 in other neighbouring countries and 2.4 million spread across Europe, North America, Oceania and elsewhere.12 Furthermore, in the Balkan region, several non-Serbian groups that are familiar with Serbia’s language and other conditions must be regarded as potential sources of emigration to the country, especially if Serbia should achieve stronger and more sustained economic growth than some of its neighbours. High unemployment and declining employment About 22% of the working-age labour force was unemployed in 2005 and 2006, falling to 19% in 2007, with a very high proportion of long-term unemployment. Some 81% of the unemployed had been so for at least a year and 37% for over five years (Table 1.2, Panel A). By comparison, the incidence of unemployment spells longer than one year was under 40% in most OECD countries, though it was 50 to 60% in several European countries, with the Slovak Republic’s 72% as the highest incidence (figures for 2006 in OECD, 2007). With a relatively high incidence of long-term unemployment, many of those who become unemployed are at risk of permanent exclusion. This concerns numerous young persons without work experience as well as middle-aged job losers. As measured by the LFS, 40% of the unemployed are seeking their first job and the proportion who are younger than 30 is 37% – in contrast to the employed workforce, where only 15% are aged under 30. Amongst unemployed persons with work experience, over half have been dismissed or their employers have gone out of business (Table 1.2, Panel B).13 Chapter 3 considers the situation of unemployed persons in more detail with the help of the National Employment Service’s administrative statistics. 12. According to Vukovic (2005), the largest communities of Serbian or Montenegrin origin outside the Balkans lived in the United States (650 000), Germany (450 000), Austria and Canada (each 200 000) Australia (150 000) and France (140 000). 13. The distribution of unemployed persons by educational attainment was similar to that of the employed in 2006, except that the share with university education was lower. However, because the unemployed were younger on average than the employed population (median age: 34 vs. 44), their educational disadvantage was greater compared with employed persons at the same age. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 28 – CHAPTER 1. SERBIA’S OVERDUE TRANSITION Table 1.2. Unemployment: duration and individual backgrounds Serbia (average) Belgrade* Vojvodina Central Serbia except Belgrade 2005 2006 2007 2005 2006 2007 2005 2006 2007 2005 2006 2007 A. Incidence of long-term unemployment Percent of total unemployment 1 year or more 80 81 81 72 83 85 75 78 78 85 84 82 2 years or more 65 67 66 54 63 65 60 62 63 71 70 67 5 years or more 36 38 37 26 29 35 34 33 36 40 43 39 10 years or more 10 14 12 5 8 8 11 13 13 12 17 12 B. Reasons for unemployment Percent distribution First-job seekers 39 43 40 40 46 45 33 31 35 42 47 41 Previously employed 61 57 60 60 54 55 67 69 65 58 53 59 Of which: Dismissed 13 14 13 15 20 10 15 13 13 12 12 14 Firm liquidated 17 17 18 14 15 17 15 16 15 18 17 20 Seasonal work 9 9 11 7 6 7 14 16 17 7 6 9 Quits and other 22 18 19 23 14 21 23 25 21 21 17 17 Total unemployment: Thousand persons 720 693 585 140 118 97 186 165 164 394 410 324 Unemployment rate 21 21 19 20 17 14 20 18 19 22 24 20 * Some of the smaller figures are uncertain, especially for Belgrade. Source: Labour force surveys in October each year. The employment rate – employed persons as percent of the working-age population – fell from 53.5% in 2004 to about 50% in 2006, before recovering in 2007 to 51.5% (Table 1.3, first panel, and Table 1.4). For youths aged 15-24, the employment rate was only 20% – lower than in any OECD country. The corresponding rate of 55% in 2006 for prime-age women and the 33.5% reported for 55-64 year-olds of both genders are also modest by OECD standards (Table 1.3, second to fourth panels). Even for prime-age men, the employment rates given for Serbia in Table 1.4 are modest, although in this case the difference between countries is smaller. As seen in Table 1.3, countries with low employment in one population group often show low employment for other groups as well, suggesting that general economic factors may be most important.14 Econometric studies in OECD countries confirm that some general characteristics of a labour market (e.g. its business cycles, taxes and benefit systems) often have similar effects for different labour market groups, although youth employment is typically more sensitive to cyclical 14. Based on Table 1.3, a correlation coefficient of 0.62 was estimated between youths’ employment rate and that of elderly persons. The correlations are weaker (about 0.45) between prime-age women and any one of the other two groups. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 1. SERBIA’S OVERDUE TRANSITION – 29 changes. But the employment effects tend to be more asymmetric for institutions that protect the employment of “insiders” (often prime-age men) at the expense of “outsiders” (often youths and women). Such asymmetric employment effects have been associated with monopolistic product markets, too-strict employment protection legislation and centralised collective bargaining.15 Table 1.3. Employment/population ratios in 2006 for the working-age population and selected age and gender groups. OECD and selected Balkan countries Countries ranked from high to low ratios Working age 15-64 Youth 15-24 Prime age 25-54 Elderly 55-64 Both genders Both genders Women Both genders Switzerland 78 Netherlands 64 Denmark 82 Sweden 70 Denmark 77 Australia 64 Sweden 82 Switzerland 66 Sweden 74 Denmark 64 Switzerland 78 Japan 65 Canada 73 Switzerland 63 Canada 77 United States 62 United Kingdom 73 Canada 59 Austria 77 Denmark 61 Netherlands 72 United Kingdom 57 Netherlands 75 Korea 59 Australia 72 United States 54 United Kingdom 75 United Kingdom 57 United States 72 Austria 54 Czech Republic 75 Canada 56 Austria 70 Mexico 45 France 73 Australia 56 Japan 70 Sweden 44 Germany 73 Mexico 55 Germany 67 Germany 44 United States 73 Germany 49 Slovenia 67 Spain 43 Australia 71 Netherlands 47 Spain 66 Japan 40 Slovakia 70 Czech Republic 45 Czech Republic 65 Slovenia 35 Romania 69 Spain 44 Korea 64 Turkey 31 Hungary 68 Romania 42 France 62 Czech Republic 28 Japan 67 France 41 Mexico 61 Korea 27 Poland 65 Bulgaria 40 Romania 59 Slovakia 26 Spain 64 Austria 36 Bulgaria 59 Italy 26 Korea 60 Croatia 34 Italy 58 France 25 Italy 59 Hungary 34 Slovakia 59 Romania 25 Serbia 55 Slovakia 33 Hungary 57 Poland 24 Mexico 50 Serbia 33 Croatia 56 Hungary 22 Turkey 27 Slovenia 33 Poland 55 Serbia 20 Italy 33 Serbia 50 Turkey 30 Turkey 46 Poland 28 Source: OECD, Eurostat and the Statistical Offices of Romania, Serbia and Slovenia. For youths, extended education is a crucial policy goal, but international experience suggests that it can and should be compatible with high employment. It is indeed possible for young people who are engaged in full-time education and training to combine this with work (often part-time) or job search.16 15. Bassanini and Duval (2006), p. 36ff; OECD (2006), p. 130 and Table 4.1. The equations were designed to explain varying employment rates over time, between countries and for different groups. 16. It is not known how many individuals are working full-time while attending education or training in Serbia. But in the 2005 LFS, only 0.3% of the employed respondents combined education with part-time work. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 30 – CHAPTER 1. SERBIA’S OVERDUE TRANSITION Table 1.4. Labour force status of the population in 2004 to 2007 Labour force participants and employed persons as percent of the population Unemployed persons as percent of the labour force Labour force/Population Employed/Population Unemployed/Labour force 2004 2005 2006 2007 2004 2005 2006 2007 2004 2005 2006 2007 Both genders 15-19 16.3 19.1 20.1 15.3 6.6 9.1 8.1 7.2 59.3 52.0 59.8 53.1 20-24 53.9 52.1 53.7 50.3 29.2 28.1 30.4 29.7 46.0 46.2 43.4 41.0 25-29 77.5 76.3 75.1 75.9 56.1 49.9 52.4 54.4 27.6 34.6 30.2 28.3 30-34 88.9 88.5 85.6 86.4 71.9 68.3 66.0 69.9 19.2 22.8 22.8 19.1 35-39 86.8 87.9 87.4 89.3 72.5 71.3 71.2 74.9 16.5 18.9 18.5 16.1 40-44 88.6 88.1 87.3 88.4 75.5 72.7 72.5 75.5 14.8 17.5 17.0 14.6 45-49 82.9 83.1 82.7 83.3 72.1 70.3 69.0 71.2 13.0 15.4 16.5 14.4 50-54 72.5 70.8 68.1 70.4 62.8 61.3 57.6 60.9 13.4 13.4 15.4 13.5 55-59 51.8 50.0 46.8 47.9 46.2 44.5 41.7 41.9 10.7 11.0 10.9 12.6 60-64 30.1 24.6 20.6 22.3 28.2 22.5 19.6 21.4 6.2 8.3 4.8 3.7 65-69 17.4 14.1 11.2 12.0 17.3 13.9 11.0 11.9 0.8 1.1 1.8 0.8 70-74 16.6 12.4 7.7 9.5 16.5 12.3 7.7 9.3 1.0 0.7 - 2.0 75+ 9.3 8.1 5.1 6.4 9.3 8.1 5.1 6.4 - - - - 15+ 55.5 53.5 51.0 51.0 45.2 42.3 40.4 41.8 18.5 20.8 20.9 18.1 15-64 66.6 65.2 63.6 63.4 53.5 51.0 49.9 51.5 19.5 21.8 21.6 18.8 Men 15-19 18.9 22.5 24.5 19.9 8.3 12.1 10.8 9.6 56.2 46.2 56.0 52.0 20-24 58.9 58.9 63.2 54.7 33.0 32.7 39.2 34.5 44.0 44.4 38.0 37.0 25-29 85.9 83.8 82.9 84.4 67.0 61.3 60.4 63.1 22.0 26.9 27.2 25.3 30-34 95.4 96.1 95.3 93.4 82.7 80.9 79.0 79.3 13.4 15.8 17.1 15.1 35-39 93.4 94.2 94.3 95.6 82.1 82.5 79.8 84.6 12.1 12.5 15.3 11.5 40-44 94.2 95.3 92.4 93.4 85.0 82.5 80.6 81.9 9.8 13.4 12.8 12.3 45-49 90.6 91.1 90.0 92.2 82.2 81.1 76.6 81.8 9.2 11.0 14.9 11.3 50-54 83.4 82.3 79.9 80.7 73.5 73.1 69.2 70.6 11.9 11.2 13.3 12.6 55-59 70.0 69.0 65.2 65.7 61.5 60.8 58.8 56.5 12.1 11.9 9.9 14.0 60-64 41.5 36.7 30.7 32.7 38.4 33.3 29.3 31.4 7.6 9.3 4.5 4.0 65-69 23.0 18.4 14.6 16.6 22.9 18.0 14.3 16.6 0.7 2.0 2.4 - 70-74 21.9 16.8 9.9 13.5 21.7 16.6 9.9 13.1 0.9 1.1 - 3.2 75+ 13.3 12.1 7.4 10.1 13.3 12.1 7.4 10.1 - - - - 15+ 64.6 63.0 60.1 59.7 54.9 52.4 49.3 50.3 15.1 16.8 17.9 15.8 15-64 75.2 74.3 72.7 71.9 63.2 61.2 59.2 60.0 16.0 17.6 18.6 16.5 Women 15-19 13.8 15.4 15.1 10.4 5.0 6.0 4.9 4.6 63.6 61.0 67.2 55.2 20-24 49.3 45.3 43.6 45.3 25.6 23.2 21.0 24.1 48.1 48.7 51.8 46.7 25-29 68.9 68.6 66.8 66.8 45.0 38.3 44.0 45.2 34.7 44.2 34.1 32.3 30-34 82.1 80.3 76.0 80.0 60.4 54.6 53.3 61.2 26.3 32.0 29.9 23.5 35-39 80.7 82.2 80.8 83.0 63.6 61.2 63.0 65.2 21.2 25.6 22.0 21.5 40-44 82.8 80.7 82.9 83.3 65.6 62.6 65.4 69.1 20.7 22.5 21.1 17.1 45-49 75.0 75.0 75.0 74.6 61.8 59.5 61.1 61.0 17.6 20.7 18.5 18.2 50-54 62.3 59.9 56.4 60.8 52.8 50.1 46.0 52.0 15.3 16.3 18.4 14.6 55-59 34.5 32.0 29.4 31.9 31.7 29.1 25.6 28.7 8.1 9.1 13.0 10.1 60-64 20.1 13.3 12.1 12.6 19.3 12.5 11.4 12.2 3.8 5.8 5.4 3.0 65-69 13.0 10.7 8.3 8.0 12.9 10.7 8.3 7.8 1.1 - 0.9 2.3 70-74 12.2 8.6 6.0 6.3 12.0 8.6 6.0 6.3 1.2 - - - 75+ 6.7 5.4 3.5 3.7 6.7 5.4 3.5 3.7 - - - - 15+ 47.0 44.6 42.5 42.8 36.3 32.9 32.0 33.8 22.9 26.2 24.7 21.0 15-64 58.1 56.2 54.5 54.9 44.1 40.8 40.6 43.0 24.1 27.4 25.5 21.7 Source: Labour force surveys conducted by the Statistical Office of Serbia. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 1. SERBIA’S OVERDUE TRANSITION – 31 Among OECD countries, those with high youth employment rates tend to display relatively high participation in education and training as well. According to the 2006 Jobs Strategy reassessment, the key challenge facing most countries concerning youth labour market performance is that too many are neither employed nor studying.17 This also seems to hold in Serbia. Based on the Jobs Strategy, the best remedies to youth unemployment appear to be, first, to facilitate the transition from school to work by ensuring flexibility in wages, working time and other conditions of employment; and, second, to reform initial education and training (e.g. apprenticeships) so that all pupils acquire work-relevant skills before entering the labour market. In addition, youths are probably the principal “outsider” group that stands to gain much from the elimination of anti- competitive practices in product markets, where such exist. Similarly, OECD comparisons show that higher female employment need not be incompatible with the care of children or elderly relatives.18 But this will require more flexible working conditions so that family responsibilities can be accommodated with the demands of work. Policy measures to promote child-care facilities would also be useful. Regarding elderly workers, it is important to design the pension system so that it becomes attractive to extend working life rather than retire early. The employment structure needs to change faster To assess the labour market’s performance, it is necessary to consider the quality of jobs as well as their numbers. For example, aggregate employment rates are often relatively high in rural districts; but this is not a sign of superior performance, considering the different job characteristics.19 Such observations point, instead, to the role of farming as a coping strategy for otherwise jobless or under-employed persons, a situation that is not sustainable or desirable: the transformation of the economy requires a large 17. OECD (2006), Chapter 4, p. 136. 18. In most of the countries for which Table 1.3 (third column) shows relatively high female employment, fertility rates have recently been a little higher than in Serbia. By contrast, a number of countries combine below-average rates of female employment and fertility (cf. Italy, Spain, Korea and Japan). 19. In 2005, the highest agricultural shares in employment (43-44%) were recorded in Jablanica, Mačva, Kolubara and Rasina districts, which all displayed above- average total employment as proportion of the adult population (age 15+). Areas with high industrial employment shares included several Vojvodina districts with high overall employment rates, as well as districts with low or very low total employment rates (North Bačka, Pčinja, Raška and Šumadija). SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 32 – CHAPTER 1. SERBIA’S OVERDUE TRANSITION shift in labour out of the agricultural sector to more high-productive jobs in industry and services. As shown below, much of the decline in overall employment has been concentrated in farming and other small-scale own-account work, often of a casual nature. This trend appears to reflect a growing reluctance on the part of most citizens to engage in low- productive work as a last resort, notwithstanding a continued severe shortage of higher-paid jobs. This section first considers the on-going shifts of employment between economic sectors and between ownership groups, and then illustrates some of the implications with the help of indicators of pay, job status, type of job contract and working-time. This analysis is based essentially on LFS data, which cover both dependent and independent employment, but are limited to the two surveys in October 2005 and October 2006. Chapter 2 will probe further into some of the questions raised here with the help of administrative employment data, which encompass the period 2001-2006 and permit more disaggregated analysis of economic sectors. One-fifth of the 2.6 million employed in October 2006 were in the primary sector, while nearly 30% were in industry and construction and 50% in services (Table 1.5). Although the agricultural employment share had declined by 3 percentage points in 12 months, or by almost 100 000 persons, it remained greater than in any OECD country except Turkey and higher than in any EU country except Romania. Manufacturing industry achieved significant net job creation in 2006, growing by about 4% on a yearly basis. But this increase was more than outweighed by job losses in construction and public services (except health and social care). Employment increased only marginally in commerce, the biggest service sector, while the trend was flat or negative in most other services. Insofar as annual relative-wage growth can be assumed to reflect successful business, and vice versa, the largest inter-sectoral shifts in employment between 2005 and 2006 went in the expected direction (Table 1.6). Job losses were thus recorded in construction, mining and education, where relative wages declined (although only the education sector experienced an absolute decline in real wages). Strong real-wage growth occurred in manufacturing and finance (banking and insurance), which increased their employment, but also in business-oriented services, where employment declined. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 1. SERBIA’S OVERDUE TRANSITION – 33 Table 1.5. Employment by economic sector and ownership A. Thousands of persons Economic sector Total employment Of which: private ownership State, social and other ownership forms Employed Change Employed in Change Employed in Change in 2006 2005-2006 2006 2005-2006 2006 2005-2006 Total 2,631 -103 1,633 -7 998 -96 Primary 540 -97 504 -92 36 -5 Industry 612 24 352 58 260 -34 Extraction 31 -2 4 -0.4 27 -1.5 Manufacturing 519 21 345 58 173 -36 Utilities 62 5 3 0.7 59 4 Construction 159 -7 129 5 30 -12 Services 1,320 -22 647 22 672 -45 Commerce, tourism 491 5 431 23 60 -18 Transports, communications 151 -1 64 0.6 87 -2 Finance, business services 113 -0.1 76 5 37 -5 Public administration 143 -16 - -6 143 -11 Education 129 -14 9 5 120 -19 Health and social care 174 15 12 -1.5 162 17 Other services 117 -10 54 -4 63 -6 B. Percent Economic sector Total employment Private ownership by sector as percent of employment in each sector 2005 2006 2005 2006 Total 100 100 60 62 Primary 23.3 20.5 93 93 Industry 21.5 23.3 50 58 Extraction 1.2 1.2 14 13 Manufacturing 18.2 19.7 58 67 Utilities 2.1 2.4 3 4 Construction 6.1 6.1 75 81 Services 49.1 50.2 47 49 Commerce, tourism 17.8 18.7 84 88 Transports, communications 5.6 5.8 42 42 Finance, business services 4.1 4.3 63 67 Public administration 5.8 5.4 4 - Education 5.2 4.9 3 7 Health and social care 5.8 6.6 8 7 Other services 4.7 4.5 46 46 Source: Labour force survey, 2005 and 2006. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 34 – CHAPTER 1. SERBIA’S OVERDUE TRANSITION Table 1.6. Relative wages and employment by economic sector, 2005-2006 Percent and thousands of employees Sectors were ranked according to the employment change Relative wage Real-wage Number of employees (Average = 100) growth Thousands Percent in percent in change 2005 2006 2005-2006 2006 2005-2006 Utilities 125 117 2 62 11 Health, social care 105 105 9 169 10 Finance 156 162 14 42 4 Manufacturing 86 90 15 473 4 Commerce 84 84 8 306 2 Transport 111 108 6 131 1 Tourism 79 79 9 70 0 Other local services 99 98 8 89 -4 Business services 121 133 20 52 -5 Construction 110 106 4 122 -6 Extraction 122 120 7 31 -7 Public service 128 126 8 143 -10 Education 117 104 -3 126 -12 Agriculture 67 71 14 96 -20 Average/total 100 100 9 1915 -1 All figures refer to the survey weeks in October, so they are sensitive to seasonal changes, especially in agriculture. Source: Calculations based on LFS data submitted by the Statistical Office of Serbia. Private-sector growth and its impact on job conditions Private ownership is now predominant in the Serbian economy, representing 65% of total employment in October 2007. In October 2006, when the corresponding figure was 62%, the private sector included most of industry, construction, commerce, tourism and business services (Table 1.5, Panel B). But non-private ownership still accounted for one-third of manufacturing employment in 2006, over half of the jobs in transport and communications and the quasi-totality of employment in mining, utilities, education, healthcare and social care. Where non-private ownership persists, it is often linked with structural problems. Chapter 2 illustrates these problems in more detail with the help of data about output and wages in enterprises of different ownership categories. But first, the following paragraphs consider the implications that can be gleaned from LFS data about qualitative job characteristics such as job duration and working time. Public and socially-owned enterprises traditionally provide stable jobs, and in their current state of stagnation they exhibit very low labour turnover rates. By October 2006, fewer than 5% of all workers in the non-private ownership sphere had been recruited in the past 12 months, while over two-thirds had been with the same employer for over ten years (Table 1.7, Panel A). SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 1. SERBIA’S OVERDUE TRANSITION – 35 Table 1.7. Employed persons by tenure in the current main job A. In selected economic sectors, by private vs. non-private ownership Percent distribution Job tenure, All sectors Agriculture Manufacturing Construction years Total Private Non- Total Private Non- Total Private Non- Total Private Non- privat private private private <1 11 14 5 7 7 12 11 16 2 16 17 11 1-5 21 27 13 11 11 8 19 26 6 32 37 13 5-10 16 16 15 10 9 10 15 18 11 18 18 15 10+ 52 43 67 73 73 70 54 41 81 34 28 61 Total, percent 100 100 100 100 100 100 100 100 100 100 100 100 000s persons 2,631 1,633 998 539 503 29 519 345 173 159 129 30 Commerce Hotels, restaurants Transport, comm. Government (central and local) Total Private Non- Total Private Non- Total Private Non- Public Health, Edu- Utilities Other Mining private private private adm. social cation local care services <1 16 17 4 25 29 7 11 15 7 5 2 6 6 7 11 1-5 36 37 19 30 36 9 20 35 9 18 11 17 11 20 11 5-10 21 22 16 18 18 18 16 15 16 18 17 17 14 25 12 10+ 27 23 60 27 17 66 54 35 68 58 70 61 68 48 66 Total, percent 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 000s persons 407 364 43 85 67 17 151 64 87 143 162 120 37 39 13 Source: Labour force survey, 2006. B. By ownership groups in primary and non-primary sectors Percent distribution Job tenure, Private State Social or other All ownership groups* years Non- Primary Non- Primary Non- Primary Non- Primary Employees Self-employed primary primary primary primary without employees <1 17 7 5 11 4 9 12 7 11 10 1-5 34 11 14 4 10 10 24 11 24 20 5-10 19 9 17 4 12 16 17 10 16 15 10+ 30 73 64 81 74 65 47 73 49 54 Total, percent 100 100 100 100 100 100 100 100 100 100 000s persons 1129 504 681 11 281 26 2090 540 1915 338 * The primary sector accounts for 3% of the employees and 52% of the self-employed without employees. Source: Labour force survey, 2006. The column for the self-employed without employees refers to age 15-64. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 36 – CHAPTER 1. SERBIA’S OVERDUE TRANSITION Within the non-private sphere, labour turnover was particularly low in socially-owned firms, but it was not much higher under state ownership (Panel B). Under private ownership, by comparison (except farming), 17% of the workers had begun their present jobs during the past year – a proportion similar to that which characterises the whole labour market in many OECD countries. In Serbia, such a large difference in average tenure according to ownership was found in all non-agricultural economic sectors. 20 Despite the private sector’s higher labour turnover, the use of fixed-term and other non-standard types of labour contract remains exceptional in most sectors. According to the LFS, the proportion of employees whose labour contracts were not of indefinite duration declined from 13% in 2005 to only 11% in 2006, with substantially higher shares only in agriculture (42%) and construction (30%).21 Weekly working time is relatively long by OECD standards for most groups in Serbia, a fact that may partly reflect the low hourly wages. For employees of both genders as well as for the self-employed and family helpers, it is more common to work over 50 hours per week than to work under 40 hours (Table 1.8, Panels A and B). In the LFS reference weeks of 2005 and 2006, the reported “usual” working time and the actual one were both at least 40 hours for all employees except about 5% of the men and 6 to 8% of the women. Engaging employees explicitly for part-time work appears to be an almost unknown practice in public and socially-owned worksites (Panel C). In OECD countries, by comparison, on average 7.5% of the men and 24% of the women worked under 30 hours per week in 2004 (OECD labour force statistics), and it is widely recognised that working time must be flexible to facilitate the employment of students, women with children and other groups with special requirements (see Chapter 2). 20. As seen in Table 1.7, Panel B, long job tenures are common under both state and social ownership, but in non-farm sectors this is particularly the case in socially- owned firms. The sector-specific estimates of job tenure refer to all employed persons; but tenure does not seem to differ greatly between employees and the self-employed, cf. the right-most part of Panel B. The average tenure given for all employed persons in Serbia – 11 years – is similar to the corresponding averages in Hungary and Portugal, but shorter than in most OECD countries (cf. OECD labour market database). 21. Dependent employment in agriculture is found mainly in Vojvodina. It is highly variable and not well measured by the present LFS, which uses a one-week reference period in October each year. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 1. SERBIA’S OVERDUE TRANSITION – 37 Table 1.8. Weekly working time Percent distributions A. All employed adults (age 15+ ) by job status B. Employees of working age ( 15-64 ) by and usual working time. 2005 actual working time in the survey week Hours per Average Self- Family Employees Hours per week employed helpers week 2005 2006 Men Men <20 2 5 6 1 <20 1 1 20-39 7 16 19 4 20-39 4 4 40-49 62 30 16 77 40-49 70 73 50-59 10 15 18 8 50-59 9 9 60+ 18 33 40 11 60+ 13 10 Absent 4 3 Total 100 100 100 100 Total 100 100 Total 000s 1,635 429 67 1,139 Total 000s 1,134 1,082 Women Women <20 3 10 7 1 <20 1 1 20-39 13 27 31 7 20-39 6 5 40-49 69 33 24 84 40-49 78 81 50-59 7 12 13 5 50-59 5 6 60+ 8 17 25 3 60+ 4 3 Absent 5 4 Total 100 100 100 100 Total 100 100 Total 000s 1,098 134 152 812 Total 000s 807 833 C. Incidence of part-time jobs by ownership group Percent of employed persons Private Non-private Total 2005 2006 2005 2006 2005 2006 Men 9.0 8.2 1.8 2.3 6.0 6.1 Women 15.5 12.6 2.3 3.0 9.7 8.6 Source: Calculations based on LFS data submitted by the Statistical Office of Serbia. Self-employment: from casual own-account work to small enterprises On the positive side, significant job growth is under way in non-agricultural micro-enterprises (self-employed persons with a few employees), although this growth is not yet sufficient to compensate for the large job losses elsewhere (Figure 1.3). The more “formal” enterprises that are registered as companies with under 50 employees have been stagnant in employment terms since 2004. As defined in the LFS – without regard to register status – non-farm self-employment represents about 10% of total employment or slightly less than the EU average. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 38 – CHAPTER 1. SERBIA’S OVERDUE TRANSITION Self-employment in general – in agriculture and elsewhere – is characterised by great variations in working conditions. Households in the poorest decile earn much of their work-related incomes in-kind, and for all deciles together, about two-thirds of the agricultural incomes are earned in- kind (judging from the household budget survey quoted in Table 1.1). This reflects at the same time the social importance of small-scale farming and its very limited productivity. Few self-employed persons and family helpers report “normal” working time: over half of the interviewed men and 30 to 40% of the women reported usual working times in excess of 50 hours per week, while over 20% of the men and almost 40% of the women worked less than the standard 40 hours (Table 1.8, Panels A and B). One in four self- employed persons is an employer, but the larger group with no employees includes almost all farmers and more than half of the non-farm self-employed (Table 1.9; see Panel C, which summarises for the whole adult population). Many have no permanent places of work: in 2006, over half of the non-farm self-employed without employees reported working at home, in street markets, from door to door, etc. (Table 1.10). Table 1.9. Employed working age and older persons by job status and main sectors Thousands of persons Total Self-employed Self-employed Family workers Employees with employees without employees Change Change Change Change Change 2006 2005- 2006 2005- 2006 2005- 2006 2005- 2006 2005- 2006 2006 2006 2006 2006 A. Age 15-64 All sectors 2517 -57 115 5 338 -10 149 -26 1915 -26 Primary sector 435 -53 12 -1 186 -4 139 -23 97 -25 Non-primary sectors 2082 -4 103 6 151 -6 10 -3 1818 -1 Of which: Industry 609 26 17 -1 25 6 2 0 565 20 Construction 159 -6 7 2 28 0 2 1 122 -8 Commerce, tourism 489 4 57 2 51 -2 5 -3 376 7 Government, education, health, social care 445 -16 5 2 3 0 0 0 437 -18 Other services 381 -12 17 0 45 -10 1 0 318 -3 B. Age 65+ All sectors 114 -45 3 -1 73 -28 32 -14 6 -3 Primary sector 106 -44 3 -1 69 -27 31 -13 3 -3 Non-primary sectors 8 -1 - - 5 -1 0.4 -0.3 3 -0.1 C. Age 15+ All sectors 2631 -103 118 4 411 -38 181 -39 1921 -29 Primary sector 540 -97 15 -2 255 -32 170 -36 100 -28 Non-primary sectors 2090 -6 103 6 156 -7 11 -3 1821 -2 Source: Calculations based on LFS data submitted by the Statistical Office of Serbia. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 1. SERBIA’S OVERDUE TRANSITION – 39 These observations suggest that farming and other own-account work is often sporadic and used as a source of extra income or as a last resort for poor households. However, the number of households that depend on such activities is falling rapidly, probably reflecting a combination of several developments including more regular pension payments, higher and more regular wages for those with jobs, and money transfers from family members living elsewhere (in Serbia or abroad). Judging from the LFS in 2005 and 2006, persons aged 65 or more reduced their employment by as much as 45 000 persons (28%) in one year. This rapid decline mainly concerned elderly peasants, who alone account for a large part of the downward trend in total employment (Table 1.9, Panel B). At the same time, self-employment in non-primary sectors displayed a welcome shift towards the group that is creating jobs for employees, thus away from the more “casual” group without employees, and also away from unpaid family work. The number of self-employed persons acting as employers increased noticeably in construction, commerce, tourism, education and health and social care. Table 1.10. Self-employed persons aged 15-64 without employees by place of work Percent distribution in 2006 Place of work Total Men Women Farm 53 53 51 Enterprise, shop 17 16 20 At home 7 6 9 Field work, "from door to door" 9 9 7 Vehicle 5 6 1 Street, market places 4 4 5 Other 6 5 7 Total, percent 100 100 100 Thousands 338 265 73 Source: LFS data submitted by the Statistical Office of Serbia. Concluding remarks A continued forceful restructuring of Serbia’s economy and labour market in line with market signals is urgently needed, even if it leads to additional job losses in the short term. There is no lack of signs that a more positive development has begun, notably in the manufacturing sector and in small businesses. But it will probably take some time before their job creation can dominate over the negative trends elsewhere in the labour market, as the restructuring must continue in line with market signals. For SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 40 – CHAPTER 1. SERBIA’S OVERDUE TRANSITION public policy to be consistent, it will be crucial to resist the inevitable pressures from special-interest groups seeking the protection of jobs that are not economically viable. Many of the data used above are available in comparable form only for the past few years. This gives reason for caution in drawing detailed conclusions. In broad terms, however, key results reported about the speed and direction of change in the employment structure appear robust and they are consistent with what can be gleaned from administrative data. Labour market policies need to pay more attention to the situation of expanding businesses. In Serbia as in many countries, there has been a tendency for the public administration to have more formal and informal contacts with big enterprises than with smaller ones. The government has recently taken new steps to develop its policy dialogue with business and worker associations, as observed in the next chapter. In the present situation, it appears especially important to ensure that a fruitful dialogue can encompass expanding and declining parts of the labour market, and so increase the chances of achieving a consensus about both economic and social objectives. According to OECD experience, a successful social compromise must reconcile workers’ demands for income security with a high degree of flexibility in business decisions. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 2. THE INSTITUTIONAL FRAMEWORK FOR EMPLOYMENT – 41 CHAPTER 2 THE INSTITUTIONAL FRAMEWORK FOR EMPLOYMENT Introduction The Labour Law, reformed in 2001 and 2005, conforms to international practice on all essential points. It generally offers a degree of flexibility that should permit employers and workers to adjust to changing economic conditions without breaking the law. Until now, however, the application of this and other laws has been uneven across the labour market, with, on the one hand, an informal sector where few rules are enforced, and, on the other, a formal sector that includes a number of severely under-performing enterprises, which have been allowed to stay in business for a long time without fulfilling their obligations. The situation has improved recently, but both informality and unresolved problems of ownership and enterprise management continue to reduce the scope for implementing legislation as it was intended. Moreover, given a persistent labour surplus, policy efforts to enforce labour standards are not always supported by market pressures on employers to make jobs more attractive. This chapter first reviews Serbia’s institutions for industrial relations and wage-setting. It then turns to labour taxation, informal work, the labour inspectorate and the anomalies found in a number of malfunctioning enterprises, after which it considers the provisions for working-time and employment protection. Concluding remarks are made in a final section. The industrial relations system is under pressure Industrial relations are marked – as in many countries – by a de facto division between a traditionally unionised sphere dominated by large public and privatised employers, where collective bargaining is widespread, and a growing number of small private firms where trade unions are weak or absent. The Labour Law stipulates that collective agreements have precedence where they exist, but the employer can SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 42 – CHAPTER 2. THE INSTITUTIONAL FRAMEWORK FOR EMPLOYMENT otherwise issue the corresponding regulations in a “labour rulebook” or via individual contracts.1 The Labour Law’s minimum requirements apply in any case. Employers and workers are free to set up associations, but if these want to participate in collective bargaining, they must be representative according to specific criteria (Article 223ff). To be representative at enterprise level, a trade union must organise at least 15% of the workers. At the level of a sector or territory, a trade union is representative if it organises 10% of the workers, while an employer association must represent 10% of relevant employers and 15% of their employment. There are numerous trade unions at enterprise and intermediate levels. At national level, there is currently one representative employer association, the Union of Employers in Serbia. The trade union movement has historically been split along political lines, but all main associations are now independent of the state and the parties. Some lower-level unions belong to no federation, but most of them adhere to one of the two that have been recognised: the Confederation of Autonomous Trade Unions of Serbia (SSSS) and Nezavisnost (“Independence”) Trade Union Confederation. A third federation, the Association of Free and Independent Trade Unions (ASNS), is not recognised at national level.2 Collective agreements are valid for three years (Article 263) and the Law presumes that a national agreement will precede lower-level ones (Article 4).3 But in recent years, the private sector’s national-level bargaining has suffered considerable delays. During 2007, none of the nationally recognised federations had signed any general collective agreement by mid-November, and there were also relatively few sector- level agreements. When a sector agreement exists, the government can 1. The Labour Law’s Articles 1-3 and 8 permit the employer alone to issue a “labour rulebook” when there is no representative trade union, when neither party initiates collective bargaining, and when such bargaining has begun but does not lead to agreement within 60 days. 2. SSSS, formerly the only allowed trade union federation, claims 850 000 members (www.sindikat.org.yu/e_licna_karta.htm accessed 21 July 2007), but other estimates are less than half as high (e.g. Upchurch, 2006). The Nezavisnost (“Independence”) Trade Union Confederation was established in 1991 in opposition to the previous regime; it probably has 150 000 to 180 000 members. The Association of Free and Independent Trade Unions (ASNS), with perhaps 40 000 members, broke away from Nezavisnost and supports the Democratic Party. 3. Collective agreements from 2001 or earlier, including a general collective agreement, became null and void as a result of the 2005 labour law (Art. 284). SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 2. THE INSTITUTIONAL FRAMEWORK FOR EMPLOYMENT – 43 extend it to cover all workers in the sector – a provision that is commonly used only in some sectors where the state is itself a predominant employer (education, culture and social protection). The legal minimum wage was 11 094 dinars per month after tax in the second half of 2007, corresponding to about 40% of the average wage. Although this proportion is close to the average for OECD countries, lower wages are found to a significant extent in Serbia (see below). But where collective agreements apply, they stipulate higher effective minima for most job categories. According to the 2005 Labour Law (Art. 106ff), collective agreements should lay down rules for setting a “basic” wage for each job category, possible performance-related elements and other wage increases. Wage increases for seniority as well as for overtime, work on holidays and night and shift work are specified in the law at certain proportions of the basic wage.4 A reform of public-sector wages is under preparation, designed to create a new framework for evaluating jobs and competences and to encourage professional development. The private sector is expected to develop its own procedures. Wage bargaining in state and socially-owned enterprises has until now resembled that of the public administration, but Ministry officials interviewed in the course of preparing this report considered that it should become aligned with the private sector. Observed relative-wage trends in the period 2001-2006 have largely matched the different employment performance of various groups of enterprises. Judging from data about registered enterprises with over 50 employees, the strongest real-wage growth was thus recorded in non-retail trade and various other private and public services, which showed smaller-than-average declines in employment.5 The opposite results – slow real-wage growth and very large job losses over the five-year period – were notable in most industry sectors and in the primary sector. These different outcomes might be taken as a sign that the wage-setting system had the flexibility to respond in the face of changing economic conditions and labour demands by sector. However, the observed actual wage outcomes 4. While partly based on tradition, these rules about wages from 2005 are more detailed than those of the 2001 labour Law. For seniority, Article 108 stipulates a 0.4% wage premium per year of service. 5. See Table 2.4 below. In the manufacturing sector, enterprises with under 50 employees showed net job creation between 2005 and 2006 while bigger firms continued to decline through 2006. Cf. Table 1.5 above, based on LFS data covering all forms of business, which indicates a significant recovery for the manufacturing sector after 2005. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 44 – CHAPTER 2. THE INSTITUTIONAL FRAMEWORK FOR EMPLOYMENT were not always attributable to the formal wage-setting because low-performing enterprises often failed to pay agreed wages, as discussed further below (cf. Tables 2.3 and 2.4). Over the past decade, Serbia’s labour market has been conflict-prone, with many strikes directed at political decisions makers, notably concerning plans for privatisation and restructuring. In individual labour conflicts, workers and employers have too often resorted to time-consuming court procedures rather than seeking peaceful solutions (Đukić, 2006; Jasarević, 2003). Provisions for peaceful arbitration in such conflicts were included in the 2001 Labour law (Article 121), but did not have the desired effects. A new Agency for Peaceful Resolution of Labour Disputes was therefore set up after legislation in 2004. It can mediate in disputes about workers’ right to organise and to strike and about the conclusion and application of collective agreements, while offering to arbitrate in individual conflicts about dismissals and the minimum wage.6 It is probably too early to assess the agency’s achievements, but it reportedly settled 2 815 individual and 25 collective disputes between its foundation and the autumn 2007. All told, the industrial-relations system is under severe pressure. Trade unions have not been very effective as counterparts to employers who operate under competitive conditions, especially small private firms. In larger firms, where unions have more power, the worst obstacles to a meaningful social dialogue are probably on the employer side, where unresolved problems of ownership and corporate governance still prevent many managers from acting effectively as employers (see below). As these problems are now being addressed, it is essential to foster a better climate for social dialogue. This is promoted by a tripartite Social-Economic Council set up in 2001, now following legislation from 2005, and by similar councils in most of the country’s 25 districts. The national-level council decides about the level of the minimum wage. But most of its functions can only be advisory, as industrial relations are largely decentralised and each actor is responsible for its own adjustment. The urgency of economic restructuring makes it crucial that the social dialogue can engage both expanding and declining segments of the labour market. It falls upon the Social-Economic Council and the principal 6. Mediation is usually voluntary. It is mandatory if a conflict threatens essential society functions, but the parties are still not obliged to reach an agreement. In individual conflicts about dismissals and the minimum wage, no party can be forced to accept arbitration – but if they do so, they must accept the verdict (Law on Amicable Resolution of Labour Disputes). SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 2. THE INSTITUTIONAL FRAMEWORK FOR EMPLOYMENT – 45 associations as coordinators to promote such changes of the industrial-relations system that the dialogue can encompass new types of business, for example via a better representation of small private firms and their employees. Labour taxes, compliance and unreported work Serbia’s labour taxation is moderate by European standards. But the cost of social insurance makes the total tax burden regressive, which encourages informality in low-wage groups. Recognising this, the authorities have introduced a series of changes that reduce the relative tax burden on low wages, with application from 2007, of which the most important are: • The standard tax rate on wages was reduced from 14 to 12%, and taxable wages are subject to a new basic deduction of 5 000 dinars per month. (Most non-wage incomes are taxed at a 20% rate.) • Lower trigger points for an additional income tax, now charged at a rate of 10% on incomes in excess of three average wages after family deductions, rising to 15% from five average wages. (The trigger points were previously four and ten average wages, respectively.) • New exemptions from wage tax and social contributions for young workers (see Chapter 3). These can last up to three years after recruitment. • Mandatory social insurance contributions must be calculated on at least 35% of the average gross wage (previously 40%). Social insurance contributions total 35.8% of earnings.7 Employers and employees share these contributions equally while the self-employed pay the full rate. There is no basic deduction from the income on which they are calculated, but no contributions are charged on the part of the income above 7. The contribution rates are 22% for old-age and disability pensions, 12.3% for sickness and maternity and 1.5% for unemployment insurance. The total falls slightly below a European average, being lower than in France, Germany, Italy and most transition countries. But it is higher than in several of those countries that, like Serbia, finance a significant part of the corresponding expenditures via the treasury (e.g. the United Kingdom and Nordic countries; www.ssa.gov/policy/docs/progdesc/ssptw/2006-2007/europe/guide.html). In Serbia, contributions cover all the spending on sickness, maternity and unemployment insurance but less than half of the pension spending. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 46 – CHAPTER 2. THE INSTITUTIONAL FRAMEWORK FOR EMPLOYMENT five average wages. Thus, although the taxation of low incomes was reduced in 2007, it remains substantial and strongly dominated by the cost of social insurance. Corporate profits are taxed at only 10%, one of the lowest rates in Europe. While this can encourage investments, the size of the difference compared with the total tax on wages can encourage under-reporting of the latter. A somewhat higher tax rate on corporate income can be justified as a means to incite employers to declare their full wage costs and deduct them from taxable profits (OECD, 2004, Chapter 5). According to OECD experience, efficient collection of taxes and social contributions is usually the most important instrument for enforcing formal employment regulations (OECD, 2004, Chapter 5). Labour or social insurance inspectorates play significant roles in detecting undeclared work in many middle-income countries with large informal economies, and, also, in some advanced countries such as Belgium, France and Spain. Elsewhere, however, as in most English-speaking OECD countries and the Nordic region, the corresponding function is performed mainly by tax authorities. The goal should be to coordinate fiscal auditing of business costs, revenues, value added and profits. The tax agency can then use various methods for tracking expenses e.g. to verify subcontractors’ accounts. Such coordination is also crucial for a correct assessment and allocation of the combined tax liability of an enterprise, its employees and its owners. For a business as a whole, it is usually tax-efficient to declare wage costs because any undeclared wages would increase taxable profits.8 Unreported work – and under-reported wages in reported jobs To assess the implications of unreported work in labour market terms, it is pertinent to distinguish between i) deliberately unreported jobs and ii) under-reporting of wages in reported jobs. The former type can be considered as a subset of informal employment, and its incidence can be partially assessed with the help of the LFS. However, an unknown additional number of informal jobs are likely to go undetected in the LFS as well as in administrative registers. Moreover, informal employment can be considered to include activities that need not be declared, e.g. production for own consumption. It will not be attempted here to assess the full extent of informal employment. 8. Once distributed to the owners, profits are taxed at a higher overall rate than wages in most countries. Exceptions exist, e.g. when social security contributions are payable on wages but not on business earnings. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 2. THE INSTITUTIONAL FRAMEWORK FOR EMPLOYMENT – 47 A partial estimate of deliberately unreported work can be obtained by comparing employment data from the LFS – a household survey – with the statistics collected from registered enterprises and self-employed persons.9 In the autumn 2006, this gave a discrepancy of 160 000 persons or 7% of the LFS-based employment total (excluding peasants; Table 2.1). Similarly, LFS-employment exceeded the number of employed contributors to public pension funds by about 230 000 or 9 to 10%.10 A majority of the non- contributors seem to be employees working full-time.11 Table 2.1. Employed persons and contributors to pension funds Thousands of persons in 2006 Data source Employees Self- Peasants* Total Total except employed* peasants 1. LFS: Employed in age 15-64 1,915 264 338 2,517 2,179 2. Administrative employment data** 1,780 239 na na 2,019 # 3. Employed contributors to pension funds 1,621 333 333 2,286 1,954 Observed discrepancies (with possible interpretations) 1 - 3: Employed but not contributing 294 -69 5 231 226 1 - 2: Employed in informal jobs (?) 135 25 na na 160 2 - 3: Working pensioners (?) 160 -94 na na 66 na = Not available. * Including family workers. ** Refers to registered enterprises, institutions and organisations in September 2006. Bulletin ZP 21, Statistical Office of Serbia. # Not counting unemployment benefit recipients, for whom the National Employment Service pays contributions. Data from pension funds were submitted by the Ministry of Labour and Social Policy. There is some evidence that such deliberately unreported employment is declining. Similar estimates by the Statistical Office of Serbia, using a method recommended by OECD (2002) for national accounts, covered only 9. The concept “unreported production” also includes a significant element not considered here that went unrecorded through no fault of the businesses concerned (around 4% of GDP in 2005 according to Ratkovic, 2008). For example, production in registered enterprises is not always recorded due to shortcomings of the official system for collecting and manipulating such data. 10. The distribution of the discrepancy in Table 2.1 suggests that some employees contribute to the fund of the self-employed. 11. Cf. Table 1.8. Only 1% of all employees counted in the LFS worked less than 20 hours per week in 2006, while about 5% worked 20-39 hours. The self-employed and family workers, whose working times are more variable, accounted for only 25 000 of the discrepancy between LFS and administrative employment data. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 48 – CHAPTER 2. THE INSTITUTIONAL FRAMEWORK FOR EMPLOYMENT construction and certain service sectors but took account of working time.12 In these sectors alone, the number of deliberately unreported full-time equivalent jobs was found to decline from about 230 000 in 2003 to 200 000 in 2004 and 163 000 in 2005. This corresponded to, respectively, 6.3%, 6% and 5% of GDP in the mentioned years (Ratkovic, 2008). The under-reporting of incomes in registered or formal jobs can also cause great shortfalls of public revenue.13 It is not known how many workers are affected. But according to Ratkovic (2008), who compared information from a survey of entrepreneurship with small firms’ financial statements, the unreported wage amounts declined from 6.8% of GDP in 2003 to 4.7% in 2004 and 4.55% in 2005. This decline suggests that small businesses improved their financial accounting. If a worker’s full wage cannot be detected, the existence of certain fixed wage components defined by law or collective agreements may serve as an imperfect basis for taxation. It has been noted in several transition countries that the use of unreported wage supplements can lead to a concentration of reported wages at the level of the minimum wage (Tonin, 2007; OECD, 2003b concerning the Baltic countries; OECD, 2004, Chapter 5). A rise in the minimum wage can then boost public revenues. To some extent, the same may hold for decisions about the lowest wage recognised in social insurance – recently reduced in Serbia, see above – and for basic wages and other standardised wage elements. In general, however, the use of such wage regulations for taxation can cause rigidities and it is unlikely to lead to a fair distribution of the tax burden. To reduce informal employment and permit a more orderly application of labour fiscal law, it will be crucial to consolidate recent improvements of business registration and financial accounting and to make sure that further rapid progress is made in these areas. While his task requires the cooperation of several public agencies, notably tax authorities and the business register, the role of the labour inspectorate has special interest in the context of this policy review. 12. This study by the Statistical Office of Serbia (2007) was a result of the OECD’s regional project for the western Balkans. The figures cited from Ratkovic (2008) concern unreported jobs in construction, commerce, hotels and restaurants, transports and communications and finance. They do not include illegal business, assessed at around 1% of GDP. 13. Serbian employers tend to report similar wage amounts to both pension funds and to the Statistical Office, so any under-reporting would affect these two data sources in equal measure. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 2. THE INSTITUTIONAL FRAMEWORK FOR EMPLOYMENT – 49 The labour inspectorate should be strengthened The labour inspectorate, organised as a division of the Ministry of Labour and Social Policy, has the right to inspect registered enterprises, and if it detects a violation of the Labour Law – such as work without a labour contract – it can order its elimination within a short period (Box 2.1). However, the inspectors’ effective powers are too narrowly constrained in two respects. First, although the Law stipulates fines, these can be imposed only by courts. To apply fines on the scale that appears necessary in order to discourage offences, inspectors would need to pursue an unrealistically large number of individual court procedures. Second, unregistered businesses can legally refuse access by the inspectors to their premises. In such cases, inspectors are also unlikely to receive assistance from the police, which in general has been reluctant to act when there is no suspicion about more “serious” offences than breaking the Labour Law. In sum, the authorities should consider possible changes in the law that would give labour inspectors a right to inspect all worksites regardless of register status, and permit them to impose penalties with a simplified procedure. It will also be important to develop the inspectorate’s cooperation with tax authorities, social insurance administrations, the police and courts. Box 2.1. The labour inspectorate The labour inspectorate was set up during 2005 and 2006 as part of what is now the Ministry of Labour and Social Policy. Headed by a director general, it has altogether 330 employees of whom 321 are labour inspectors, with offices in each of the 25 administrative districts and Belgrade, including some units in the Ministry’s headquarters. Over half of the inspectors are lawyers, most of the others being engineers. The inspectorate enforces labour law, collective agreements and recognised standards for health and safety at work. It also promotes technical measures to improve safety. Strategic objectives – defined in accordance with the EU accession plan and the Poverty Reduction Strategy – are to minimise work-related risks, to fight against informal work and violations of industrial- relations rules, and to promote smoke-free worksites. A priority task at present is to check that employees have labour contracts. In 2005, about 60 000 worksite inspections targeted industrial relations and employment conditions and led to the detection of 29 000 workers without contracts, whom about 21 500 received contracts after the inspections. The corresponding numbers in the first ten months of 2006 were a little lower, with 47 000 inspections that detected 13 600 workers without contracts, for whom contracts were subsequently established in 9 500 cases. In addition, some 24 000 health and safety inspections were completed in 2005 and 20 000 in the first ten months of 2006. When workers are found to lack labour contracts, the employer is given a date before which these workers must be declared or dismissed. Within 15 days from this deadline, the employer must notify the inspectorate about the steps taken, after which a follow-up inspection is carried out. According to Ministry officials interviewed by the OECD team, this procedure has the intended effect for many of the workers concerned, but it is nevertheless unlikely to discourage future offences because there is no credible threat of sanctions. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 50 – CHAPTER 2. THE INSTITUTIONAL FRAMEWORK FOR EMPLOYMENT The Labour Law (Article 273-274) stipulates penalties of 600 000 to 1 000 000 dinars (EUR 12 500 at the current exchange rate) for enterprises that commit various offences, notably lack of labour contract, lack of social insurance, unpaid wages and wages below the legal minimum, while responsible managers are fined at rates of 30 000 to 50 000 dinars. The corresponding penalties for entrepreneurs (self-employed persons with employees) are 300 000 to 500 000 dinars. But to impose these penalties, inspectors must bring each case to an administrative court, a procedure that is seldom deemed efficient enough to merit the effort. Only 2% of the cases reported in 2005 were dealt with in this manner.14 As a further limitation, worksites not belonging to registered companies or entrepreneurs cannot be legally inspected, except to some extent with a view to enforcing health and safety standards. The problem concerns many small businesses and some larger ones, especially construction sites. Officials of the Ministry of Labour and Social Policy consider that inspections of unregistered employers in general receive too little support from the courts and police, which do not regard violations of the Labour Law as criminal offences. As a possible solution, it has been suggested that some offences against the Labour Law should be covered by criminal law; but no such reform appears to be under preparation. The inspectorate has limited ICT capacity. Electronic links have been established with the Business Registers Agency, and an in-house database is being developed for analytical and planning purposes. But the inspectorate cannot connect directly with the files of enterprises or tax authorities. The problem of unreformed enterprises Over the past decade, a shrinking but still-significant group of formal enterprises that have not been restructured or privatised has shown many signs of malfunctioning. They have frequently imposed unpaid leaves on their workers or defaulted on wage payments, while postponing or neglecting to take the necessary steps to adjust. Such practices are hardly compatible with the current Labour Law or other regulations, notably bankruptcy law, but they have often been tolerated in the face of threats of large-scale lay-offs and liquidation. The Labour Law’s Chapter IX provides for compensation of workers for unpaid wages via a public Solidarity Fund – but this applies only if a bankruptcy procedure has been initiated. As seen below, non-payment of wages is most widespread in socially-owned enterprises that operate in competitive product markets. The privatisation policy pursued since 2001 has been partially successful. Between 2001 and 2005, the employment share of private ownership increased from 15 to 54% in all registered companies, while social ownership declined from 45 to 13%. The incidence of mixed and state ownership changed less. Privatisations slowed after 2003, when 14. The inspectorate can itself impose sanctions against some less-serious offences, e.g. working without a work-book. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 2. THE INSTITUTIONAL FRAMEWORK FOR EMPLOYMENT – 51 many of the remaining non-private firms were deemed to be so unprofitable that they needed restructuring before they could be sold, or otherwise should be liquidated. To facilitate restructuring in selected state and socially-owned firms, a “Social Programme” set up in 2002 can pay various subsidies including generous severance pay, self-employment support and other active measures. About 130 000 beneficiaries were reported in 2002-2005, when the estimated annual spending was about 0.5% of GDP, with severance benefits as the biggest item, typically paid as a lump sum (World Bank, 2006).15 Phasing out social ownership The government’s current plans envisage the abolition of social ownership by 2009, along with most subsidies to enterprises.16 However, to the extent that socially-owned firms are merely transferred to state or mixed ownership – as often happened in the past, at least temporarily – political decision makers would still be responsible for many of the necessary management changes. According to companies’ annual reports for 2005, socially-owned enterprises had 144 000 employees and their value added per worker was only 40% of the private-sector average (Table 2.2). 15. According to a tracer survey of participants in selected enterprises, reported by the World Bank (2006, p. 76), the average severance benefit corresponded to nine average monthly wages. This was a high amount compared with their own recent wages, which had often been very low if paid at all, but in most cases it could not be combined with unemployment benefits. Almost all were registered as unemployed, but a majority showed little interest in job-search assistance and only 5% participated in any active programme. By late 2005, less than one-third of a group of persons laid-off in 2002-2003 had found work or started a business. 16. The 2001 Privatisation Law gave enterprises four years to implement privatisation, after which the Privatisation Agency would take over the remaining firms. This was not fully accomplished. In July 2007, the Minister of Economy and Regional Development, Mr. Mlađan Dinkić, announced a new plan to pursue privatisation with the deadline postponed till the end of 2008. According to this plan, bankruptcy procedures should be started whenever three attempts to sell a company have been unsuccessful. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 Table 2.2. Number of employees and value added (VA) per employee in registered enterprises by ownership as reported in company statements for 2005 Sector, Private State Social Mixed Total firm size Employees, VA per Employees, VA per Employees, VA per Employees, VA per Employees, VA per (no. of employees) thousands employee, thousands employee, thousands employee, thousands employee, thousands employee, thousand thousand thousand thousand thousand dinars dinars dinars dinars dinars Manufacturing 277 661 5 556 70 200 90 342 442 522 0-9 24 560 - - - - - - 25 550 10-49 45 653 - 540 3 239 4 282 52 602 25-249 68 646 - 626 19 164 28 290 117 480 250+ 139 688 4 541 48 212 57 371 247 521 Commerce 150 751 - - 12 164 20 283 184 676 0-9 54 620 - - - - - - 57 610 10-49 43 871 - - - - 3 312 49 798 25-249 27 899 - - 6 68 7 234 40 665 250+ 25 664 - - 3 320 10 296 39 629 Other non-agriculture 157 692 167 916 49 379 53 1,029 428 791 0-9 32 677 - - - - - - 35 737 10-49 40 721 4 702 5 328 5 397 55 652 25-249 34 714 20 593 18 315 14 427 87 556 250+ 50 664 143 967 25 439 33 1,379 251 910 Of which: Mining 3 669 37 1,153 6 142 - - 47 987 Utilities - - 40 1,270 4 1,028 - - 46 1,226 Transports 33 897 56 652 8 380 22 1,698 119 895 Agriculture 23 535 9 685 13 387 8 421 64 478 0-9 2 424 - - - - - - 4 375 10-49 3 760 - - - - - - 10 439 25-249 8 546 - - 7 375 4 487 24 460 250+ 9 474 8 693 5 403 3 417 25 527 Total 607 686 183 909 144 275 170 551 1,118 648 0-9 114 619 - - - - - - 121 626 10-49 132 748 5 684 12 287 12 321 165 666 25-249 138 707 22 592 50 237 53 334 268 530 250+ 224 671 156 962 80 297 103 692 563 702 - (Dash): indicates insignificant employment (up to about 2 000 persons). Source: Calculations based on data submitted by the Statistical Office of Serbia. The figures cover registered companies but not public administration or the employees of entrepreneurs (self-employed persons). SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 2. THE INSTITUTIONAL FRAMEWORK FOR EMPLOYMENT – 53 Social ownership was associated with poor economic performance in most sectors and in enterprises of all size categories. Private firms generally showed better results.17 Firms under “mixed” ownership – often unreformed – also performed worse than average, at least in manufacturing and commerce. (However, the highest value added per worker was reported by big state and mixed-owned firms in mining, utilities and transports – a result that probably reflects special market conditions.) Socially-owned firms also stand out by paying the lowest and most irregular wages. In February 2007, over 25% of their employees received no wage (Table 2.3). The corresponding proportion was then 10% on average for all ownership groups, rising to 15 to 20% in manufacturing, construction and commerce. Table 2.3. Incidence of unpaid wages Employees concerned in February 2007 Workers with Percent of all unpaid wages employees Ownership Social 32,924 26 Mixed 35,791 14 Private 14,906 11 Family 1,913 34 State 9,124 2 Total 94,658 10 Sector Agriculture, forestry 7,554 19 Fishing 360 56 Extraction 713 3 Manufacturing 49,752 16 Utilities 573 1 Construction 9,902 20 Commerce 8,565 15 Hotels, restaurants 2,339 16 Transport, comm. 5,045 6 Finance 410 2 Business services 2,642 10 Public administration 997 2 Education 1,130 1 Health, social care 3,988 3 Other services 688 2 Total 94,658 10 The figures refer to registered companies. Source: Data submitted by the Statistical Office of Serbia. 17. In general, the observation that private firms perform better also holds at more disaggregated levels, not shown in Table 2.2. But some notable exceptions in 2005 concerned private (often privatised) firms in the textile and apparel sector, which reported very low value added per worker. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 54 – CHAPTER 2. THE INSTITUTIONAL FRAMEWORK FOR EMPLOYMENT Disaggregated average wage statistics – taking account of zero and non-zero wages – indicate large differences by ownership and sector. During 2006, serious problems were evident in “light” manufacturing sectors such as textiles, apparel, wood and furniture production as well as in commerce (Table 2.4). In light industry except for food and publishing, the firms counted as socially-owned (some of which had recently been privatised)18 paid on average 5 600 dinars per month, which was only 18% of the overall average gross wage – much less than either the minimum wage or the 40% of the average wage that was then the income limit for social insurance. (The minimum wage was then about 37% of the average wage in after-tax terms.) Socially-owned retail trade firms also paid on average less than the social-insurance limit, as did mixed-owned firms in the textile, apparel and wood industries. Available statistics do not reveal for how many workers the wages were incompatible with the law or collective agreements. But this was undoubtedly the case for more than the 10% of the total workforce in registered companies who received zero wages in February 2007. By each of the measures considered here, the situation was most preoccupying in socially-owned firms in light industry and commerce, and to a lesser extent in mixed-owned firms. Even regardless of ownership, Serbian firms in these consumer-oriented sectors have largely failed to benefit from the recovering household incomes of the period 2001-2006. (Cf. Table 2.4. The rightmost column shows an accumulated employment decline of 44% for these sectors, while the middle column shows a real-wage rise of only 41% compared with an average of 71% in manufacturing and 93% in the whole labour market.) Such results demonstrate the urgency of completing the planned privatisations. But in addition, they indicate that further policy efforts will be needed to encourage better management after privatisation. Judging from the experience of other transition countries, it may be most essential to ensure that businesses face hard-budget limits in the form of well-enforced bankruptcy legislation, taxation and labour rights. 18. These wage statistics are based on data collected from employers in March and September 2006. The enterprise-level information was not fully updated with respect to recent changes of ownership. By contrast, the above-quoted value- added statistics reflect actual ownership in 2005. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 2. THE INSTITUTIONAL FRAMEWORK FOR EMPLOYMENT – 55 Table 2.4. Relative wages by sector and ownership* Sector Relative wages in 2006 Real-wage Memorandum items: 100 = Overall average (31,745 dn/month) growth Employees in Sept.'06 (000s) Employment change Total Social Private Mixed State 2001-2006 Firms Of which: All 2001-2006 (%) covered by socially registered Covered All regis- (%) wage data owned firms firms tered firms Whole economy 100 64 100 89 117 93 1,172 163 1,472 -24 -16 Primary sector 81 79 61 80 89 78 49 14 59 -36 -31 Agriculture, fishing 73 70 61 79 74 72 41 11 50 -38 -31 Forestry etc. 120 119 60 110 120 109 8 3 9 -24 -24 Extraction 124 82 92 82 143 58 28 6 29 -21 -22 Fossil fuels 143 - - - 143 45 19 - 19 -13 -15 Metal ores and other 84 82 92 82 - 86 9 6 10 -15 -12 Manufacturing 82 49 97 84 134 71 339 68 421 -40 -32 Food, beverages, tobacco 108 61 127 113 65 75 71 9 84 -28 -21 Publishing, printing 98 91 67 107 118 105 10 2 18 -37 -15 Other light industry 44 18 63 53 45 41 68 20 91 -54 -44 Textiles 29 19 53 31 - 21 15 6 18 -60 -55 Apparel 27 11 44 31 - 41 18 7 24 -62 -54 Leather, footwear 47 30 52 51 - 70 10 2 12 -40 -33 Wood products excl. furniture 38 21 49 37 36 41 6 - 11 -56 -35 Pulp, paper 64 28 109 45 - 5 6 - 9 -47 -34 Furniture; manufacturing n.e.c. 56 13 67 75 81 42 12 3 16 -47 -37 Industrial inputs etc. 105 69 116 108 153 79 86 19 101 -36 -30 Coke, petrol processing 151 120 111 106 155 53 4 - 4 -25 -22 Chemicals 124 101 88 145 113 69 23 5 27 -36 -31 Rubber, plastics 89 54 102 98 - 90 16 4 20 -22 -12 Other non-metal mineral prod. 82 46 125 83 - 25 21 6 24 -43 -37 Basic metal processing 113 80 133 85 - 143 20 4 22 -42 -40 Engineering industry 70 53 80 68 148 81 104 19 127 -33 -26 Metal prod. excl. machinery 67 40 71 71 153 61 23 4 31 -47 -36 Machinery, equipment n.e.c. 73 58 102 65 - 115 27 8 31 -27 -21 Office machines, computers 148 - 137 176 - 141 2 - 5 55 48 Electrical machinery n.e.c. 74 55 78 76 - 56 12 2 14 -30 -25 Radio, TV, phones 56 35 44 62 - 126 3 - 4 -60 -47 Medical, precision instruments 54 53 74 51 - 40 4 - 6 -51 -36 Motor vehicles, trailers 66 59 62 67 - 71 25 2 27 -29 -27 Other transport equipment 66 57 65 68 79 83 9 2 9 -31 -31 Utilities 133 103 85 107 138 87 42 5 44 -12 -10 Electricity, gas, hot water 148 137 - 107 151 67 25 2 26 -23 -20 Water 112 87 85 106 118 116 17 3 18 9 9 Construction 89 74 79 81 149 93 64 15 86 -22 -12 Commerce 92 39 108 79 140 121 86 14 198 -25 -8 Motor vehicle sale, service 111 55 147 97 139 88 13 3 24 -38 -12 Wholesale trade 112 45 119 103 153 138 32 3 114 2 23 Retail trade excl. vehicles 70 31 80 60 134 119 41 8 60 -42 -30 Hotels, restaurants 68 77 71 56 71 76 20 8 25 -43 -35 Transport 115 87 112 87 129 78 91 12 110 -21 -13 Land transport 93 92 53 79 108 59 47 9 56 -23 -13 Air transport 203 180 - - 205 61 2 - 2 -60 -56 Support, travel agencies 145 69 165 108 192 135 15 3 23 -27 -16 Post, telecommunications 130 - 208 39 128 79 26 - 28 -5 0 Other services 117 124 150 159 109 100 451 4 501 -1 2 Banking 229 188 268 215 246 161 24 2 24 -31 -35 Insurance, other finance 190 171 138 203 - 97 4 2 6 -5 11 R&D 145 145 182 149 142 32 9 2 10 -19 -19 Business services 108 88 95 101 157 92 31 8 57 -3 29 Public administration 128 130 - 139 127 92 69 - 69 8 8 Education 104 83 140 - 104 108 120 2 125 6 5 Healthcare, social work 103 52 90 - 104 93 154 2 158 -5 -3 Other 110 201 80 135 94 102 40 4 52 -2 4 * The wage statistics used here count many private firms as non-private due to incomplete updating. For estimates of the ownership structure, see Tables 2.7 and 2.11. The wage data are collected twice-yearly (March and September) from registered enterprises, institutions and organisations with over 50 employees and from a sample of smaller firms. Small firms are thus under- represented, causing a bias that can affect estimates of average wages. Note. The table shows wages before tax. The legal minimum wage was 35-40% of the average after tax in 2006, while the minimum wage for social insurance was 40%. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 56 – CHAPTER 2. THE INSTITUTIONAL FRAMEWORK FOR EMPLOYMENT Employers often disregard the Labour Law – but seldom use the flexibility it allows Working time The standard working week is 40 hours. Enterprises can shorten it to 36 hours, an option that has been little used until now. By the same token, the Law’s provisions for part-time work are liberal and neutral – all rights and duties being proportional to working time, unless other conditions are agreed upon (Article 40) – but LFS data suggest that this legal flexibility is not much used. Part-time work remained unusual for private-sector employees and almost non-existent in the public sector in 2006, as seen above in Chapter 1 (Table 1.8, Panel C). Paid overtime is limited to 8 hours per week. But the law seeks to limit the need for overtime by offering generous possibilities for employers to reschedule work over a six-month period, with up to 60 hours worked in any one week (Article 50 ff). Judging from Table 1.8 (Panels A and B), both limits are frequently exceeded.19 Actual working times of over 50 hours were reported by 9% of female employees and 19% of male employees, of whom over half even exceeded the 60-hour limit that applies to rescheduled work. Moreover, reported usual working times (Panel A) differed only a little from actual ones in the survey week (Panel B), suggesting that the rescheduling option was seldom used. These results point to the need for increased inspectoral supervision. Employment protection legislation (EPL) As noted in Chapter 1, the OECD Jobs Strategy Reassessment (OECD, 2006) identified EPL as one of a set of institutional factors behind international differences in labour-market performance. Econometric studies found that strict protection against dismissal of workers has been associated with below-average employment rates and slow job creation. As a result, it protects “insiders” at the expense of “outsiders” such as youths. Given the unfinished state of the transition in Serbia, and the imperative need to speed it up, it would be justified to make EPL more flexible than in 19. It is unclear how secondary jobs affect these LFS data. According to the 2005 LFS, just under 6% of all employed persons (including self-employed and family workers) had secondary jobs. This could explain some of the cases when actual working time exceeded the legal limits, but far from all of them. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 2. THE INSTITUTIONAL FRAMEWORK FOR EMPLOYMENT – 57 most EU countries. However, based on the OECD’s method for “scoring” the strictness of EPL, the relevant regulations in Serbia’s 2005 Labour Law (Articles 37, 153-160 and 175-191) appear on average about as stringent as they are in typical Continental European countries. They are thus more stringent than in most transition economies and much more so than in English-speaking countries, Switzerland, Denmark and Japan (Table 2.5). Table 2.5. The strictness of the employment protection legislation (EPL) Scores 0-6 from lowest to highest strictness. Country ranking by average score and for each EPL type Total: 1. Termination of indefinite contracts Weighted average 1a) Procedural 1b) Difficulty of 1c) Notice periods score inconvenience dismissal (criteria; trial and severance pay periods; penalties and reinstatement) United States 0.7 United States 0.0 United States 0.5 United States 0.0 United Kingdom 1.1 Switzerland 0.5 United Kingdom 1.3 New Zealand 0.4 Canada 1.1 Belgium 1.0 Denmark 1.5 Italy 0.6 New Zealand 1.3 Canada 1.0 Switzerland 1.5 Ireland 0.8 Ireland 1.3 Denmark 1.0 Belgium 1.8 Austria 0.9 Australia 1.5 Mexico 1.0 Australia 2.0 Korea 0.9 Switzerland 1.6 United Kingdom 1.0 Canada 2.0 Canada 1.0 Hungary 1.7 Australia 1.5 Ireland 2.0 Norway 1.0 Japan 1.8 Hungary 1.5 Poland 2.3 Australia 1.0 Denmark 1.8 Italy 1.5 Turkey 2.3 Finland 1.0 Czech Republic 1.9 Greece 2.0 Hungary 2.5 United Kingdom 1.1 Korea 2.0 Ireland 2.0 New Zealand 2.7 Serbia 1.2 Slovak Republic 2.0 Japan 2.0 Finland 2.8 Germany 1.3 Finland 2.1 New Zealand 2.0 Slovakia 2.8 Poland 1.4 Poland 2.1 Norway 2.0 France 3.0 Switzerland 1.5 Austria 2.2 Spain 2.0 Serbia 3.0 Sweden 1.6 Netherlands 2.3 Turkey 2.0 Greece 3.0 Hungary 1.8 Serbia 2.4 Austria 2.5 Korea 3.0 Japan 1.8 Italy 2.4 Serbia 2.5 Germany 3.3 France 1.9 Germany 2.5 France 2.5 Italy 3.3 Netherlands 1.9 Belgium 2.5 Finland 2.8 Netherlands 3.3 Denmark 1.9 Norway 2.6 Poland 3.0 Spain 3.3 Mexico 2.1 Sweden 2.6 Sweden 3.0 Japan 3.5 Greece 2.2 France 2.9 Korea 3.3 Mexico 3.7 Belgium 2.4 Greece 2.9 Czech Rep. 3.5 Austria 3.8 Spain 2.6 Spain 3.1 Germany 3.5 Czech Rep. 3.8 Czech Rep. 2.7 Mexico 3.2 Portugal 3.5 Norway 3.8 Slovakia 2.7 Portugal 3.5 Netherlands 4.0 Portugal 4.0 Turkey 3.4 Turkey 3.5 Slovakia 5.0 Sweden 4.0 Portugal 5.0 SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 58 – CHAPTER 2. THE INSTITUTIONAL FRAMEWORK FOR EMPLOYMENT Table 2.5. The strictness of the employment protection legislation (EPL) (cont) 2. Temporary jobs 3. Additional constraints 2a) Fixed-term contracts 2b) Temporary-work on collective dismissals (valid reasons, agencies cumulation rules, maximum duration) Canada 0.0 Australia 0.5 New Zealand 0.4 Poland 0.0 Canada 0.5 Japan 1.5 United States 0.0 Czech Republic 0.5 Korea 1.9 Slovak Republic 0.3 Denmark 0.5 Czech Republic 2.1 United Kingdom 0.3 Finland 0.5 France 2.1 Czech Republic 0.5 Serbia 0.5 Ireland 2.4 Japan 0.5 Hungary 0.5 Turkey 2.4 Ireland 0.8 Ireland 0.5 Slovak Republic 2.5 Korea 0.8 Slovak Republic 0.5 Finland 2.6 Netherlands 0.8 United Kingdom 0.5 Australia 2.9 Australia 1.3 United States 0.5 Canada 2.9 Switzerland 1.3 New Zealand 1.0 Hungary 2.9 Belgium 1.5 Switzerland 1.0 Serbia 2.9 New Zealand 1.5 Austria 1.3 Norway 2.9 Austria 1.8 Sweden 1.5 United Kingdom 2.9 Germany 1.8 Netherlands 1.6 United States 2.9 Hungary 1.8 Germany 1.8 Netherlands 3.0 Portugal 1.8 Italy 1.8 Spain 3.1 Sweden 1.8 Greece 2.0 Austria 3.3 Denmark 2.3 Japan 2.0 Greece 3.3 Italy 2.5 Norway 2.5 Portugal 3.6 Mexico 2.5 Poland 2.5 Germany 3.8 Spain 3.0 Korea 2.6 Mexico 3.8 Finland 3.3 France 3.3 Denmark 3.9 Norway 3.3 Belgium 3.8 Switzerland 3.9 France 4.0 Portugal 3.8 Belgium 4.1 Serbia 4.3 Spain 4.0 Poland 4.1 Turkey 4.3 Mexico 5.5 Sweden 4.5 Greece 4.5 Turkey 5.5 Italy 4.9 Source: OECD Employment Outlook 2004, Chapter 2, Annex 2.A; Serbia’s 2005 Labour Law. In practice, some of the detailed EPL provisions may be less strict than they appear. For example, the use of fixed-term contracts is subject to a one-year limit – stringent by OECD standards – and some sectors use seasonal and “casual” employment to a considerable extent.20 In practice, 20. The use of different types of temporary contract also seems to depend on qualifications. According to the LFS in 2005, persons with secondary or higher education were over-represented among those with fixed-term contracts, but under-represented among seasonal and casual workers. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 2. THE INSTITUTIONAL FRAMEWORK FOR EMPLOYMENT – 59 temporary contracts of one kind or the other are common in sectors where they are justified by particular conditions, while elsewhere their spread has been largely prevented. According to the LFS in 2006, some two-thirds of the employees in agriculture and almost one-third of those in construction had temporary jobs. But the overall average was 11%, which was lower than the OECD average of 14%.21 In some part, the limited use of temporary contracts reflects the legal restrictions governing their use. But it also suggests that employers regard the standard indefinite-duration contract as flexible enough. Employers are usually free to terminate employment if they follow the rules and pay severance benefits.22 The approved causes of dismissal include both individual factors, such as poor performance, and economic and technical redundancies. Procedures can be somewhat complex, as is the case in most of the EU – not least concerning collective redundancies, for which Serbia’s rules have been aligned with the corresponding EU directive.23 These procedures involve a duty to cooperate with the employment service and trade unions. To some extent, the practical implications are therefore likely to vary with the strength of the unions, the positions they take and the general climate of industrial relations. Taken together, these EPL rules have proved compatible with a good deal of labour turnover. About 450 000 job separations are reported per year, or about 25 per 100 employees (Table 2.6). But about half of this turnover is concentrated in the small group of workers with temporary contracts. Moreover, only around 50 000 separations per year are reported as dismissals, i.e. employer-initiated terminations of indefinite contracts – a very low figure, considering the large restructuring needs. 21. The reported incidence of temporary job contracts is often high in countries with rigid rules about indefinite contracts (e.g. Spain, Portugal and Sweden) and low where indefinite contracts are more flexible (English-speaking countries and Denmark). 22. The severance pay stipulated in the Labour Law is normally one-third of a monthly wage for each of the first ten years of service, plus one-fourth of a monthly wage for each subsequent year of service. Cf. above concerning the more generous severance benefits paid as part of the Social Programme for restructuring of state and socially-owned firms. 23. When the number of redundant workers exceeds certain limits (ranging from 10 to 30 depending on the size of the firm), the employer must prepare a redundancy programme in co-operation with the trade union and the National Employment Service. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 60 – CHAPTER 2. THE INSTITUTIONAL FRAMEWORK FOR EMPLOYMENT Table 2.6. Reported job separations Percent distribution by reason Reason 2004 2005 2006 2007 January-June January-June Dismissals 11 11 11 9 (employer-initiated, including bankruptcy etc.) Quits 27 25 26 27 (voluntary or consensual, including retirement) End of temporary jobs 50 49 53 52 Other reasons in law or collective agreement 12 15 10 12 Total, percent 100 100 100 100 Separations per year: - total in thousands 348 373 453 459 - separation rate per 100 employees 19 20 25 26 Source: National Employment Service. Concluding remarks This chapter has found many of Serbia’s legal employment regulations relatively well adapted to foreseeable needs, and they essentially correspond to international standards. On some points, especially EPL, Serbia’s rules are too rigid for a transition economy, a fact that may partly reflect a somewhat premature adoption of certain standards promoted by EU countries with more advanced economies. As shown in the reassessed OECD Jobs Strategy Reassessment (OECD, 2006), relatively rigid EPL rules has for a long time had a negative impact on job creation in western Continental Europe. Transition countries like Serbia, which need to sustain high economic growth and rapid restructuring in order to catch up, should aim at a significantly higher degree of labour market flexibility than is currently found in most of western Europe. In any case, some of the most serious shortcomings of Serbia’s labour market institutions belong to the implementation stage. Further reforms should therefore pay special attention to the respective roles of the principal enforcement bodies, including tax and social insurance administration as well as the labour inspectorate and the industrial relations system. Their tasks are particularly challenging in the present difficult labour market situation, where most workers have a weak bargaining position due to the large labour surplus, and where corporate governance is still poor in some enterprises. The labour inspectorate should be strengthened with additional resources and decision powers, for example concerning the use of fines and the right to inspect unregistered businesses. The inspectorate’s co-operation with other enforcement bodies such as tax authorities, police and courts should also be developed. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 3. THE NATIONAL EMPLOYMENT SERVICE AND ITS PROGRAMMES – 61 CHAPTER 3 THE NATIONAL EMPLOYMENT SERVICE AND ITS PROGRAMMES Introduction The National Employment Service (NES), a public agency under the Ministry of Economy and Regional Development,1 administers unemployment insurance along with job-brokering and counselling services and other active labour market programmes (ALMPs), mostly regulated in the 2003 Law on Employment and Unemployment Insurance. With 26 main offices2 and at least 130 small outlets, the NES has about 2 100 staff members, of whom 1 100 are counsellors and related groups in contact with clients. The average number of registered unemployed persons per counsellor can then be roughly assessed at about 500, or more than twice as high as in most western European countries. However, if the total staff resources could be related to the number of unemployment benefit recipients alone, the ratio would decline to about 50, which is similar to the corresponding ratios for the best-equipped employment service agencies in OECD countries (e.g. those in the United Kingdom, Germany and Nordic countries).3 The NES’ total expenditures in 2006 were just under 1% of GDP, of which four-fifths concerned unemployment benefits, which until now have been paid with significant delays. Most of the remaining funds were used for 1. Responsibility for the NES was transferred from the Ministry of Labour and Social Policy to the Ministry of Economy and Regional Development in the summer of 2007. 2. In addition, the NES operates five offices in Kosovo. 3. Such estimates have been made in a series of thematic reviews of the public employment service, mostly during the 1990s. For a summary table, see Grubb (2007), Table 3. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 62 – CHAPTER 3. THE NATIONAL EMPLOYMENT SERVICE AND ITS PROGRAMMES running the office network, while other “active” programmes accounted for barely 0.07 percentage points. Available estimates for 2007 suggest about the same total spending relative to GDP, but with a rise in the “active” programme share to about 0.1%. Employer and employee contributions (at a total rate of 1.5% of earnings; see Chapter 2) cover about half of the total expenditures, the remainder being financed via the state budget. Like its counterparts in other former Yugoslav republics, the NES inherited an unemployment insurance programme and an office organisation that had been developed in the socialist period. This legacy included a set of administrative registers that had been designed not merely to monitor job seekers and vacancies, but to keep comprehensive records on employers and their employees, recruitments, job terminations and changes in social insurance enrolment. The above-mentioned Law from 2003 (Article 100) still obliges the NES to keep such registers. The system has become somewhat bureaucratic and the information it provides is not as useful for client services as it could be, a problem that is now recognised and has motivated a number of recent initiatives. Nonetheless, a key shortcoming discussed below is that unemployment registers include numerous individuals who are not seeking jobs. Similarly, the NES’ monitoring of employers’ vacancies and recruitments has traditionally served to support social-insurance enrolment. As a result, it lists many vacancies for which the employers have already selected candidates, while giving too few details about the jobs for which employers wanted help in finding suitable applicants. The vacancies reported as filled – almost 700 000 on an annual basis in 2007 – probably represents the bulk of actual recruitments in the formal sector during the year; registered job seekers took up a little less than half of these jobs (Figures 3.8 and 3.9). The Change Strategy of the National Employment Service 2006-2008 (NES, 2005) places much emphasis on client-oriented services, but it also calls for better labour market data and analytical labour market indicators, which should follow international and scientific standards.4 Many analytical indicators are best produced in sample surveys such as the LFS. But OECD experience shows that the quality of public employment-service registers is 4. The Change Strategy of the National Employment Service 2006-2008, adopted by the NES in the fall 2005, had been initiated under a programme for institutional capacity-building supported by Sweden. It developed the implications for the NES of Serbia’s National Employment Strategy 2005-2010 and its National Employment Action Plan for 2006-2008, both drawing on the European Employment Strategy. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 3. THE NATIONAL EMPLOYMENT SERVICE AND ITS PROGRAMMES – 63 increasingly important as well, both for helping clients and for monitoring the performance of employment-service agencies and their various programmes.5 As far as possible, these registers should be treated as operational instruments and not as legal obligations. The NES should therefore be given a wide remit to develop its registers and information systems and to adjust them to changing needs, while discontinuing any practices it no longer finds worthwhile. Serbia’s National Employment Strategy 2005-2010 (MLSP, 2005) singles out the promotion of active job search as the most important and probably most efficient active labour market policy (ALMP) (ibid., pp. 30-33). As developed in the Strategy, this crucial function requires a capacity to organise targeted activities at limited cost as a direct complement to basic job counselling, e.g. skill assessment, CV drafting, interview techniques, individual action plans, job clubs and special activities with vulnerable groups. Concerning other ALMPs – which generally are more expensive – the Strategy takes note of the mixed results documented in the international evaluation literature, which give reason for caution with proposals for large-scale job-subsidy and training schemes. Recognising this, the Strategy favours a moderate expansion of carefully targeted programmes.6 As an example of recent Serbian experience, it mentions the pilot project “Beautiful Serbia”, supported by Spain and the UNDP, which was found to give modest but promising results by combining public works with training for construction jobs. However, the subsequently adopted National Employment Action Plan for 2006-2008 (MLSP, 2006) favours “broadening the scope and types” of ALMPs, with more subsidies towards self-employment, job creation and training. To support this stronger emphasis on ALMP spending compared with the Strategy, the Action Plan cites the EU’s 70% target for the employment rate, which in Serbia’s case has been deemed to require an annual increase in this rate by 1.5 percentage points. 5. Cf. OECD (2005b), Chapter 4, concerning the evaluation of labour market programmes; Chapter 5 concerning the use of employment-service registers for managing performance. 6. Commenting on the principal types of ALMP, the Strategy (p. 31) notes: i) public works should target the most vulnerable groups and help them gain work experience and training; ii) recruitment subsidies have modest net employment effects due to their inevitable side effects; iii) only a limited proportion of the unemployed have entrepreneurial skills, but self-employment support may nevertheless be the only realistic solution in some cases when large enterprises are restructured; and iv) training schemes should be small-scale and organised in co-operation between employers, the unemployed and specialised institutions. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 64 – CHAPTER 3. THE NATIONAL EMPLOYMENT SERVICE AND ITS PROGRAMMES Additional programme spending would target, for example, the long-term unemployed, unemployed youths and vulnerable groups such as the disabled, Roma and refugees. Until now, however, Serbia’s budgetary situation has prevented any large increase in public spending on ALMPs, which therefore continue to play a very marginal role compared with the scale of unemployment. Before discussing the principal programme types, the following pages consider the NES client registers and what can be gleaned from them about different categories of job seekers. Who are the registered unemployed? For the whole transition period, Serbia’s unemployment registers have been dominated by what appears to be a large pool of long-term clients at the margins of the job market. The median duration of current unemployment spells is about three years. This concerns a very high proportion of first-job seekers along with persons who have lost their jobs. But numerous individuals in both groups do not seem to seek jobs actively, perhaps because they have given up looking for work or are busy in informal jobs or education. Following a period of increase after 2000, the number of registered job seekers stabilised at approximately one million during 2005 and 2006. It declined during 2007 to about 850 000 at the end of the year, partly due to administrative changes considered below (Figure 3.1). Over 90% of the registered job seekers are classified as unemployed, implying that they must report regularly to a NES office and show proof of active job search (Article 91 of the Law on Employment and Unemployment Insurance). Most of the unemployed need not come to the office more than once every three months. Those who receive unemployment benefits – only about 70 000 persons (see below) – must report about their job search every month. The remaining about 65 000 job seekers, who are not considered as unemployed, are not obliged to keep contact with the NES and most of them are not ready for work.7 7. In December 2007, 62 500 of the 65 700 non-unemployed job seekers were “temporarily incapable or unavailable”; under 2 000 were recorded as employed while 1 500 were waiting for decisions. Available statistics do not always make the distinction between unemployed and other job seekers, despite its operational importance for the NES. It is given in stock terms (Figures 3.1 and 3.2), but not in flow statistics (Figure 3.8). SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 3. THE NATIONAL EMPLOYMENT SERVICE AND ITS PROGRAMMES – 65 Figure 3.1. Registered job seekers: stock data 1,200,000 1,000,000 800,000 600,000 400,000 Not unemployed 200,000 Unemployed with work experience Unemployed first-job seekers - Source: NES registers. Figure 3.2. The unemployed by duration of unemployment 1,000,000 900,000 800,000 700,000 600,000 500,000 400,000 300,000 200,000 5 years or more 100,000 2-5 years 1-2 years - Up to 1 year Source: NES registers. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 66 – CHAPTER 3. THE NATIONAL EMPLOYMENT SERVICE AND ITS PROGRAMMES Registered unemployment has fallen steadily from the beginning of 2007 until November, when it was under 800 000. The decline initially concerned those with less than one year of unemployment, so the relative incidence of longer-term unemployment increased further (Figure 3.2). In December 2007, as many as 77% of the unemployed had been on the register for over one year, while 57% had been so for over two years and 28% for over five years. On first sight, these unemployment statistics from the NES correspond well with those of the LFS, which indicated somewhat lower unemployment (falling from about 700 000 in 2006 to 600 000 in 2007), but an even higher incidence of unemployment spells longer than two years (66% in 2007). However, it is possible that many of the registered unemployed are not LFS- unemployed and vice versa. An estimate for 2002 suggested that only one-third of the registered unemployed were LFS-unemployed, while over one-third were inactive and almost as many were employed, mostly in informal jobs.8 Compared with employed workers, the unemployed by any definition include much higher proportions of young people and women (Figures 3.3 to 3.5). But youths are not as strongly over-represented in registered unemployment as they are in LFS-unemployment, indicating that many jobless youths who are looking for work do not find it worthwhile to register at the NES. By contrast, elderly persons and women (except youths) are more likely to register as unemployed than to be LFS-unemployed, suggesting that many are not very active as job seekers. This over-representation of elderly and female clients on the register has increased during 2007. Figure 3.3. Women as percent of the employed and the unemployed 60 50 40 30 20 10 0 LFS-employed, LFS- Registered Registered Oct. 2006 unemployed, unemployed, unemployed, Oct. 2006 Nov. 2006 Aug. 2007 Source: NES. 8. World Bank (2003, Vol. II, p. 69) compared data from the Survey of Living Standards with LFS data for 2002. However, the LFS definitions used then were different from those applied since 2004. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 3. THE NATIONAL EMPLOYMENT SERVICE AND ITS PROGRAMMES – 67 Figure 3.4. Employed and unemployed men: age distributions 100% 90% 80% 70% 51+ 60% 50% 41-50 40% 31-40 30% 26-30 20% 15-25 10% 0% LFS-employed, LFS-unemployed, Registered Registered Oct. 2006 Oct. 2006 unemployed, unemployed, Nov. 2006 Aug. 2007 Figure 3.5. Employed and unemployed women: age distributions 100% 90% 80% 70% 51+ 60% 41-50 50% 31-40 40% 26-30 30% 15-25 20% 10% 0% LFS-employed, LFS- Registered Registered Oct. 2006 unemployed, unemployed, unemployed, Oct. 2006 Nov. 2006 Aug. 2007 Source: Statistical Office of Serbia; NES. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 68 – CHAPTER 3. THE NATIONAL EMPLOYMENT SERVICE AND ITS PROGRAMMES Insofar as young job seekers have finished their schooling, they tend to have higher average educational qualifications than older workers (Figure 3.6). But the difference is moderate, e.g. the proportions with secondary or higher qualifications range from 22% to 33% for men in the shown age classes and from 24% to 45% for women.9 Moreover, persons with only primary education are over-represented on the register regardless of age – both compared with the employed workforce and compared with the LFS-unemployed (Figure 3.7). Comparing the registered unemployed with the LFS-unemployed, most of this difference in education remains even if one controls for the age distribution (rightmost columns for men and women, respectively).10 In conclusion, numerous young LFS-unemployed persons are not registered at the NES, and their propensity to be registered is particularly low in the group with secondary or higher education. Figure 3.6. Registered unemployed persons in August 2007: distribution by gender, age and education attainment 100% 90% 80% Higher 70% 60% Secondary 50% Primary 40% 30% 20% 10% 0% Source: NES. 9. Following Serbian practice, the term “secondary” refers to the upper-secondary level (age 15-19). 10. Due to data availability, this comparison is made only for October 2005. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 3. THE NATIONAL EMPLOYMENT SERVICE AND ITS PROGRAMMES – 69 Figure 3.7. Employed and unemployed persons in October 2005: distributions by gender and education attainment 100% 90% 80% 70% Higher 60% Secondary 50% Primary or less 40% 30% 20% 10% 0% LFS-U=LFS-unemployed; RU=registered unemployed. The “age-adjusted” bars reflect actual education by age class, while assuming the same age distribution as in LFS-unemployment. Source: NES; Statistical Office of Serbia. The relatively low number of unemployment benefit recipients – less than 10% of the registered unemployed – is explained by the eligibility rules (see Box 3.1). These are not particularly restrictive by international standards, apart from the absence of any flat-rate benefit to first-job seekers, as found in several European countries. However, amongst the registered unemployed in Serbia, a large majority are either first-job seekers, long-term unemployed persons whose benefit rights have been exhausted, or persons with only informal work experience.11 It falls upon the NES to assist all these groups and promote their entry into jobs in the formal economy. Until now, however, their registration has served mainly as a justification for access to other social programmes. Especially health-care, but also pension and disability insurance, have until now been provided more or less automatically to the registered unemployed, making it difficult for the NES to refuse registration of persons who need social insurance, even if they do not intend to seek jobs. That such apparent misuse of unemployment registers has been tolerated for a long time, if not encouraged, can thus be explained by policy concerns with the population’s 11. Following previous rules, some older beneficiaries who became unemployed before 2003 can continue drawing benefits until they become eligible for pensions. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 70 – CHAPTER 3. THE NATIONAL EMPLOYMENT SERVICE AND ITS PROGRAMMES social protection, combined with the limited impact on unemployment benefit spending. However, this practice has distorted unemployment registers and placed an administrative burden on the NES that is irrelevant to its main functions. All told, unemployment data must be used with caution – but there is no doubt that Serbia suffers from massive under-employment. This is also evident from the low employment-population ratio (see Chapter 1). The register system should be redesigned to identify those individuals who are motivated to seek jobs. Box. 3.1. Unemployment insurance The NES administers unemployment insurance according to the 2003 Law on Employment and Unemployment Insurance. It is financed by a 1.5% contribution, shared equally by employers and employees. Insured self-employed persons pay the whole contribution. Eligible persons: those who contributed for at least 12 months, or for 18 months with interruptions. Workers who quit jobs voluntarily are excluded, as are those dismissed for justified reasons related to their behaviour. Duration of benefits: • three months for persons with up to five years of insurance. • six months after over five to 15 years of insurance. • nine months after over 15 to 20 years of insurance. • one year after over 20 years of insurance. This is the maximum benefit duration up to age 51. • two years a) after over 20 years of insurance, applicable from age 61 (men) and 56 (women); b) after insurance periods of 38 years (men) or 33 years (women) from age 51; and c) after 25 years of insurance from age 55. Benefit calculation: Benefits depend on the beneficiary’s average wage in the last six months. This amount is replaced at a rate of 60% for the first three months of unemployment, thereafter 50%. However, the benefit cannot be higher than the average wage or lower than the minimum wage. The benefits are tax-free and the NES pays employer contributions to social insurance. In effect, the benefits are quite generous by OECD standards in most situations, but especially for those who previously earned the minimum wage. If wages were not paid on time, benefits depend on the wage amounts for which contributions were paid. Special groups: Unemployed persons who start self-employment can collect their benefit entitlement as a lump sum. Some persons in training receive unemployment benefits augmented by 10%. Payment arrears affect unemployment insurance, like other social insurance. In the autumn of 2007, benefits were paid with a delay of four months. During 2006, around 75 000 beneficiaries received on average just over 10 000 dinars per month (EUR 130 or 47% of the average net wage) after deduction of social insurance contributions.12 12. In addition, the NES paid monthly benefits to 23 000 persons in Kosovo. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 3. THE NATIONAL EMPLOYMENT SERVICE AND ITS PROGRAMMES – 71 New principles for registration from 2007 The observed trend break at the beginning of 2007 was brought about by a new Law on Healthcare that extends medical insurance to additional groups, notably students. These should now have less reason than previously to seek registration as unemployed unless they actually want to work. According to interviewed NES officials, the reform was phased-in during 2007, with the NES taking gradually a more restrictive approach towards accepting new clients as unemployed. The agency has also begun an effort to “clean” the existing registers of persons not actively seeking work. As a result, registered unemployment has declined for both first-job seekers and those with work experience. There have been fewer new registrations and more deletions from the register. Positive trends in vacancies and hirings have facilitated the change, but most of the rise in deletions appears unrelated to hirings (Figures 3.8 and 3.9). Figure 3.8. Registered job seekers: monthly flow data A. Newly registered job seekers B. Removed from job seeker registers 40,000 40,000 35,000 35,000 30,000 30,000 25,000 25,000 20,000 20,000 2005 2005 2006 2006 15,000 15,000 2007 2007 10,000 10,000 I II III IV V VI VII VIII IX X XI XII I II III IV V VI VII VIII IX X XI XII C. Recruited from job seeker registers 35,000 30,000 25,000 20,000 2005 15,000 2006 2007 10,000 I II III IV V VI VII VIII IX X XI XII Source: NES. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 72 – CHAPTER 3. THE NATIONAL EMPLOYMENT SERVICE AND ITS PROGRAMMES Figure 3.9. Registered vacancies: monthly flow data A. Newly registered vacancies B. Filled vacancies 75,000 75,000 70,000 70,000 65,000 65,000 60,000 60,000 55,000 55,000 50,000 50,000 2005 2005 45,000 45,000 2006 2006 40,000 2007 40,000 2007 35,000 35,000 I II III IV V VI VII VIII IX X XI XII I II III IV V VI VII VIII IX X XI XII Source: NES. A greater element of self-selection by motivated clients is welcome and merits being promoted, for example by publicity to inform the general public about the NES, its services and its new principles for registration. Individuals’ desire to be registered at the NES should normally depend on their entitlement to unemployment benefit, their motivation to seek jobs and their assessment of the usefulness of available services. Under such conditions, registered unemployment might well stabilise at a level somewhat lower than LFS-unemployment, as it has done in many countries. A difficult challenge for job counsellors To support client-oriented services, the register should thus focus on active job seekers and it should record details of importance for dealing with them. Much of this information is best collected in direct connection with job counselling, so its quality will largely reflect the frequency and sincerity of the client contacts. In OECD countries – where most of the unemployed receive cash benefits – job-search assistance has been found most cost-effective when the client-friendly service role is combined with effective controls on job-search behaviour, including a credible though moderate threat of benefit sanctions against non-compliance with the rules (OECD, 2005b, Chapter 4). The legal requirement to seek jobs actively is crucial, but it is also important to act SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 3. THE NATIONAL EMPLOYMENT SERVICE AND ITS PROGRAMMES – 73 against other abuses such as drawing benefits while working.13 Both controls and client-friendly services – “sticks” and “carrots” – tend to be ineffective if the client contacts are not sufficiently regular and candid, especially when the employment service cannot steer clients to enough vacant jobs and/or vacant slots in ALMPs. According to a recent questionnaire, about half of the OECD countries require the unemployed to report every two weeks or monthly about their job search (OECD, 2007, Table 5.2). Such reports by individuals can take different forms, ranging from telephone contacts or email to interviews with counsellors. In over half of the OECD countries, job seekers must also attend intensive interviews at least quarterly, lasting typically around 30 minutes. Elsewhere, the length of intervals between meetings can vary more, often depending on client profiles and individual action plans. Profiling and selection of priority groups As mentioned above, there are still over 500 registered unemployed per NES counsellor, of whom about 50 are benefit recipients. It is impossible with such heavy caseloads to conduct intensive interviews at sufficiently regular intervals with all those not receiving benefits, so it is essential to prioritise. As proposed in the above-cited Change Strategy (NES, 2005), the NES has begun to use a form of profiling as part of a first interview with new clients, to be held within 90 days from registration. According to Ministry officials, some 350 000 clients have been profiled during 2007 until mid-November, a figure that is somewhat higher than the number of newly registered persons. Four client categories are distinguished according to an assessment of employability and risks of long-term unemployment: • Easily employable: requiring only basic mediation services (about 10%).14 • Employable: requiring basic support in active job search, e.g. assessment of skills and capabilities, CV writing, interview techniques, job clubs or job-search training (about 15%). 13. Drawing unemployment benefits while working, if detected, is likely to be treated as insurance fraud in most OECD countries. But this is not always the case elsewhere, especially when unemployment benefits are modest. China’s unemployment insurance, for example, appears to tolerate that beneficiaries work informally – but not formally – in order to gain a supplementary income (OECD, 2002a, Chapter 16). 14. The percentages given in brackets refer to preliminary assessments of the outcomes of profiling in 2007. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 74 – CHAPTER 3. THE NATIONAL EMPLOYMENT SERVICE AND ITS PROGRAMMES • Employable: requiring more substantial ALMPs such as training, job subsidies or public works (about 45%). • Hardly employable: not possible to integrate in the labour market in the short run (about 30%). For clients assigned to the first two categories, the Change Strategy envisages the establishment of “individual employment plans” at the first meeting. But for the other two categories – comprising 75% of the persons concerned – the employment plans can be postponed if further meetings are scheduled. The third category alone, for which more expensive ALMPs are deemed necessary, appears substantially greater than the current capacity of the Serbian economy to either fund or deliver such programmes. From an OECD perspective, it is surprising that the above-cited strategy papers do not suggest more special treatment of benefit recipients. In most countries, recipients of unemployment benefits and other social benefits are the main targets for activation policies, if not the only targets. In a budgetary context, targeted spending on activation measures can be justified by the need to control benefit spending. This linkage also makes it legitimate and feasible to put pressure on benefit recipients to co-operate with the employment service, as many countries do by spelling out mutual obligations in individual action plans. Following the international trend, Serbia obliges unemployment benefit recipients to agree with the NES about mutual obligations (Law on Employment and Unemployment Insurance, Article 14). But it appears questionable if agreements about mutual obligations can be considered as binding for unemployed workers whose benefits are paid with several months’ delay. In the future, however – when the NES can hopefully fulfil its part of the mutual-obligations bargain – such agreements should strengthen the recipients’ commitment to job search, and the NES can then respond by giving them priority in access to slots in ALMPs. However, the Law’s general provisions about active labour market programmes (Article 31ff; cf. below) only suggest a broad targeting of all the unemployed or large sub-categories such as the long-term unemployed, first-job seekers, youths and older workers. Any unemployed person is obliged to accept training if the NES proposes this (Article 37). Benefit recipients are probably given priority in terms of participation in ALMPs, especially the more expensive ones, but this priority could be made more explicit and linked systematically with individual employment plans. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 3. THE NATIONAL EMPLOYMENT SERVICE AND ITS PROGRAMMES – 75 It will also be important to develop a consistent service approach to registered persons without benefits – insofar as they want to seek jobs. Those who appear motivated should receive personal attention, but they should not be classified as unemployed without evidence that they are available for and seeking a job. A full assessment of a client’s labour market situation appears necessary if he or she is considered for access to targeted programmes. For those who do not find work within a few months, the above-described procedures for new clients – profiling and individual employment plans – need to be followed-up in counselling interviews at regular intervals. However, a higher proportion of clients could be registered only as job seekers and asked to provide the NES with personal information for job matching, without having to make further obligations. For those who do not keep regular contact with an office, the NES can do little more than remove them from the register, and they would still be eligible for general services such as vacancy information. Active labour market programmes (ALMPs) The budget for ALMPs in 2006 amounted to a modest 1.3 billion dinars or 0.07% of GDP, rising to an estimated 2.3 billion dinars or 0.10% of GDP in 2007. Training, self-employment and job subsidy programmes had altogether some 40 000 participants during 2006 (Tables 3.1 and 3.2). As mentioned above, this budget allocation does not cover NES staff and other costs of running NES offices, which represent an additional 0.1% of GDP and include most costs of the above-mentioned intensified counselling measures. Spending per participant is moderate for many training courses and public works, probably reflecting their relatively short duration (2-3 months for training courses), but it is much higher on average for job subsidies. In addition to the programmes mentioned in Tables 3.1 and 3.2, some active measures have been financed under a Social Programme for laid- off workers in certain enterprises, as mentioned in Chapter 2. Moreover, a new hiring subsidy has been introduced in the form of exemptions from wage tax and employers’ social insurance contributions. These can last for two years after hiring registered unemployed persons aged under 30, or for three years if they are recruited as trainees.15 The employer is obliged to retain the worker for another three years after the subsidy period. 15. Trainee jobs can usually last up to a year. They are intended for first-job seekers and others who lack the work experience required for particular jobs, as specified in law or enterprises’ organisation plans (Labour Law, Article 47). SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 76 – CHAPTER 3. THE NATIONAL EMPLOYMENT SERVICE AND ITS PROGRAMMES Judging from OECD experience of targeted exemptions from social insurance contributions, they may soon carry a higher budgetary cost than all the other ALMPs. As with most other hiring subsidies, tax exemptions are liable to have only a small net effect on employment due to significant “deadweight” effects (subsidies to recruitments that would have happened anyway). But they may be worthwhile if they reduce informality. To increase the chances of success in this respect, the exemptions should be linked with systematic checks of employer compliance with other tax obligations and labour law. Table 3.1. Active labour market programmes: estimated spending in 2006 Million Euros per dinars participant* Training 400 377 Vocational training 264 520 Courses for complementary skills 114 219 Other 22 681 Support of self-employment 258 931 NES Training for entrepreneurs 4 36 Activities under other bodies 4 .. Subsidy to self-employment 250 1,512 Employment subsidies 511 1,261 Subsidies to new jobs 202 1,109 Regional programmes 225 1,088 Jobs for the disabled 48 2,902 Measures for groups at risk 36 .. Public works 100 403 Contribution to job creation by other 26 .. bodies Total 1,295 650 * Using current exchange rates and the participant numbers in Table 3.2. Source: National Employment Action Plan for 2006-2008. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 Table 3.2. Participants in active labour market programmes 2006 2005 (budget Total Key target groups as percent of the participants estimate) Age <25 Age >50 Women Low- LTU First-job skilled* (>2yrs) seekers Activities in the employment service .. 100,555 30% 11% 54% 19% 19% 50% Group information .. 34,946 28% 15% 52% 27% 12% 55% Individual assessments and plans .. 20,730 30% 10% 59% 18% 15% 50% Job fairs 36,000 27,851 25% 11% 48% 20% 29% 38% Job-search courses 39,750 17,028 42% 3% 63% 1% 22% 61% Training 12,834 11,325 31% 3% 58% 1% 20% 60% Further education .. 50 56% 0% 28% 92% 34% 60% Vocational training 6,135 6,357 35% 1% 59% 0% 16% 72% Courses for complementary skills 6,309 4,414 27% 4% 60% 0% 27% 45% Foreign languages .. 919 25% 2% 68% 0% 23% 48% IT .. 1,456 24% 6% 69% 0% 34% 37% Specialised IT .. 1,225 27% 4% 45% 0% 24% 48% Other .. 867 35% 4% 55% 2% 25% 54% Training at work 390 504 24% 5% 38% 13% 12% 32% Support of self-employment** .. 26,165 14% 19% 41% 18% 32% 16% Business centres (IT, legal support) .. 8,332 12% 14% 42% 19% 21% 11% Training in setting up a business ---by the State Agency for SMEs 1,350 1,112 17% 15% 46% 12% 29% 14% ---by the NES .. 6,742 14% 19% 41% 15% 31% 16% Lump sum for self-employment*** .. 305 0% 44% 37% 8% 0% 0% Subsidy to self-employment 2,000 9,674 15% 23% 41% 19% 44% 21% Employment promotion 4,900 1,167 31% 7% 39% 27% 35% 50% Subsidies to new jobs 2,200 133 30% 11% 39% 13% 38% 57% Regional programmes 2,500 937 31% 6% 39% 29% 34% 47% New jobs for the disabled 200 97 41% 9% 34% 25% 36% 65% Public works 3,000 - - - - - - - Total .. 139,212 27% 12% 52% 17% 22% 44% Memorandum: The corresponding groups as percent of all registered unemployed persons (Dec. 2005) 19% 19% 54% 37% 52% 53% * No secondary education; ** In addition to the mentioned programmes, the Serbian Development Fund gave microcredits to about 3 500 unemployed persons in 2005; *** Unemployment benefits paid as a lump sum. LTU: Long-term unemployed. Source: NES. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 78 – CHAPTER 3. THE NATIONAL EMPLOYMENT SERVICE AND ITS PROGRAMMES Compared with 2006, the above-mentioned National Employment Action Plan for 2006-2008 envisaged a rapid expansion so that ALMP spending would be almost twice as high in 2007 and nearly three times as high in 2008. This increase would be similarly strong for each of the main programme types. But although the government first endorsed this plan, it was not fully implemented in 2007. The state budget for 2007 was not approved until June, and the authorities may have under-estimated the difficulties that must be overcome at the implementation stage. Nevertheless, a government memorandum from July 2007 announced that ALMPs – including all categories cited in Table 3.2 – were estimated to concern 636 700 participants and cost 12.5 billion dinars over the three years 2008-2010 (Government of the Republic of Serbia, 2007, p. 69).16 But according to the Ministry of Economy and Regional Development (2007), the numbers of ALMP participants in 2007 are likely to be just a little higher than in 2006. In terms of their general orientation (if not their volume), reported activities in 2005 and 2006 correspond with the principles of the National Employment Strategy 2005-2010, which as mentioned give top priority to the employment service itself and related activities such as group information, individual assessments and plans, job fairs and job-search courses. But even for these important activities, the participant numbers shown in Table 3.2 are low compared with the stock and flows of registered unemployed persons. At least one of the above-mentioned items, individual assessments and plans, should involve all newly registered unemployed persons – 300 000 to 400 000 persons per year – unless they find work within a few months. The participant figure given for 2005 was probably not even sufficient for the inflow of new unemployment benefit recipients. Against this background, job-search counselling in various forms should remain the top priority in the near future. According to international experience, not only are these activities generally cost-effective in their own right, they are crucial for the efficiency of more expensive ALMPs that should be targeted according to careful assessments of individual needs, which in practice depend on counsellors’ client contacts. Policy decisions about the scale of ALMPs should take account of the organisation’s capacity to make such assessments. It is also relevant to consider to what extent various sub-groups among the unemployed participate in the existing programmes. In 2005, every programme type except self-employment support showed an 16. Concerning the financial management of the NES, this memorandum stresses the goal of higher formal employment so that more of the spending can be financed via contribution revenues (Government of the Republic of Serbia, p. 78). SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 CHAPTER 3. THE NATIONAL EMPLOYMENT SERVICE AND ITS PROGRAMMES – 79 over-representation of youths and first-job seekers (Table 3.2, six rightmost columns). Elderly persons and the long-term unemployed are correspondingly under-represented. Unemployed women are more likely to participate in training than in subsidised jobs or self-employment schemes. On the whole, simple demographic distinctions play a moderate role (except that, as in most countries, few elderly persons participate in training). More disturbing, however, is that almost no vocational training course receives pupils who lack secondary education. Such persons are found only in a small further-education scheme, which targets the low-skilled, and an on-the-job training programme. Considering that 37% of the registered unemployed are low-skilled, it will be important to develop more further-education and training options that can help low-skilled adults to overcome their educational handicap. Courses of this nature have played a role for a considerable time in, for example, Germany, Ireland, the Netherlands and the Nordic region. But while some countries have targeted such options on the unemployed, they can also be considered as a complement to initial education that should be provided by the regular education system. Concluding remarks A key challenge considered in this chapter concerns the need to prioritise within an exceedingly large pool of registered unemployed clients, which includes many who are not actively seeking work. Appropriate principles for keeping NES client registers must be consistently applied. This should help bring registered unemployment down to a level that corresponds to the actual number of unemployed persons who are motivated to seek jobs via the NES. But even then, the number of unemployed clients will remain high compared with available resources. Serbia’s plans to modernise its employment service take account of recent international experience, for example concerning the links between job counselling and other programmes. The country’s situation differs compared with almost all OECD member countries insofar as relatively few clients receive unemployment benefits, and any such benefits are paid out with a delay of four months. The NES therefore has less scope for putting pressure on clients who do not want to seek jobs. But this makes it all the more justified to concentrate on those who are motivated to co-operate. Recipients of unemployment benefits must be a priority group, considering the need to control benefit spending. In the future, the proportion of newly registered persons who are eligible for benefits should increase if and when the level of formal employment recovers. Should this prove difficult to accommodate in budgetary terms, the authorities may SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 80 – CHAPTER 3. THE NATIONAL EMPLOYMENT SERVICE AND ITS PROGRAMMES consider a possible shortening of the maximum duration of the benefits, which is longer than in most comparable economies. An expansion of the training programmes could be justified, provided that they are carefully designed to address specific needs that can be identified as a result of the employment service’s interactions with local employers and job seekers. Existing programmes for self-employment and job subsidies are relatively expensive unless they have short duration. Such programmes cannot solve a large-scale unemployment problem, a goal that is best promoted by persevering in the policies to establish good institutional conditions for creation of jobs that are productive enough to be financed on market conditions. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 ANNEX. RIGHTS AND RISKS: LABOUR MARKET CHALLENGES IN A POST-SELF-MANAGED ECONOMY – 81 Annex RIGHTS AND RISKS: LABOUR MARKET CHALLENGES IN A POST-SEL-MANAGED ECONOMY by Vladimir Gligorov1 Introduction With the dissolution of Socialist Yugoslavia, social ownership and self-management of enterprises came to an end as well, leaving only some anomalous remnants. Research about this type of “market socialism” declined at the same time. Some studies of the transition in Yugoslav successor states have considered their institutional path-dependency, especially when it comes to privatisation and labour market performance. This path-dependency has weakened with the passage of time, if it ever existed. But it can still be expected to play some role in countries like Serbia, where the transition has been delayed. To clarify the implications, this Annex first gives a stylised account of the former Yugoslav “market socialism”. It describes the micro and macro-economics of social ownership and self-management and compares the successor states’ experience of transforming this system. Serbia’s development is then briefly reviewed and the remaining economic impact of the discussed path-dependency is assessed. Some policy suggestions conclude the Annex. Model and muddle The former Yugoslav system of market socialism has been extensively analysed from different theoretical, institutional and empirical points of 1. Senior economist at the Vienna Institute for International Economic Studies (WIIW). SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 82 – ANNEX. RIGHTS AND RISKS: LABOUR MARKET CHALLENGES IN A POST-SELF-MANAGED ECONOMY view. The theoretical debate was mostly about the consistency with which self-management was applied and how it affected efficiency. The interest in efficiency was partly comparative, i.e., whether a self-managed firm is more or less efficient than a profit-making one. Analysts of the institutional setup, and especially its development, were mainly interested in how it dealt with the fundamental problems of the socialist federal state. Empirical research was aimed at checking the model’s expected consequences against actual outcomes in terms of resource allocation, income distribution and economic growth. Two main questions emerged from this voluminous literature on the Yugoslav experiment. First, how suitable was the model of a self-managed economy, used in theoretical debates, for understanding Yugoslavia’s economy? Second, how important was the formal institutional framework for the actual functioning of this economy? Both could be hard to assess in consistent empirical terms against a reality that presented itself as an ever- changing muddle of regionally diverse developments. But the two questions may still be important to answer if it is hypothesised that the system has partly persisted, institutionally or behaviourally, long after the collapse of the Yugoslav state and the progress of transition. In other words, the two questions boil down to one: how did the model relate to the muddle? Before going into that question, it is worth noting that relatively few studies have combined micro and macro-economic approaches. The work of the late professor Aleksander Bajt stands out in that respect.2 Most other researchers showed little interest in the macroeconomic foundations of the self-management system, tending, instead, to draw conclusions about macroeconomic performance from the Yugoslav economy’s microeconomic foundations. Some, like the late Branko Horvat, believed that the inconsistency of the macroeconomic and indeed the political system interfered with the performance of the micro-economy, and so was responsible for the unsatisfying outcomes (Horvat, 1982). However, both the institutional setup and economic policies were important. A complete model of Yugoslav market socialism would have to combine its macroeconomic institutions and policies with the constraints and incentives that resulted from microeconomic arrangements. Such a model of the political economy might be more useful in explaining the facts of Yugoslav economic development. It could also provide some insights into the paths of transition that the different successor states took. What then was the model of the Yugoslav muddle? The model evolved from the early 1950s to its “classic” setup of the late 1970s, after which it was 2. The crowning achievement is Bajt (1988). SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 ANNEX. RIGHTS AND RISKS: LABOUR MARKET CHALLENGES IN A POST-SELF-MANAGED ECONOMY – 83 modified in a piecemeal manner in the 1980s and finally reformed almost out of existence in 1989 and 1990. What was left was subsequently transformed in different ways in each successor state. Throughout its history, the model relied on the assumption that five main principles could be followed: 1. Productive resources should be owned socially and managed collectively. 2. Markets were to allocate goods, labour and most services. 3. Capital goods and investments should be financed through the banking system rather than state budgets, while banks were under the control of enterprises and the six republics (federal states).3 4. The fiscal system was to be decentralised, republics having the dominant taxing and spending powers. 5. Foreign trade and financial flows were to be increasingly liberalised as the economy developed, substituting tariffs and capital-account restrictions for the state monopoly on foreign trade. For corporate governance and the management of enterprises, these key principles were institutionalised in different ways in different periods. But again, two institutional assumptions were invariably present despite ever-changing practical arrangements: • Market self-management: enterprises were to be autonomous, their transactions voluntary and their employees empowered to decide about employment and investments. • Social ownership: enterprises’ assets, and all other assets that were socially owned, were not tradable. Throughout the existence of the system, it proved difficult to find a consistent codification of these assumptions. As an apparent result, the final version – formulated in the Law on Associated Labour in 1976 – was excessively detailed and full of unnecessary regulation. Laws adopted in 1997 about banking and the financial system tried unsuccessfully to combine the principles of commercial banking with social ownership and self-management. At the same time, an increasing de facto concentration of political and economic power at the level of the six republics must be taken into account. 3. The Federation consisted of six republics: Bosnia and Herzegovina, Croatia, Macedonia, Montenegro, Serbia and Slovenia. Serbia’s two autonomous regions, Vojvodina and Kosovo, had almost the status of republics until 1989. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 84 – ANNEX. RIGHTS AND RISKS: LABOUR MARKET CHALLENGES IN A POST-SELF-MANAGED ECONOMY The system as it existed was not identical or even very similar to those that could be found in theoretical models of worker-managed or “Illyrian” firms, which often guided the understanding of the Yugoslav economy.4 Moreover, as the formal institutions laid down in different laws were not always consistent, explanations of the economic performance could not be simply deduced from the legally defined system. The actual model that could best capture the performance of Yugoslav market socialism would also have had to reflect its macroeconomic characteristics and its response to various shocks, some of which were exogenous to the institutional setup. One shock with far-reaching consequences was the rise in international labour mobility. In the mid-1960s, Yugoslavia allowed outward migration and a sizeable number of people left – mainly for Germany and some other western European countries – with profound consequences for domestic labour markets and macroeconomic developments. Studies are lacking on the impact on employment, wages and income, but more is known about the consequences for the balance of payments and the exchange rate, and thus for monetary policy. One result was that the Yugoslav economy relied increasingly on the Deutschmark as its preferred currency from the mid-1960s and on. Another important shock was connected with industrialisation, which brought changes in productivity and urbanisation and had all the usual consequences of rapid development. This was accompanied by growing trade and financial relations with the rest of the world, forcing domestic producers to deal with foreign competition. Yugoslavia was far from being a closed socialist economy, even though the level of protectionism was rather high throughout the period. Also, as with most socialist countries, the fact that Yugoslavia was politically and socially unstable had significant consequences for policy making. Ever-present centrifugal forces had to be accommodated in fiscal, economic and social policies. The socialist federation had some success in attenuating tensions between regions and ethnic groups – especially in the period from the early 1950s to the late 1970s, when it managed to sustain both peace and strong economic growth. But the historic tensions did not 4. The Illyrians, an ancient tribe, lived in the western Balkans before the arrival of the Slavs in the 6th century. In economic literature, the “Illyrian firm” refers to theoretical models of worker-managed firms, where workers decide locally about both output and prices. Ward (1958), who developed such a model, pointed out that it was not identical to the actual situation in Yugoslavia. An important difference was that outside pressures, especially by local governments, seemed to have considerable influence on Yugoslav firms’ decisions (p. 585). SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 ANNEX. RIGHTS AND RISKS: LABOUR MARKET CHALLENGES IN A POST-SELF-MANAGED ECONOMY – 85 disappear, and they were accompanied by important regional differences in economic development. A federal fund transferred money from developed to less-developed republics, which helped the latter to quell some social tensions. But these subsidies were unpopular elsewhere, adding to the increasing difficulty of sustaining political cohesion at federal level. Finally, interest in Yugoslavia’s actual economic performance increased in the early 1980s, when a balance-of-payments crisis occurred and turned into a systemic crisis, prompting eventually the attempt at systemic reform that failed because the country disintegrated. However, even this disintegration was then often explained as a consequence of the failure of the economic system, though other explanations were clearly more fundamental. Taking account of all these factors, the fact remains that the Yugoslav system developed significant macroeconomic disequilibria. Relatively high unemployment was endemic from the mid-1960s and on. Unemployment may have been moderate in Slovenia and Croatia, the most developed republics, but there is no doubt that the bulk of the country was marked by labour surplus. Inflation also tended to be relatively high in the 1960s and 1970s. It accelerated in the 1980s when the balance-of-payments crisis erupted and reached hyperinflationary rates just before a major reform effort in December 1989. The sources of inflationary pressures were not well understood and much of the discussion of possible causes was focused on self-managed enterprises’ behaviour rather than on the stance of macroeconomic policies. The two factors were arguably inter-connected; but it was the macro-economic stance that most directly caused an inflationary bias. Perhaps the most important shocks were the development of the balance of payments. Following the liberalisation of foreign trade and financial relations in the 1960s, and with a steady growth of remittances from emigrants, the trade deficit widened as imports grew much faster than exports. In the mid-1970s, Yugoslavia also started borrowing in commercial markets. The inflow of foreign currency led to growing currency substitution, with the result that monetary authorities preferred to peg the exchange rate to the Deutschmark. This, in turn, facilitated imports of goods and services. The balance-of-payments crisis of 1981-1982 was triggered by a sharp increase in world interest rates, which had similar effects in several emerging market economies, e.g. Poland and various Latin American countries. Thereafter, the Yugoslav economy went through a decade of stagflation, growing at less than one percent per year on average for the SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 86 – ANNEX. RIGHTS AND RISKS: LABOUR MARKET CHALLENGES IN A POST-SELF-MANAGED ECONOMY 1980s. It was also in that period that unemployment increased significantly in the less-developed regions, while major fiscal imbalances had to be financed via ever-rising inflation. Yugoslavia was not heavily indebted when the balance-of-payments crisis began: its foreign-debt-to-GDP ratio was hardly higher than 20%. The problem was not solvency but liquidity – and in that respect, social ownership proved to be the real constraint on the country’s ability to service its foreign obligations. Indeed, the main issue of debate throughout the decade after the balance-of-payments crisis was how to make socially- owned assets marketable. This eventually turned into a debate on privatisation, to be carried out later in the successor states. Rights and risks in enterprises Given the history of Yugoslav market socialism, what are the characteristics that can be said to have endured even after the disintegration of the country and the transformation of its economy and institutions? There are basically two ways in which path-dependency could work: through specific rights and obligations, or through institutional inertia. In socialist Yugoslavia, a specific structure of rights and obligations could be traced to certain attitudes towards ownership and rights to employment, which were inherent in the concept of social ownership. The definition was negative: a socially-owned asset was neither private nor state-owned. Society delegated the right to manage firms to their employees, so that a firm became nothing but the association of its employees. No explicit distinction was made between employers and employees, except that firms employed managers who were responsible not only to the self-management bodies but also to the law. The right to self-manage also implied a right to the proceeds. Firms did not distinguish between wages and profits, but treated them together. Each self-managed firm would distribute corporate income – sales minus purchases and taxes – between wages and investments. Given that the wage sum depended on corporate income, while employment did not, it was assumed that all workers as a “team” would have a common interest in making the best possible use of the capital and to make new investments to enhance their future wages. This system provided a mix of rights to ownership, employment and management, combined with some positive incentives. But workers had no clear obligations as owners because there was no clear responsibility for risks. The hiring of additional employees, like investments, was to be guided by workers’ interest in a growing income. But there was no clear provision for firing of employees or for risk-taking in investment SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 ANNEX. RIGHTS AND RISKS: LABOUR MARKET CHALLENGES IN A POST-SELF-MANAGED ECONOMY – 87 decisions. A loss-making firm could do little to restructure its costs; all it could do was to stop investing, and perhaps to stop increasing the wages. And if a firm could not service its debt, its bank had few options but to refinance it. All told, the system was strong on workers’ rights and weak on the allocation of risk. This could be expected to lead to low productivity and misallocation of investments, a problem that was addressed in various ways in different periods. In the late 1970s, each firm was obliged by law to maintain its value, at the very least; but this merely highlighted the question of how to determine the value of a firm when its assets were not valued on the market. There were also some provisions about the circumstances in which a firm could fire employees, and legislation was eventually passed in the late 1980s about some kinds of bankruptcies. But given the underlying structure of rights and incentives, all these attempts to institutionalise risks and obligations proved ineffective. In principle, labour and capital mobility and the entry of new firms could mitigate such institutional deficiencies. Studies of labour mobility were scarce, but a bit more was known about the performance of new firms. Some research was devoted to the mobility of managers, who as mentioned were hired by the firms so that a market for management jobs could be expected to develop. Indeed, some regional differences in enterprise performance were attributed to the labour markets for managers. By implication, the rigidities of social ownership and self-management were not absolute: competition for goods, services and labour did have some impact in the direction of better resource allocations. In the voluminous literature on self-management, theoretical models identified several problems with labour and capital markets. The functioning of the labour market would reflect a tendency for labour-managed firms to maximise average wages (value added per worker) and not to equate marginal costs with marginal revenues: the so-called Ward effect (Ward, 1958). The firm would then behave as a monopoly rather than increasing production and employment as much as a competitive, profit-maximising firm would have done. Aggregate employment would then be lower than in a competitive market economy with profit-maximising firms. In addition, wages might not tend to equalise across firms, considering the limited labour mobility in monopolistic firms. However, empirical evidence about these predicted effects was not altogether convincing. Wages were found to display systematic differences across firms, but it was not clear if this was an effect of the goal functions of self-managed firms. Studies did tend to show, however, that capital- intensive firms paid higher wages than did labour-intensive firms at the SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 88 – ANNEX. RIGHTS AND RISKS: LABOUR MARKET CHALLENGES IN A POST-SELF-MANAGED ECONOMY same skill levels, suggesting that workers could capture some profits at enterprise level. Another well-known problem was that any right to capture proceeds from private investments would favour these over socially-owned investments in the long run: the so-called Furubotn-Pejovich effect.5 Both entrepreneurs and households were expected to prefer private ownership of their capital if this was allowed. So if an employee could choose between higher wages and higher socially-owned investments in his or her firm, he or she would prefer higher wages, which could then be invested in private assets or consumed. If private ownership were not allowed, social ownership would merely favour consumption over investment. But if both ownership forms were allowed to compete, the prediction was that private firms would drive out socially-owned firms. These two effects have been used as arguments against social ownership and self-management, indeed from both state-socialist and liberal points of view. Much of the debate in Yugoslavia was exactly about whether one or the other shortcoming could be deduced from deficiencies inherent in the self-managed economy. If the labour market was inefficient due to the Ward effect, it could be argued that privatisation would have helped – but so would some conceivable state interventions.6 Similarly, if investments were too low due to the Furubotn-Pejovich effect, privatisation could be an answer – but so could a return to state-led investments, which might have required a nationalisation of socially-owned assets. The third alternative was developed by Aleksander Bajt to defend social ownership in the late 1980s.7 His idea was that self-managed socially-owned firms would be pushed towards improved efficiency if they had to compete in a completely free and open market economy with a government that was constrained, perhaps by law, to pursue prudent macroeconomic policies, 5. According to Furubotn and Pejovitch (1972, p. 1155), Yugoslav workers’ rights to a company’s capital and profits differed from those of a stockholder on two counts. First, Yugoslav workers had some possibilities to influence the managers, but it was uncertain to what extent they could do this in practice. Second and most important, a worker’s share of the capital could not be sold and the right to a share in profits applied only as long as the worker remained in the firm. To attract capital in competition with private investments, a socially-owned firm would therefore have to give much higher annual returns than the private investments. 6. The state could offer job subsidies, or merely exercise political pressure on self-management bodies to hire more workers. 7. Primarily in Bajt (1988). SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 ANNEX. RIGHTS AND RISKS: LABOUR MARKET CHALLENGES IN A POST-SELF-MANAGED ECONOMY – 89 i.e. to maintain balanced current and fiscal accounts. There were obvious problems with this proposal, but the point was that the Yugoslav economy’s underperformance might have been due primarily to bad economic policies. According to Bajt, the experience of the short reform period in the 1960s (see above) proved that a combination of market liberalisation and responsible fiscal and monetary policies could substantially improve the efficiency of both capital and labour allocations. The performance of such an economy would still be lower than what could be achieved in a privately- owned economy, but many argued that it would have preferable social characteristics. In sum, the Yugoslav heritage could give rise to three alternative “path-dependencies”: 1. To nationalise risk without interfering with the rights of employees. 2. To liberalise internal and foreign trade, privatise enterprises and quickly establish a market economy in line with the international main stream. 3. To restore macroeconomic balances and quickly increase market competition, while changing rights and risks only gradually when it becomes socially and politically acceptable. Successor states’ experience provides material for comparing the three types of path-dependency. As seen in the following section, this contributes to the understanding of the deficiencies of the previous system and the strength of path-dependencies. Comparing transitions Yugoslavia disintegrated and new states chose quite different transition strategies. Putting aside the still-conflictual developments in Bosnia and Herzegovina, the transitions of Slovenia, Croatia, Serbia and Macedonia can be usefully compared. Slovenia is where the break with the previous system was most gradual. The country came out of socialism with a much more developed economy than other socialist countries, being comparable to South European EU members. There was a social consensus that the inherited system was not grossly inefficient. In fact, it was believed that most enterprises were sound and managerial skills well allocated. It was also deemed important to avoid conflicts and social disintegration. As a result, Slovenia’s transition strategy embraced a rather strong path-dependence. It had the following key elements: SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 90 – ANNEX. RIGHTS AND RISKS: LABOUR MARKET CHALLENGES IN A POST-SELF-MANAGED ECONOMY 1. Insider privatisation to turn self-managed firms over to their employees, i.e. rights and risks became symmetrically allocated. 2. Liberalisation of foreign trade in order to maintain the competitiveness of the Slovenian economy. 3. Monetary policies aimed at the stability of the real-exchange rate and thus at achieving a balanced current account. 4. Balanced fiscal policy. The outcome was generally positive. Slovenia’s transitional recession came to an end after barely one year as it successfully reoriented its exports from the collapsed Yugoslav market to the EU, while avoiding any great volatility in real or nominal values. Unemployment increased from just over 1% to nearly 10%, but it came down and has now been around 6 and 7% for quite some time. The developments on the labour market can be explained by the increased efficiency of enterprises and have no obvious relation to either Ward or the Furubotn-Pejovich effects. Recently, there has been a debate about a lack of incentives for innovation in Slovenian firms, which may be a consequence of both the unreformed institutions and the economic policy stance. But it is hard to say whether, and to what extent, this can be attributed to the persistence of a structure of rights and risks inherited from social ownership and self-management. The experience of self-management was less positively assessed in Croatia, which chose to nationalise all socially-owned assets before privatising them in various ways. It also chose to rely on an economic policy mix that led to significant instability and macroeconomic imbalances. Its privatisations led to massive resource misallocations that resulted in widespread deindustrialisation and bad performance in the banking system until the latter was completely sold to foreign banks. Croatia thus did away with its legacy of self-management, but it initially retained much of the economic policies that had been characteristic of Yugoslavia. In contrast to Slovenia, it is difficult to detect any path dependence in terms of attitudes about rights or the distribution of risks. In fact, its transitional employment decline and the resulting rise in unemployment were quite similar to what happened in, for example, Poland and Slovakia. Macedonia is an example of a third type of transition. It privatised its socially-owned assets, but mostly to managers. It was decided that the rights to socially-owned assets were basically to be assigned to the managers who were expected to start behaving like risk-conscious entrepreneurs. As in Croatia, the economic policy mix was similar to the previous Yugoslav one. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 ANNEX. RIGHTS AND RISKS: LABOUR MARKET CHALLENGES IN A POST-SELF-MANAGED ECONOMY – 91 It thus created a rather closed and monopolised product market, while combining a fixed exchange rate with a lax fiscal policy. This led to successive attempts at fiscal consolidation and a restrictive monetary policy, which, together with the unreformed product market, contributed to the country’s disappointing economic performance and a record high unemployment rate. It is striking that Slovenia, which displayed the strongest path dependence with respect to social ownership and self-management, did better than the successor states that opted for a more clear rupture with this heritage. Two main factors seem to account for this difference. The first was the belief in Slovenia that employees already held de facto ownership rights, which made it legitimate to entrust them with these rights de jure as well along with the risks that enterprises faced. Elsewhere, ownership rights were often seen as belonging to the state or to the managers, if not to individuals who were close to the government in one way or the other. Second, Slovenia relied from the beginning of its transition on open market competition and hard-budget limits to discipline enterprises, principally through the need for them to ensure a growth in exports. This approach proved effective even with imperfect corporate-governance arrangements. In the other states, one or the other type of protectionism was applied but did not prevent a surge in imports. Serbia’s transition In Serbia’s case, discontinuity with the self-management system was for a period observed mainly as an increase in the role of the state, somewhat similar to the soviet system. The Serbian constitution adopted in 1990 – before the breakup of the federation – declared all types of ownership equal. But the state nationalised public utilities and a number of large companies, while a subsequent law on privatisation allowed a voluntary transition from social to private ownership. Moreover, Serbia took over an already-existing law that had legalised the setting-up of private firms from the end of 1980s. Indeed, once private ownership became legal, numerous private companies were registered immediately. The private sector was soon dominant in terms of the number of firms, but the state and social sectors remained much more important in employment terms. Voluntary privatisations were not very popular. Selling companies to outsiders was resisted by insiders (employees and managers), who saw it as a threat to their employment, while insider privatisation was regarded as a formula for taking over risks without any strong expectation about returns. It is hard to say if this scepticism reflected a real lack of interest in ownership or if it resulted from deteriorating economic conditions and growing state SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 92 – ANNEX. RIGHTS AND RISKS: LABOUR MARKET CHALLENGES IN A POST-SELF-MANAGED ECONOMY interventions. Also, some insider privatisations produced mixed ownership, whereby – in typical cases – a private firm owned by managers would create a joint company together with a socially-owned firm, and then channel assets from the socially-owned to the private firm. Such methods could also involve state-owned companies. As a result, privatisation was widely seen as a way to steal and cheat employees out of their previous ownership rights. Throughout the 1990s, the state did not allow socially or state-owned firms to fire workers. The state thus took over the obligation to pay their wages directly or indirectly, which it could not always do on time. As the economic situation deteriorated, the health and social benefits associated with employment could seldom be supplied without individual participation in money or in-kind. Thus, employees’ rights became increasingly nominal while their risks became increasingly real, though unrelated to investment decisions. Many were listed as formal employees while making their living in the informal economy. In sum, the period between 1990 and 2000 saw a transformation from predominantly social ownership to a mixture of state, social and private sectors, both formal and informal. Although many firms remained under social ownership, this could not be described as a persistence of the rights and risks that had previously been associated with this ownership form, considering that few legal or state institutions were functioning by the time of the regime change in 2000, when the economy had been devastated by several wars and international isolation. To the extent that traces of the previous system could still be found, this was mostly due to institutional inertia and obstacles faced in reforming them. In any case, the legacy of socially-owned and self-managed enterprises cannot explain the many economic disasters of the 1990s: the decline of GDP, deindustrialisation, destruction of the banking system, rising unemployment and the growth of the informal economy. Nor did it require the above-mentioned increase in state ownership or the accommodative economic policies. Policies and institutions Comparing the experiences of the different successor states, it becomes clear in retrospect that the legacy of political and economic mismanagement was more important, and more problematic, than that of social ownership and self-management. The performance of the socially-owned and self- managed Yugoslav economy had always depended rather less on its institutions than it did on policies, among which those relating to external trade and finance played a critical role. Throughout its history, socialist Yugoslavia relied significantly on foreign financial support. Much of its economic history can be understood SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 ANNEX. RIGHTS AND RISKS: LABOUR MARKET CHALLENGES IN A POST-SELF-MANAGED ECONOMY – 93 only if its changing foreign economic relations are taken into account. For example, the crucial reforms of the mid-1960s were driven largely by the need to take up foreign loans – from sovereign lenders or international financial institutions – as a substitute for foreign aid, which had been substantial until then. This was found to require a liberalisation of foreign trade, and, later, a commercialisation of banks and financial markets. Similarly, the need to meet foreign obligations required firms to become efficient. When this put pressure on the labour market, a relief was found in emigration, which led, in turn, to adjustments in the banking system so that it could handle growing amounts of remittances. This involved the commercialisation of banks, which eventually were given access to the international market for private credits. With the growing role of the Deutschmark and the reliance on the IMF, fixed exchange rates became a preferred policy, leading to a trend of nominal convergence due to Germany’s lower inflation. Yugoslavia’s inflation may have served to attenuate the fiscal deficit, though the size of the latter is hard to assess due to a lack of reliable data. Some labour market developments can thus be explained by macroeconomic trends. Labour markets also displayed important regional differences, with low unemployment rates in Slovenia and Croatia and higher rates elsewhere. In other words, Yugoslavia was not an optimal currency area. The reliance on the Deutschmark tended to distort wages in the less-developed republics, with negative consequences for their competitiveness. It appears likely with hindsight (despite a lack of relevant data) that wages were in line with productivity in Slovenia, and, probably, in Croatia. But wages were often inflated in the rest of the country including Serbia, with negative effects on the competitiveness of the Serbian economy. Indeed, Serbia and Macedonia tended to show significant trade deficits, while Slovenia and Croatia had more balanced trade balances. To the extent that Serbia’s higher unemployment was due to an excessive price level, it can thus be explained by a misaligned exchange rate. Labour market rigidities may also have reduced Serbia’s competitiveness, but the nature and importance of these rigidities have yet to be determined. Transition and labour markets In the labour market as in the economy in general, comparisons between for example Slovenia and Croatia suggest that the performances do not depend primarily on the structure or inertia of labour market institutions; they depend rather more on initial conditions and on macroeconomic policies. Before commenting on this hypothesis, a general comment on the behaviour of labour markets in transition may be in order. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 94 – ANNEX. RIGHTS AND RISKS: LABOUR MARKET CHALLENGES IN A POST-SELF-MANAGED ECONOMY If the experience of Central European transition economies is anything to go by, most of the growth in GDP and industrial production is due to rising productivity. After initial declines in output and employment, the former recovers quickly and decisively while the latter stagnates for a longer time. This suggests an improved utilisation of those labour resources that remain employed, while too few new jobs are created. Croatia’s experience corresponds to this Central European pattern of productivity-driven growth. But in Slovenia, employment declined only a little and unemployment has remained low for a transition economy. This is remarkable not least because Slovenia has a relatively small agricultural sector and cannot hide unemployment there. Moreover, the labour market is probably more rigid in Slovenia than in Croatia. So what caused the underperformance of transition economies’ labour markets? Serbia is still at the stage when an employment decline can be expected. The transition invariably leads to job losses not only during the transitional recession, but also during an initial period of recovery, when privatisation and foreign investments lead to higher production mainly by enhancing productivity. Many of the job losses now observed in Serbia are due to closures of unprofitable firms and to rising productivity in restructured ones. As a sign of labour market rigidity, it has been pointed out that real wages tend to grow while employment declines. The labour market disequilibrium may reflect excessive wage increases, which insiders might be able to extract as a result of institutional rigidities that bar outsiders from competing with them for existing jobs. However, the recent wage growth in Serbia started from a very depressed level, and it was strongest in service sectors whose employment trends, while mostly negative, showed smaller negative values than the average for all sectors.8 It is also not clear whether the assumption about labour market rigidity goes together with the large increases in labour productivity. In any case, while wage growth may occasionally match or even exceed labour- productivity growth, it is likely as a rule that real wages do not grow as fast as productivity does. The long-term growth trend should be a little lower for real wages than for labour productivity in order to encourage investments and job creation. This will be important not least in Serbia, 8. By comparison, wage growth in Serbia since 2001 has been relatively moderate in most of the industry sectors that experienced steep employment declines (cf. Chapter 2, Table 2.4). SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 ANNEX. RIGHTS AND RISKS: LABOUR MARKET CHALLENGES IN A POST-SELF-MANAGED ECONOMY – 95 where much of the employment decline can be attributed to a worrying scarcity of new start-up firms. Another possible problem is that public-sector wages may grow faster than productivity, but here the evidence is mixed. Macedonia, like some countries outside the former Yugoslavia, introduced a cap on public-sector wage growth; but this had limited effect on employment there. Elsewhere, public-sector wages have proved to be a problem. But perhaps the main reason for this has been the slow democratisation. Political instability and insufficient transparency make it difficult to control public-sector wage growth. In Croatia, for instance, fiscal consolidation has been possible only in the last few years. Poor governance can also contribute to low labour turnover in the public sector. Serbia’s lack of political stability has clearly had effects on the speed of the public sector’s transformation, and probably on its relative wages as well. In conclusion, the experience of transition economies in Central Europe and the Balkans suggests the following as key causes of low employment: 1. Rising productivity in existing enterprises, especially for tradable goods and services. 2. Slow growth of the new private sector. 3. Decreased employment in the public sector, combined with low labour turnover and the ability of insiders to restrict outsiders’ access to employment. 4. Upward pressure on the real exchange rate due to growing inflows of foreign investments and remittances, often leading to currency substitution and uncompetitive prices and wages. 5. Mismatch between labour supply and demand resulting from population ageing and the skill structure of the labour force. In addition to these factors, rigidities in product and labour markets are likely to contribute. However, with the possible exception of some sectors, the role of labour-market rigidities is probably moderate as long as high unemployment prevails. Social justice and labour markets Serbia’s new constitution, adopted in October 2006, includes a general declaration committing the state to social justice. This is followed by several provisions of relevance to the labour market that indicate a dependency on previous constitutions. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 96 – ANNEX. RIGHTS AND RISKS: LABOUR MARKET CHALLENGES IN A POST-SELF-MANAGED ECONOMY The new constitution does away with social ownership by not mentioning it in a list of possible ownership types. But another article states that existing socially-owned firms will be transformed into other of types of ownership in accordance with the law. (Such laws already existed.) It is also stipulated that all ownership forms will be equally treated. It is not fully clear that the latter applies to social ownership, although this appears important in view of the number of people still employed in such firms. Other provisions grant citizens a right to work, with equal rights to all jobs, and envisage job subsidies to support these rights. The constitution stresses the importance of social dialogue and collective bargaining about wages, and it provides framework rules about social insurance and healthcare. In all these areas, the constitution thus stipulates ambitious goals, while – like similar laws in many countries – it leaves numerous questions open concerning the practical implications. Against the background of the actual situation in labour and product markets, such constitutional promises can be expected to have limited impact in the near future. The present labour surplus means that the right to employment is at best a distant goal. Collective bargaining occurs mainly in the public sector and some large firms, with limited impact on wages even there. What these constitutional provisions primarily seem to reflect is an understanding – to an extent inherited from the socialist past – that there is a demand for social justice, job security and social protection. This results in high claims to rights, especially the right to work and to be socially secure. But there is less understanding of how risks can be distributed in a market economy. Though Serbia’s labour market is quite flexible in reality, not least as a result of a labour surplus that reduces workers’ bargaining power, the legal language is strong on social solidarity and labour rights. Policy options Because Serbia is still at the beginning of transition, significant restructuring has yet to happen and will cause further job losses. The private sector is getting stronger, but there are too few new private firms to compensate for the downsizing of older firms. The paramount goal must be to improve the conditions for creation of new and more productive jobs. The small share of new firms in total labour demand may have contributed to a tendency for real wages to outpace productivity after 2000, especially in the public sector but also in private service sectors. In principle, a disconnect between wages and productivity in non-traded sectors (the “Balassa-Samuelsson effect”) suggests that an overvalued real exchange rate is causing inflationary pressures. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 ANNEX. RIGHTS AND RISKS: LABOUR MARKET CHALLENGES IN A POST-SELF-MANAGED ECONOMY – 97 Such a policy stance makes it unnecessarily expensive to create new jobs. It implies an inefficient distribution of rights and risks, with, on the one hand, excessive wage growth for insiders (strong rights, few risks), and, on the other, insufficient job creation and higher unemployment for outsiders (high risks, few rights). The functioning of the labour market could be improved by a set of policy measures that target competitiveness, establish an appropriate relation between productivity and wages and an appropriate combination of risks and rights. As in most transition economies, unemployment rates are particularly high for youths (first-job seekers), the unskilled, women, and those who have been laid off. Members of these “outsider” groups are unlikely to demand as high wages as do “insiders” (often skilled middle-aged men), who frequently have jobs in the public sector or in firms that face relatively little competition, permitting them to demand higher wages without taking much account of product markets. As a result, a significant proportion of the employed workforce – apart from small-scale services and the informal economy – may be in a position to demand higher wages than would be realistic for “outsiders”, causing little risk for themselves but negative effects for the economy as a whole. In this situation, the policy options boil down to three: 1. Increase competitiveness. Avoid misalignment of the real exchange rate, which is problematic in combination with the present structure of labour demand. 2. Invest in human capital. It has proved difficult for Serbia to compete in low-skill sectors. 3. Increase product-market competition. Many sectors are monopolistic and competition policies are weak or non-existent, permitting companies to pocket profits without taking risks. Some further labour market reforms could be envisaged with specific view on the distribution of rights and risks. Clearly, the new constitution goes in the direction of a stronger affirmation of rights to work, and some provisions can possibly be understood as a promise to limit the market’s influence on wage setting. This is understandable, given the social pressures and the underlying uncertainty facing an economy in rapid transition. In practice, however, unemployment works as an unintended source of labour market flexibility in a situation where employment and wages are not sufficiently responsive to market forces. A better solution, if it can be achieved, would be to adopt reforms that permit the necessary flexibility within the framework of regular employment. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 98 – ANNEX. RIGHTS AND RISKS: LABOUR MARKET CHALLENGES IN A POST-SELF-MANAGED ECONOMY Conclusion Employees should take labour market risks and employers should take product market risks. Workers need to be protected by rights when they are affected by changes in product markets, which they cannot control. But the two types of risk are related, and sometimes impossible to distinguish. Moreover, enterprise managers must have the right – indeed, the duty – to concentrate on their business, which is to compete in product markets, and to make whatever changes are necessary to achieve this goal. The legacy of socialist labour markets and of worker-managed product markets is not helpful in this respect. The institutional path-dependence is not strong in Serbia, but there is a persistent confusion about the proper distribution of risks and rights in a market economy. It is unfortunate that its macroeconomic policies have been most path-dependent, with inflation as a chief constraint on real wages and a fixed exchange rate to discipline monetary and fiscal policy. This combination continues to cause imbalance in the economy and in the labour market. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 REFERENCES – 99 REFERENCES Alfieri, A. et al. (2005), Issue Paper: Definition of Remittances and Relevant BPM5 Flows, United Nations Statistics Division, http://unstats.un.org/unsd/tradeserv/TSG%2002-05-Paris/tsg0502-16.pdf accessed 6 April 2007. Bajt, A. (1988), Samoupravna oblika družbene lastnine (Self-managed Form of Social Property), Globus, Zagreb. Bassanini, A. and Duval R. (2006), “Employment Patterns in OECD Countries: Reassessing the Role of Policies and Institutions”, OECD Economics Department Working Paper, No. 486, OECD, Paris. Belev, B. (ed.) (2003), The Informal Economy in the EU Accession countries: Size, Scope, Trends and Challenges in the Process of EU Enlargement, Centre for the Study of Democracy (in partnership with the World Bank and the Bertelsmann Group (www.csd.bg – publications), pp. 139-174. Center to Liberal-Democratic Studies (2002), Competition Policy in FR Yugoslavia: Market structures and Competition Institutions, Belgrade. Đukić, P. (2006), “The Organised World of Labour in Serbia Today: Between Economic Reforms and Populism”, South-East Europe Review for Labour and Social Affairs, No. 4/2006, Hans Böckler Stiftung, Baden-Baden. Eurostat (2005), “Eurostat’s Tabular Approach to Exhaustiveness: Guidelines”, mimeo, 15 March, Luxembourg. Furubotn, E. and S. Pejovitch (1972), “Property Rights and Economic Theory”, Journal of Economic Literature, Vol. 10. No. 4. Gonzaga, G. (2004), “Labor Turnover and Labor Legislation in Brazil”, Economía, Latin American and Caribbean Economic Association (LACEA), Rio de Janeiro. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 100 – REFERENCES Government of the Republic of Serbia (2003), “Poverty Reduction Strategy Paper for Serbia”, Belgrade. Government of the Republic of Serbia (2007), “Memorandum on the Budget and Economic and Fiscal Policy for 2008 Including Projections for 2009 and 2010”, 5 July, www.mfin.sr.gov.yu/eng/631/, accessed 3 November 2007. Grubb, D. (2007), “Audit du service public de l’emploi au Luxembourg”, OECD Social, Employment and Migration Working Paper, No. 59, OECD, Paris. Horvat, B. (1982), The Political Economy of Socialism: A Marxist Social Theory. M.E. Sharpe, Inc., Armonk, NY. Jasarević, S. (2003), “Labour Dispute Settlements in Serbia”, South-East Europe Review for Labour and Social Affairs, No. 3/2003, Hans Böckler Stiftung, Baden-Baden. Ministry of Economy and Regional Development (2007), Mere aktivne politike zaposljavanja, www.merr.sr.gov.yu/dokumenta/Mere%20aktivne%20politike%20zaposlj avanja.doc, accessed 22 November 2007. Ministry of Labour and Social Policy – MLSP (2005), National Employment Strategy 2005-2010, Belgrade. Ministry of Labour and Social Policy – MLSP (2006), National Employment Action Plan for 2006-2008, Belgrade. Nagyszombat, A.G. (2008), “Measurement of the Non-observed Economy in Croatia”, Paper No. 4 for the EFTA-OECD Workshop on the Non-observed Economy for the Western Balkan Region in February 2008, OECD, Paris. National Employment Service – NES (2005), Change Strategy of the National Employment Service 2006-2008, Belgrade. OECD (1997), Labour Market Policies of Slovenia, Paris. OECD (2002a), China in the World Economy: The Domestic Policy Challenges. Paris. OECD (2002b), Handbook for Measurement of the Non-observed Economy, Paris. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 REFERENCES – 101 OECD (2003a), OECD Economic Surveys: The Federal Republic of Yugoslavia, Paris. OECD (2003b), Labour Market and Social Policies in the Baltic Countries, Paris. OECD (2004), OECD Employment Outlook, Paris. OECD (2005a), OECD Economic Surveys: Brazil, Paris. OECD (2005b), OECD Employment Outlook, Paris. OECD (2006), OECD Employment Outlook – Boosting Jobs and Incomes, Paris. OECD (2007), OECD Employment Outlook, Paris. Ratkovič, D. (2008), “Measurement of the Non-observed Economy in Serbia”, Paper No. 8 for an EFTA-OECD Workshop on the Non-observed Economy for the Western Balkan Region in February 2008, OECD, Paris. Tonin, M. (2007), “Minimum Wage and Tax Evation: Theory and Evidence”, William Davidson Institute Working Paper No. 865, Ann Arbor, MI. Upchurch, M. (2006), “Strategic Dilemmas for Trade Unions in Transformation: The Experience of Serbia”, South-East Europe Review for Labour and Social Affairs, No. 4/2006, Hans Böckler Stiftung, Baden-Baden. Vukovič, D. (2005), “Migrations of the Labour Force from Serbia”, South East Europe Review, No. 4, p. 139ff. Ward, B. (1958), “The Firm in Illyria: Market Syndicalism”, American Economic Review, Vol. 48, No. 4, pp. 566-589. Winkler, R. 1997, “The Size and Some Effects of the Underground Economy in Mexico”, in O. Lippert and M. Walker (eds.), The Underground Economy: Global Evidence of its Size and Impact, The Fraser Institute, http://collection.nlc-bnc.ca/100/200/300/fraser/underground/index.html accessed 16 October 2004. World Bank (2003), Serbia and Montenegro Poverty Assessment, Vol. I: Executive Summary and Vol. II: Main Report, Washington D.C. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 102 – REFERENCES World Bank (2004), Serbia Investment Climate Assessment, Washington D.C. World Bank (2005), Doing Business in 2006: Creating Jobs, Washington D.C. World Bank (2006), Doing Business in 2007: Serbia, Washington D.C. World Bank (2006), Serbia: Labour Market Assessment, Washington D.C. World Bank (2007), Doing Business in 2008: Serbia, Washington D.C. SERBIA: A LABOUR MARKET IN TRANSITION – ISBN-9789264045798 © OECD 2008 OECD PUBLICATIONS, 2, rue André-Pascal, 75775 PARIS CEDEX 16 PRINTED IN FRANCE (81 2008 08 1 P) ISBN-978-92-64-04579-8 – No. 56229 2008 OECD Reviews of Labour Market and Social Policies SERBIA A LABOUR MARKET IN TRANSITION To catch up with more advanced economies, Serbia urgently needs to improve the functioning of its labour market. Despite many reforms, new business growth until now has been far too slow to compensate for job losses elsewhere. Recent reforms of labour law should be followed up by further efforts to improve the climate for business and productive work. Labour regulations must be ﬂexible, but they should also be enforced more consistently. For all this to happen, it is essential that an effective social dialogue can take place and that it encompasses expanding and declining segments of the labour market. The full text of this book is available on line via these links: www.sourceoecd.org/employment/9789264045798 www.sourceoecd.org/transitioneconomies/9789264045798 Those with access to all OECD books on line should use this link: www.sourceoecd.org/9789264045798 SourceOECD is the OECD online library of books, periodicals and statistical databases. For more information about this award-winning service and free trials, ask your librarian, or write to us at SourceOECD@oecd.org. ISBN 978-92-64-04579-8 ����������������������� 81 2008 08 1 P -:HSTCQE=UYZ\^]:
Pages to are hidden for
"OECD Reviews of Labour Market and Social Policies Serbia 2008"Please download to view full document