SP DISCUSSION PAPER NO.0310 - ~26307 IU3nemployment Benefit Systems in Central and Eastern Europe: A Review of the 1990s M ilan Vodopivec, Andreas Worgotter and Dhushyanth Raju March 2003 rot i n RKETS, PENSIONS, SOCIAL ASSISTANCE T HE W OR L D B A N K UNEMPLOYMENT BENEFIT SYSTEMS IN CENTRAL AND EASTERN EUROPE: A REVIEW OF THE 1990s Milan Vodopivec', Andreas Wbrg6tter,*' and Dhushyanth RajunS* 1. INTRODUCTION One of the most conspicuous consequences of the transition of former socialist economies has been the emergence of large-scale, open unemployment - a phenomenon unheard of before the transition. These economies have thus been confronted with the difficult task of protecting the unemployed while avoiding undue fiscal costs and minimizing work disincentives created by such protection. Faced with the prospect of high unemployment, many transition economies introduced traditional, OECD-style unemployment insurance programs. The purpose of this paper is to evaluate those programs by examining their distributive and efficiency effects. To address distributive issues (an aspect so far neglected by researchers), we analyze data from household expenditure surveys and try to answer the following two questions: Which groups of workers benefited most? How have these programs changed the pre-transfer distribution of income? To examine efficiency effects, we review the existing literature. The questions that have received the most attention are: Have unemployment benefits created work disincentives? In particular, have more generous replacement rates and longer benefit durations affected the length of unemployment spells? We also examine whether the introduction of unemployment benefit programs has helped to speed up enterprise restructuring. (We found no study that examined the possible positive effects of * The authors would like to thank Gordon Betcherman, Carlos Silva-Jauregui, Wayne Vrornan and two anonymous referees for valuable conmments on an earlier version of the paper, Jakob Tomnse for excellent research assistance, and Leszek Kucharski, Gyula Nagy, and Jiri Vecemik for providing information on expenditures and participants of unemployment benefit programs in Poland, Hungary, and the Czech Republic, respectively. The views expressed in the paper do not necessarily reflect the opinion of the World Bank and the OECD. * Mr Woergoetter, OECD contributed to this paper while he was affiliated with the Institute for Advanced Studies in Vienna and the Central European University in Budapest. He expressed his private views which have no official character with respect to his current affiliation. *-- When writing the paper, Dhushyanth Raju, Cornell University, was a World Bank consultant. I unemployment benefit systems - for example, better job matching as reflected by increased reemployment earnings.) In this study, we would like to shed light on the question of the suitability of traditional unemployment insurance programs for transition economies, including those with poor administrative capacities and low incomes per capita. This concem relates to distributive issues: unemployment insurance benefits may be regressive, that is, mostly paid to better-off workers. Moreover, because of the lack of administrative capacity in these countries, there may be significant "leakage" of these benefits. And unemployment benefits may also hurt efficiency. They may create work disincentives and, by increasing the bargaining power of workers, contribute to higher equilibrium unemployment and increased employment in the informal sector. The paper is organized as follows. We first describe how the reduction of output translated into the increase of unemployment, and discuss the nature of unemployment in transition countries as it evolved in the first ten years (Section 2). We focus on Central and East European as well as Baltic countries, which constitute a relatively homogeneous group. We emphasize differences in the responsiveness of employment reduction to output decline across these countries, which were undoubtedly at least partly produced by different policy choices. We then describe formal unemployment benefit progms introduced by transition countries (Section 3) and, in the core section, evaluate these systems by presenting their distributive and efficiency effects (Section 4). We conclude with a summary of main findings and a discussion of emerging policy issues. 2. UNEMPLGYMENT AND TRANST(N Transition reforms have drastically reduced output and severely affected employment.' The cumulative GDP decline was about 25-35 percent for Central and Eastem European (CEE) economies and 40-50 percent for the Baltic republics (Table 1). Growth tumaround was first achieved by Poland (1992), followed shortly thereafter by the Czech Republic and Slovenia (1993). With the exception of the Czech Republic, Romania and ' Output decline was predominately related to supply side shocks, and long-standing structural imbalances under the socialist regime (see Holmann et al, 1995). 2 Bulgaria, GDP in all countries grew in the late 1990s; still, by 2000, GDP levels surpassed their respective 1989 levels only in Poland (by a substantial 26 percent), Slovenia, Slovakia, and Hungary. How did the reductions in output affect employment and, ultimately, unemployment? Some countries protected their workers from unemployment by reducing average wages and keeping reductions of the workforce to a minimum even in the wake of output reductions. Another way was through the use of government subsidies to promote early retirement. This section examines how labor market stocks adjusted to the decline of output. It describes both employment responses to output decline, as well as trends in unemployment. It also analyzes the structure of unemployment by duration, age and gender, which has important implications for the design of income support systems for the unemployed. Table 1: GDP Index (1989 = 100) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Bulgaria 90.9 80.3 74.4 73.3 74.6 76.2 68.5 63.7 65.9 65.9 69.5 Czech R. 98.8 87.4 84.6 85.1 87.8 93.4 96.9 97.2 95.0 95.0 98.1 Estonia 91.9 79.4 68.1 62.0 60.8 63.4 65.8 72.8 75.7 75.7 81.1 Hungary 96.5 1 85.0 82.4 181.9 84.3 1 85.5 86.6 90.6 95.2 98.1 101.7 Latvia 102.9 92.2 60.0 51.1 51.4 51.0 52.7 57.2 59.2 60.1 64.2 Lithuania 95.0 89.1 70.1 58.9 53.3 55.2 57.9 62.2 65.4 65.4 67.9 Poland 88.4 82.2 84.3 87.6 92.1 98.6 104.6 111.8 117.1 121.2 126.0 Romnnia 94.4 82.2 75.0 76.1 79.1 84.7 88.2 82.1 76.1 73.0 74.3 Slovakia 97.5 83.3 77.9 75.0 78.6 84.1 89.6 95.4 99.6 101.4 103.6 Slovenia 95.3 86.8 82.0 84.3 88.8 92.5 95.7 100.1 104.0 107.6 112.5 Source: Central European Countries' Employment and Labour Market Review, EUROSTAT, Theme 3, 1999-1; CANSTAT Statistical Bulletin No.2/2002. 2.1 Responses of employment to output reductions Reductions of output invariably reduced employment and increased both the number of the unemployed and inactive individuals. But the mode of adjustment differed significantly across countries, both in terms of how strongly employment was affected and which non-employment destinations were chosen. While all countries reduced their employment by less than they reduced their output, there are significant differences among them. Mencinger (2000) estimates that in 1989-97 period, the elasticity of employment to 3 output was high in Bulgaria, amounting to 0.9; in a medium range of 0.4 to 0.6 in Hungary, Poland, Slovakia and Slovenia; and low in the Czech Republic, amounting to 0.3 (his estimate for Romania is also low but insignificant, and he finds low elasticities for Russia and Ukraine as well). Why are these elasticities so different? Apparently, some countries used the approach of "job preservation," that is, of keeping open unemployment low by discouraging labor shedding, thus increasing hidden unemployment. For example, Slovenia used explicit "employment preservation" subsidies to prevent increases in unemployment. Another example is Russia, where in the early, transition period, various practices (resulting from peculiar corporate governance) kept unemployment at a low level (see, for example, Gaddy and Ickes, 1999). Although a thorough evaluation of "job preservation" approaches is outside the scope of this paper, we provide a few comments. While job preservation is appealing from a fairness point of view, it may have serious efficiency consequences. For example, Haltiwanger and Vodopivec (2002) and Vodopivec (2000) find a much faster job reallocation pace and even more favorable outcomes for young workers in Estonia, a well-known radical reformer, than in Slovenia. On a more general level, Caballero and Hammour (2000) show that labor reallocation - the central issue of the transition - can be greatly hampered by poor institutions. Such institutions may reduce cooperation among factors of production and, by favoring some of them over others, contribute to underemployment of the factor which reaps disproportionate gains, impeding technological innovations and job creation. Literature on job creation and destruction also shows that in order for a market economy to function properly, many jobs (perhaps 10 percent of the total stock a year) must be destroyed and new jobs created; and that these newly created jobs are much more productive than those destroyed (see, for example, Davis et al, 1996). It is important to bear in mind, however, that the analysis of stocks, and even of worker flows independent of their driving forces, canmot be taken as a basis for evaluation of the success of labor market adjustment. Interestingly, Russia's labor market adjustnent in the early 1990s has been praised by Layard and Richter (1994). They point to a high hiring rate as a cause of low inflow into unemployment and to the significant changes in the sectoral 4 structure of employment. But such a positive evaluation may be reversed once the extremely low job creation rates of Russia are taken into account (see Acquisti and Lehmann, 1998). Indeed, the rate of unemployment is not a good proxy for the scale of worker and job reallocation: the same rate of unemployment is consistent with very different labor market characteristics, and thus, with a dynamic or a static labor market (see Blanchard and Portugal, 1998, for the comparison of the U.S. and Portugal). Moreover, note that large worker flows do not necessarily mean that labor is being reallocated - that is, that jobs are being destroyed in one firm or sector and created in another. Undoubtedly, developments on the job creation and job destruction front are crucial for evaluating the success of labor reallocation in transition economies.2 2.2 Responses of unemployment to employment reductions Above we showed that the output decline led to substantial reductions in employment - were these reductions absorbed by increases in unemployment or in inactivity, or both? Interestingly, there are considerable differences in the intensity of the use of each channel across transition economies. In some, adjustment affecting unemployment has been less intense: for example, in the Czech Republic, Estonia, and Hungary, unemployment increased by 27 to 42 per reduction of employment by 100. In contrast, the reduction of employment by 100 resulted in a much larger increase in unemployment in some other transition economies - by 85 in Poland, by 75 in Bulgaria, and by 66 in Slovakia (Blanchard, 1997, p.1 1). In Slovenia, there has been virtually no increase in survey unemployment (but registered unemployment increased by 44 per reduction of employment by 100 over the period 1990-2001). The ability to channel redundant workers to inactivity depended partly on the scale of adjustment; obviously, small adjustments could have been achieved by regular outflows from the labor market. But outflows to inactivity were undoubtedly partly policy driven: some countries tried to avoid unemployment by sponsoring early retirement or allowing more 2 There is a substantial (and growing) anount of literature on this topic: see Acquisti and Lehmnn (1998) for evidence on Russia; Bilsen and Konings (1998) on Bulgaria, Hungary, and Romnania; Bojnec and Konings (1998) on Slovenia; Haltiwanger and Vodopivec (2002) on Estonia, and Konings, Lehmann, and Schaffer (1996) on Poland. 5 workers to withdraw from the labor market by claiming disability benefits. As shown in Figure 1, Poland, Hungary, Slovakia and Slovenia have spent considerable resources on promoting early retirement (in 1992, their expenditures on early retirement reached 0.8 percent of GDP). As we show below for Slovenia, the early retirement route, however, proved both fiscally expensive as well as ineffective as far as promoting the employment of young workers. gmingre I1 EZPeladnAtres on Early aetrememt (¢as percemt of GlD)P) 0.841- -;X i-;0- 0 . 8 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . . H ungary 0.6 -..- Poland 0 4~ . . \ . - Slovak R. 0.2 -- 1 1 - > Slovenia 90 91 92 93 94 95 96 97 Source: Employment Outlook, OECD, various issues, Yearly Work Report of Employment Office of Slovenia, various issues. Experenmce wMt1h early retnyemenat: these' of 0 $oDvemfi To protect workers from the increased hardship brought about by transition reforms - and possibly to "make room" for the -employment of young workers - the Slovenian govermenet subsidized early retirement. Women qualified for early retirement at the age of §°, and men at the age of 55, five years before their respective regular retirement ages. Those who qualified had to have sufficient years of service and buy missing pension credits - at a price that had no relation to the actuarially fair price. As a rule, employers paid the missing pension credits for early retirees, and the government compensated employers for about 50 percent of the costs. Early retirement was certainly a good deal for workers: pension levels for early retirees were only slightly reduced (by one percentage point of fall pension for each missing year of pension credits), with the reduction effective only until reaching normal retirement age. In 6 addition, pension levels for newly retired workers tended to be better protected from inflation than were wages, and for some groups of workers, pension levels at retirement even exceeded wages received immediately before retirement. It thus comes as no surprise that the early retirement policy was very effective. The proportion of those employed at the beginning of the year who retired during the year rose significantly in both 1990 and 1991. For example, the proportion of women with 30-34 years of experience who retired during the year increased from 0.164 in 1989 to 0.425 in 1990 and 0.40 in 1991 (see Orazem and Vodopivec, 1995). The story for men is the same, but with a five year lag. The sharp increase began five years earlier in the experience profile for women than men, coinciding with the five year difference in minimum age and experience requirements for the receipt of pensions. This produced dramatic employment reductions for workers in the highest experience groups (measured by the length of total employment), accompanied by equally dramatic wage increases for persons of these groups who remained working. By 1992, employment of males with 30-34 years of experience fell to 72 percent of the 1987 level, while employment for those with 35 or more years of experience fell to 33 percent of the 1987 level. Similar reductions were seen with women's employment, but they began five years earlier, at 25-29 years of experience. Apart from being expensive, the early retirement program failed to "make room" for the employment of young workers. The share of workers under the age of 20 fell from 34 percent in the late 1980s to 1.5 percent in 1992 (Vodopivec, 2002). Estonia offers a sharp contrast: although the govemment did not sponsor early retirement (and pensionable age was even increased early in the transition), the share of employed workers under 20 years increased early in the transition (from 3.1 in 1989 to 3.5 in 1993). Obviously, employment opportunities of young workers are affected by other, much more powerful forces than those induced by early retiremnent - the forces connected with job creation in general.3 In fact, if 3 Employment protection legislation - the Slovenian one being much stricter than the Estonian one -- is a plausible candidate to account for job creation and destruction. Namely, the model of Blanchard (1998) shows that higher employment protection costs lead to impaired access to jobs for narginal groups of workers (productivity of these workers before hiring is not easily revealed and therefore their probability of being hired in the presence of large firing costs is lower). 7 younger workers are complements for - and not substitutes of - older workers, early retirement programs may have a negative effect on the employment of young workers. 2.3 The smrge of ueploymera Perhaps the most dramatic development in the labor market of transition economies has been the emergence of large-scale unemployment. Below we examine the evolution of overall unemployment as well as its structure by age, gender, region, and duration of unemployment spells. From an income support perspective, not only the level, but also the structure of unemployment is of interest: different groups may have different escape rates from unemployment, and, hence, may require specific policy responses. $oirvey nmemlpoymeima. Labor force surveys show persistent and, in some countries, very high levels of unemployment (Table 2).4 With a few exceptions (the Czech Republic, Slovenia, and Romania), unemployment rose to double digit levels. In counties with a large initial increase, unemployment started to fall in the second half of the 1990s (but in Poland and Slovakia, it than again increased by 2000). Table 2: memplloymemt Riee Labon IFource Snveys) 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Bulgaria 21.4 20.5 14.7 13.7 15.0 16.0 14.1 18.7 Czech R. 3.9 3.8 4.1 3.9 4.8 6.5 8.5 8.8 Estonia 1.5 3.7 6.5 7.6 9.7 10.0 9.7 9.6 11.8 13.5 Hungary 9.3 11.9 10.7 10.2 9.9 8.7 7.8 7.0 6.6 Latvia 18.9 18.3 14.4 13.8 13.9 14.4 Lithuania 17.4 17.1 16.4 14.1 13.5 10.4 15.9 Poland 13.7 14.9 16.5 15.2 14.3 11.5 10.6 12.6 16.6 Romania 8.2 8.0 6.7 6.0 6.3 6.9 7.7 Slovakda 12.2 13.7 13.1 11.1 11.6 11.9 16.0 19.1 Slovenia 7.3 8.3 9.1 9.0 7.4 7.3 7.4 7.9 7.5 7.1 Source: Same as Table 1. We rely on survey data, because data on registered unemployment is likely to be distorted. There are numerous country specific incentives to register which invalidate cross-section comparisons, and these incentives may vary through time, which also makes comparisons through time unattractive. 8 Evidence suggests that a significant increase in unemployment in a transition economy can be postponed, but not avoided. The Czech Republic, for example, initially kept unemployment low, but in the second half of the 1990s struggled with the consequences of postponed microeconomic restructuring and excessive income increases. Sirnilarly, the relatively low unemployment of Romania largely reflects a lack of enterprise restructuring and labor hoarding. Registered unemployment. Most of the transition countries experienced a rapid increase in the registered unemployment rate to two digit levels (Table 3). While in some countries the rates in the second half of the 1990s started to decline (Hungary, Latvia, Romania, and Slovenia), in some others they grew throughout the 1990s (the Czech Republic, Lithuania and Estonia). Table 3: Registered Unemployment Rate 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Bulgaria 11.1 15.3 16.4 12.8 11.1 12.5 13.7 12 14.1 20.8 Czech R. 4.1 2.6 3.5 3.2 2.9 3.5 5.2 7.5 8.4 8.8 Estonia 5.0 5.1 5.0 5.5 4.6 4.7. 6.7 6.6 Hungary 7.4 12.3 12.1 10.4 10.4 10.5 10.4 9.1 9.7 9.0 Latvia 0.6 3.9 8.7 16.7 18.1 19.4 14.8 13.8 10.7 9.3 Lithuania 0.3 1.3 4.4 3.8 6.2 7 5.9 6.4 7.7 11.4 Poland 11.8 13.6 16.4 16.0 14.9 13.2 10.5 10.4 12.7 14.4 Romania 3.0 8.2 10.4 10.9 9.5 6.6 8.8 10.3 9.2 8.9 Slovakia 12.2 13.7 13.1 11.1 11.6 11.9 18.9 20.6 Slovenia 8.2 11.5 14.4 14.4 13.9 13.9 14.4 14.5 11.6 10.3 Source: EBRD, Transition Report 1999, Employment and Labour Market in Central European Countries 2001/2, EUROSTAT, Theme 3, 2001. It seems that the generosity of benefits affects the rate of registration of the unemployed at employment offices. For example, in Estonia, where the level of benefits until 2003 was extremely low, the registered unemployment rate was much below the survey rate of unemployment. In contrast, Slovenia seems to have introduced the opposite 9 incentives.5 The rate of registered unemployment remained fairly steady during 1995- 98 at around 14 percent - the highest registered level of unemployment in the 10 transition countries in our sample and 6-7 percentage points above the survey rate of unemployment, but it seems that the 1998 reform of unemployment benefit program and stricter monitor of non-recipients helped to reduce the number of registered unemployed in the post-1998 period. The discrepancy between the survey and register numbers is attibutable mainly to the group of registered unemployed who are not considered unemployed according to survey criteria. In 1999, more than half of the registered unemployed did not qualify as unemployed by survey criteria, 70 percent of them because they did not actively search for a job (Employment Office of Slovenia, 2000). 2A4 &SFtMCMe o MmemaPlymew What are the demographic and other characteristics of the unemployed? Given their low mobility, how many of them are long-term unemployed? Are young and old workers disproportionately represented among the unemployed? What is the share of women among the unemployed? How high are regional disparities in unemployment rates? All these aspects have an important bearing on the design of income support programs for the unemployed. Lomg-term unmemipfoyminet. Apart from the high levels of unemployment, long-term unemployment has also become a serious problem across many European transition economies. During the period 1993-2000, the share of the long-term unemployed ranged from a third to more than half, and this share in the majority of countries increased in the late 1990s (Table 4). ln Slovenia, incentives to register are very diverse. As in other countries, registration is a prerequisite for the receipt of unemployment insurance benefits (which consists of cash payments and payments of social security contnbutions for old-age and health insurance) and of employment office services (counseling, training, and erployment subsidies). But in Slovenia, registration at the employment office also brings protection from layoffs for a working spouse; reimbursement of moving expenses connected with reemployment; eligibility for health care services after unerployment insurance eligibility expires; advantages in accessing social housing; ability to enroll in evening post-elementary education (only day-time enrollment is permitted otherwise); subsidies for child care and eligibility for child allowances; and eligibility for voluntary old-age insurance. In addition, newly declared disabled persons waiting to be positioned to new worldng places that correspond to their disability also have to register with employment offices (in early 2000, they constituted about 10 percent of all registered unemployed). 10 Table 4: Shares of Long-term Unemployed* 1993 1994 1995 1996 1997 1998 1999 2000 Bulgaria 53.8 60 65.6 59.9 57.6 54.4 58.3 53.0 Czech R 27.6 29.5 36.6 50.0 Estonia 28.1 39.6 31.8 55.3 45.8 45 42.2 47.3 Hungary 32.2 41.3 45.6 49.8 46.5 44.3 47.9 47.9 Latvia 56 53.0 55.9 Lithuania 38 38.8 52.4 Poland 33.5 38.6 40.5 40 39.1 37.9 41.6 44.6 Romania 44.3 45.2 49.2 Slovakia 30.2 41.6 53.1 52.7 51.1 49.7 47.6 54.7 Slovenia 54.8 62.1 59.0 53.8 59.6 57.1 41.8 62.7 Source: Central European Countries' Employment and Labour Market Review, EUROSTAT, Theme 3, 1999-1; OECD: Country Surveys, Economic Outlook; Employment and Labour Market in Central European Countries 2001/2, EUJROSTAT, Theme 3, 2001. *Unemployment duration exceeding 12 months. One of the reasons for the high increase in the proportion of long-term unemployed workers has been the low probability of transition from unemployment, particularly to employment. Boeri (1996) shows that this probability is much lower in transition than in OECD countries. Individuals who become unemployed may encounter a host of factors which act to lower the probability of transition from long-term unemployment, particularly to employment. Labor shedding unmatched by sufficient job creation contributed to a higher incidence of long-term unemployment in all economies; in addition, generous unemployment benefits may have played a role in some of them as well (see below). A large proportion of the long-term unemployed are relatively unskilled workers (European Commission, 1999). Structure of unemployment by age. Young workers faced more serious impediments in securing and maintaining employment than other groups of workers, including those over 55 years of age. For all countries, the unemployment rates for youth, that is, for those individuals below 25 years of age, were substantially higher than the average unemployment rates (compare Tables 2 and 5), with the trends in the two being quite similar. The Czech Republic was the only country that had single digit youth unemployment rate till 1997, but it increased to 17 percent by 2000. In line with the falling average unemployment rates in the second half of the 1990s, the youth unemployment rate in Hungary, Romania and Slovenia was also falling The countries with the most sever youth unemployment problem in 2000 were Bulgaria, Poland, and Slovakia. 1993 1994 1995 1996 197 2998 2999 2000 Bulgaria 47.0 44.9 37.7 33.5 36.0 36.0 31.3 39.4 Czech R. 7.7 7.9 7.2 8.6 12.4 16.6 17.0 Estonia 11.0 11.6 14.1 16.0 14.4 14.5 22.1 23.7 Hungary 21.3 19.4 18.6 18.0 15.9 13.5 12.3 12.3 Latvia 30.1 29.0 24.9 27.1 23.4 21.2 Lithuania 32.1 31.6 27.4 26.2 22.9 21.3 27.5 Poland 30.0 32.5 31.2 28.5 24.8 23.3 29.6 35.7 Romania 22.5 20.6 20.2 18 18.3 17.3 17.8 Slovakia 25.7 27.6 24.8 20.6 22.4 23.5 32.0 36.9 Slovenia 24.2 22.2 18.8 18.8 17.6 18.3 18.5 16.4 Source: Same as Table 1. * Unemployment rates of workers below 25 years of age (according to labor force surveys). The share of unemployed workers older than 55 years in total unemployment has been low (see Table 6). In all countries of our sample except Latvia, this share has been below 10 percent throughout the 1993-2001 period, and several countries have managed to keep this share below 5 percent (Hungary, Poland, Romania, Slovalda and Slovenia). Talbe 6: Mhanres of 1LmemmpDoyed clldeir 3Enm 05 i lmepRoy emt 1993 2994 1995 1996 1997 2998 2999 2800 2881 Bulgaria 6.6 5.9 4.5 4.2 3.8 5.0 4.6 5.4 7.9 Czech R. 8.2 6.8 6.3 8.5 6.4 5.4 5.3 5.3 5.7 Estonia 6.0 7.6 8.2 8.2 7.8 7.6 7.5 8.8 9.6 Hungary 5.2 4.5 3.1 3.4 4.1 3.9 2.1 3.0 3.6 Latvia 16.6 11.3 9.1 8.2 7.2 8.8 11.5 Lithuania 4.7 4.9 5.0 7.7 7.6 Poland 4.4 3.7 3.8 3.8 3.8 4.4 4.7 4.0 3.6 Romania 1.6 1.9 1.1 1.0 0.9 0.9 1.3 1.6 Slovakia 3.2 2.7 3.0 2.6 3.0 2.8 3.1 3.1 Slovenia* 3.5 4.5 4.2 5.3 2.4 2.1 2.8 4.4 4.9 Source: ILO, LABORSTA, National Statistical Offices of Estonia and Slovenia. *Low reliability because of the small proportion of individuals older than 55 in the sample. $tn Uire of melmpRoymemt 1by germdeir In most countries, the unemployment rate for women is higher than for men, with the notable exception of Hungary where the female unemployment rate was consistently and significantly below the male unemployment rate between 1993-1998. Slovenia, Estonia, and Latvia also reported, at times, lower female unemployment rates than for males (European Commission, 1999). But women's share in 12 unemployment in most countries is lower than men's (except in the Czech Republic and Poland). The lowest women's share is in Hungary, with around 40 percent (Table 7). Table 7: Share of Women in Unemployment 1993 1994 1995 1996 1997 1998 1999 2000 Bulgaria 48.3 47.0 48.1 47.4 47.8 47.0 46.7 46.0 Czech R. 55.9 53.7 52.9 52.7 54.6 56.5 53.2 53.6 Estonia 47.7 49.0 42.7 43.9 45.4 43.3 42.5 41.3 Hungary 39.1 43.3 37.2 39.1 38.6 39.5 40.1 39.3 Latvia 44.3 45.6 48.5 48.4 45.7 44.8 Lithuania 47.1 44.0 43.2 41.4 Poland 51.3 51.2 50.9 51.8 54.0 53.3 49.2 52.0 Romania 49.7 49.6 49.5 48.5 44.1 41.6 43.0 Slovakia 43.8 46.0 46.9 50.7 49.0 47.3 45.4 44.7 Slovenia 42.3 43.5 44.3 44.9 48.6 46.8 47.1 47.0 Source: Same as Table 1. In summary, the low propensity of the unemployed to take jobs has produced a stagnant pool of unemployed, with a large share of long-term unemployed. Among the unemployed there is also a disproportionate share of young workers. Due to a loss of earnings, the unemployed may experience a sharp decline in their consumption expenditure, particularly if they are not compensated for this loss. We focus on income support programs in the rest of the paper. 3. INCOME SUPPORT PROGRAMS FOR THE UNEMPLOYED Prior to the transition, "open" unemployment in East European countries - with the exception of Yugoslavia - was virtually non-existent. Consequently, at the beginning of the transition, income support programs for the unemployed were unavailable.6 In all countries, the network of employment offices existed, but these offices primarily focused on labor exchange for employed workers and were unprepared to offer services for large numbers of unemployed. Anticipating the emergence of widespread unemployment, in the early 1990s (as early as 1989 in Poland), all transition countries enacted legislation for the provision of income support for the unemployed, using the West as their blueprint. This income support included 13 unemployment benefit and social assistance programs, as well as active labor market programs ranging from training to subsidized employment programs to labor-intensive public works programs. From the incipient stages, both unemployment benefit and social assistance programs have undergone several, often radical reforms largely aimed at curbing expenditures in the face of an upsurge in unemployment and severe fiscal constraints, as well as to reduce the work disincentives created by such programs. Reforms commonly comprised of tighter eligibility requirements for the receipt and maintenance of benefits, and a reduction in the length of the maximum potential duration of entitlement. In this section, we first describe the rules of these systems in transition economies and discuss issues arising in their implementation. We then discuss the costs of these systems and compare them to the costs of unemployment benefit programs in OECD economies. 3.1 ]Desc4:pdon of incomDse sMppoFt sys9emsJbv 9e uOem§oy Below we focus on unemployment benefits as they are the primary program of income support for the unemployed, but we also discuss social assistance as an increasing number of the unemployed participate in this program as well. UEme mnpoyment beimeIths Similar to most OECD countries, unemployment benefit programs in transition countries are typically mandatory and cover the majority of emnployed persons, irrespective of industry or occupation (the most notable exception being the self-employed). Benefit levels are earnings-related and the duration of entitlement is tied to previous employment history, but at the same time, benefit floors ensure that the benefits of those at the bottom of the wage distribution do not fall beneath an officially detennmined minimum. In addition, as a significant deviation from the insurance principle, some special groups such as school eavers (in the Czech and Slovak Republics, Estonia, Poland, and Romania) and others are entitled to benefits at fixed rates (Boeri, 1997). Moreover, after eligibility to eamings-related benefits under unemployment insurance expires, the unemployed may continue to receive benefits under a means tested follow-up program called unemployment assistance. Because it pays a 6The exceptions were Yugoslavia, where unemployment insurance existed since the early 1970s, and Hungary, where benefits were offered to redundant workers dismissed in mass layoffs since 1986 (Boeri, 1997). 14 flat rate and is financed from general revenues, Estonia's program was sometimes classified as unemployment assistance (Vroman, 2002).7 Some of the salient institutional features of unemployment benefit systems of transition countries are detailed below (they are summarized in Table 8) 7 In one of the most radical reforms implemented by a transition economy, Estonia substantially revamped its benefit program in 2001, and the first unemployment benefits under the new regime were paid out in January 2003. The program introduced a proper, mandatory public unemployment insurance scheme financed from contributions by workers and employers. It is expected that under the new program unemployment benefits will increase by several times, particularly for better paid workers (in contrast to the flat-fee rate of the previous program, the new benefits are earnings-related). 15 'IPahe 0: IMgoin eDauires of Un¢moDoyment 1Menefl1 Systens In CEEVC (Latest Legl.Slilon In 3o1d) Date Reference Required min Max duration of Relation to individual's gross earnings Unemployment benefit levels (minimum and period employment benefits maximum, expressed in % of minimum wage) record Bulgaria 1989 12 months 6 months 6 months' 100% of last monthly wage for first 100%C month, then 10% less for next 5 monthsb 1991 12 months 6 months 12 months Equal to minimum wage 1992 12 months 6 months 12 months 60%d 90% 140% A1990 22 months 9 months 12 months 60 %d 85% 140% Czech R. 1991 3 years' 12 months 12 months 60% first 6 months none, but 70% of minimum living standard 50% following 6 months (MLS) if not employed before (70% in case of retraining course) 1992 6 months 60% first 6 months none, but 70% of 150-180%f 50% following 6 months minimum wage if not (70% in case of retraining course) employed before 1996 3 yearse 12 months 6 months 60% first 6 months none, but 70% of 150-180% of MLSf 50% following 6 months MLS if not employed (70% in case of retraining course) before 1990 3 yearse 12 months 6 mounths §0% flrst 6 months none (but 70% of T50-180% of MLST 40% foflowing 6 months FtLS oT not enmpRoyed (60% Dn ease of retrnlnnp conurse) beore) lstonln 11991 12 months 180 days 6 months Flat rate, detenrined as 60% of minimum wage 1995 12 months 180 days 6 months (3 months Flat rate, determined as 60% of extensions minimum wage considered on individual basis) 2001 24 months 12 months 12 months 50% n tOhe frst 100 days, 40% of the overage 150 percent of the (effective of the reeelpt, 40% therenfter wage average wage 2003) 16 Table 8: Main features of unemployment benefit systems in CEEC (cont'd) (latest legislation in bold) Date Reference Required minim Maximum duration Relation to individual's gross earnings Unemployment benefit levels (minimum and period employ,ment of benefits maximum, expressed in % of minimum wage) record Hungary 1989 3 years' 18 months 24 monthsg 70% first 6 months 80% since 1990i 300% 60% following 6 months 45% following 12 months 1991 4 years 360 days 70% first half of entitlement period 100%i 50% second half period 1992 70% first half of entitlement period none 200% 50% second half period 1993 4 years 90 days 360 days 70% during phase 1 8600 fiorints 18000 fiorints during 50% during phase lilh phase I 1500 fiorints during phase ll 1997 4 years 90 days 360 days 65%h 90% of minimum 180% of minimum old-age pension old-age pension Latvla 1993 6 months 90% of minimum wage (70% for new 70 % of minimum 140 % of minimum entrants) wage wage Litlhuania 1993 6 months 70 %, later reduced to 60 % and 50% Poland 1989 None None None 70% first 3 months 100% average wage 60% following 6 months 45% afler 9 months 1992-94 1 year' 180 daysi 12 months 2 36% of national average wage none none years in exceptional cases 1997 18 months I year 18 months flat rate amount paid at 378,2cz none none Romania 1996 1 year 6 months 9 months 50-60% for 9 months 75-80% 200% 1 year 1998 1 year 1 year 9 monthsk 50-60% for 9 months 76-92% 210% 17 '1I ble 0: MOnun featnires of emelpRoymeunt beneffe systemns mD CIEE (¢cont'd) (mlnest RegsIgtmom !nn bo2d) Date Reference Required minim Maximum duration Relation to individual's gross earnings Unemployment benefit levels (minimum and period employment of benefits maximum, expressed in % of minimum wage) record SD0VEk IR. 1991 3 years 12 months 12 months 65% first 6 months none None 60% following 6 months 70% during retraining 1992 6 months 65% first 6 months 45%1 150-180%f 60% following 6 months 70% during retraining 1995 3 years 12 months 12 months 60% first 3 months nonem 150% 50% following 9 months 1997 3 years 12 montas 12 monthDs 60% flirst 3 months nonenn 150% 50% foluowlng 9 months SgovenOl 1996 18 months 9-12 months 24 months 70% first 3 months 80%p 320%p 60% following 3 montlhs0 1990 10 months 9-12 months 24 months 70% flrst 3 msonths 100% 300% 60% ofolowing 3 months0 Plus an additional three months of unemployment assistance. b Unemployment insurance equal to the minimum wage plus 20% of the difference between the avcragc wagc and thc minimum wagc Since October 1990. d Average of last six months' wage; an additional 15% is awarded upon completion of a training course. e Not required if enrolled in a training course. 'The recipient receives 180% if enrolled in a training course. ' One year until January 1990 when it was extended to two years. h Unemployed earning from casual work not more than half of the minimum wage per month remain entitled to full Ut 'If previous earnings were lcss than minimum wagc then the benefit is set equal to previous camings. Introduced in September 1990. h Some of the unemployment benefit exhaustees (after 9 months) qualify for a SUPPORT ALLOWANCE PROGRAM for an additional (maximum) 18 months. This program is means tested, 0nd the level of allowance is 60% of the level of tnemployment benefit they initially receive 'Minimum applies only to first time unemployed and school leavers and is paid for a period of six months. Net monthly wage if lower than the minimum pension income. n First three months paid at 70%, then remaining months paid at 60% of average wage. 0 Recipients can receive a supplement for each family member to raise the average income per family member to 80% of the gross min wage. P As a percentage of the guaranteed minimum wage. Sources: Employment Observatory, no 8, OECD Short-term Economic Indicators. Sources and Definitions, national labour ministries; Rutkowski (1996), Micklewright and Nagy (1996), Terrell, Erbenova and Sorm, (1996), Vodopivec (1995), Lubvyova and Ours (1996), Kwiatkowski (1998). 18 Benefit eligibility requirements. To qualify for benefit receipt, the unemployed must register at the local employment office and have typically worked in covered employment for a minimum period of 9 to 12 months within a reference period of 12 to 36 months, depending on the country. As a significant exception, Hungary requires the minimum employment period of 90 days in the 4 years preceding unemployment. Workers who quit are either not eligible for benefit receipt (e.g., the Czech Republic) or observe a waiting period (e.g., Bulgaria, 5 months; Hungary, 180 days; and Poland, and Slovakia, 90 days). Continuing eligibility typically requires claimants to be actively seeking, capable of, and available for work. Employment offices also require the claimant be willing to accept suitable job offers. Furthermore, some countries (e.g., Poland) also require claimants to be willing to participate in vocational training or public works programs, and not collect any other public cash transfers at the same time. Non-compliance with these requirements, according to the legislation, results in disqualification. Because of the low level of benefits, transition countries often allow for a significant "eamings disregard," that is, allow unemployment benefit claimants to eam and continue to maintain benefit eligibility. In Bulgaria, for instance, an individual can continue to claim unreduced benefits so long as earnings do not exceed 150 percent the official minimum wage. Likewise, the maximum admissible level of earnings is 100 percent of the minimum wage in Hungary and 50 percent in both Poland and Romania. Scarpetta and Reutersward (1994) express concern that the widespread under-reporting of wages may mean that some claimants unfairly benefit at the expense of those claimants with the greatest need, resulting in the poor allocation of funds. This problem is certainly being compounded by increasing participation rates in the informal economy. Benefit levels. Benefits are usually a proportion of average earnings over some stipulated period of the most recent employment spell. The initial replacement rate ranges between 50 and 70 percent of average gross earnings, and is often degressive over time. For example, in Slovenia, the replacement rate is 70 percent in the first 3 months, followed by 60 percent in the remaining months. A notable exception is Poland where the benefit level is not related to previous earnings but rather set at 36 percent the national average wage; until 2003, a flat fee benefit set at a very low level (below 10 percent of the average wage) was 19 also in place in Estonia. Some countries such as Bulgaria and the Czech Republic reward those who attend or complete training courses by offering them a higher replacement rate. Benefit ceilings and floors are used to limit the range of benefits. When present, minimum benefit levels are usually at either official minimum wage or slightly below (75-90 percent of minimum wage), while the maximum benefit level is typically at 150 percent of the minimuim wage. However, the benefit ceiling exhibits greater variation, ranging from 140 percent in Bulgaria to 300 percent in Slovenia. The compression of the benefit range, particularly the establishment of low ceilings, have helped contain outlays. In addition, benefit ceilings play an important redistributive role. Under increased fiscal pressures, many countries have found it hard to sustain the benefit levels as set at the introduction of the programs. Several countries therefore reduced their unemployment insurance replacement rates by the late 1990s (see Figure 2). gFAgire 2: Repl nceemmt IRte of UJiemlloymnenmt H ramne IPaynme emets, 11 lnisfition Ecoimorlnes9 lEarny aXnt LgRe 3199f$* 0100 80 ~70< w z Early 60 ~~~~~~~~~~~~1990's 50 2 40 ELa 30 r~~~~~~~~~~ri ~~~~1990's 20; 10 i~- 0 ~ - Source: Table 8. *Average replacement rate in the first six months of benefit eligibility. For Estonia, the benefit is flat, so the rate is calculated as the level of the benefit divided by the average wage Beimeflt tiiniradtio. With the exceptions of Bulgaria (and since 2003, Estonia), all transition countries have reduced the maximnum potential duration of unemployment benefit payments (see Figure 3). As it stands, the maximum potential duration of benefits is typically 12 months. Notable exceptions include the Czech Republic (6 months) ox the 20 lower side and Slovenia (24 months) on the higher side. Durations are often tied to the length of previous employment history and sometimes to age (e.g., in Bulgaria, Estonia, Slovakia, and Slovenia). In Slovenia, for example, the potential benefit duration for those with work experience of 1-5 years is 3 months; in contrast, for those with work experience above 25 years and older than 55 years, it is 24 months. Figure 3: Maximum Potential Duration of Unemployment Insurance Payments, Transition Economies, Early and Late 1990s 30 25 EJ Eariy 20 1901S 15 Source: Table 8. Special rules. As in OECD countries, transition countries give special dispensation to school leavers, the older unemployed, and the unemployed in "high unemployment" regions. School leavers who are unable to find suitable jobs within a certain period of time (e.g., in Bulgaria, 1 month, and in Poland, 4 months) are entitled to unemployment benefits albeit for a shorter maximum potential duration than normal. These benefits are provided as a 'job search allowance" and tend to be some percentage of the official minimum wage (e.g., Hungary, 75 percent of minimum wage; Bulgaria, 90 percent). Options for the older unemployed are more varied, but typically consist of extended benefit durations (see above). Furthermore, in many countries, those individuals that are a couple of years away from retirement age are entitled to unemployment benefits until then. Similarly, those individuals unemployed in regions where the unemployment rate is particularly high or increasing more rapidly than the national rate are entitled to extended benefits. For example, in Poland, the 21 benefit duration is related to the relative incidence of unemployment in the local labor market and selectively to a number of other criteria.8 Fknamcihmg. Similar to most OECD countries, unemployment benefits are financed through extra-budgetary means based on regular contributions from employers (and some in cases, employees).9 These contributions are channeled into designated funds which are often used to finance both passive (including unemployment benefits) and active labor market programs. In Hungary, a separate fund (the Solidarity Fund) was set up for unemployment benefits exclusively. As aforementioned, all countries require employers to contribute for unemployment benefits, while only some require the same of employees (e.g., the Czech and Slovak Republics). The contribution rate for employers varies between 3 percent of payroll in Poland and Slovakia to 7 percent in Bulgaria and Hungary. Where applicable, the contribution rate for employees are generally lower, either 1 or 2 percent of wages. The rates for both employers and employees are comparable to that of OECD countries, but unlike the latter, unemployment benefit programs in transition countries often fail to be financially self- sufficient and consequently, program deficits have to be financed out of state budgets. For example, in 1992, Hungary and Poland financed 30 and 70 percent of total unemployment benefit costs respectively out of general tax revenues. JnemnpBoy]naent aMsssawtaee In some transition countries, unemployed workers who have exhausted their eligibility to unemployment insurance benefits can continue to receive benefits under a means-tested program called unemployment assistance. For example, in Slovenia, unemployment assistance benefits are awarded to individuals whose actual income per family member is below 80 percent of the minimum wage; regardless of the gap, the same amount of 80 percent of the guaranteed wage is paid to all who qualify. The benefit is paid for 15 months. 8 According to the 1996/97 act, if the local unemployment rate is less than the national average, then the resident unemployed is entitled to a maxinmm of 6 months; if the local rate exceeds the national average, the resident unemployed is entitled to a maximum of 12 months; and finally, if the local rate is more than double the national average, and the unemployed has more than 20 years of service, an unemployed spouse, and supports a child of less than 15 years of age, then s/he is entitled to a maximum duration of 18 months (Kwaitkowski, 1998). 8 One notable exception is Estonia where unemployment benefits is financed entirely from general tax revenues. 9 One notable exception is Estonia where unemployment benefits is financed entirely from general tax revenues. 22 Under unemployment assistance, the same rules for continuation of eligibility apply as under unemployment insurance. Unemployment assistance programs are also in place in Bulgaria and Hungary. Social assistance. Partly as a result of the declining coverage rate of unemployment benefit programs, transition countries have relied increasingly on social assistance - a means-tested income support scheme of the last resort - to provide income support to the unemployed. Largely as a reflection of its growing rolls, expenditures on social assistance, although less than 1 percent of GDP generally, has increased steadily in several countries. Social assistance typically takes the form of guaranteed minimum income schemes (e.g., the Czech and Slovak Republics, Bulgaria, and Romania) and are provided to all those in need, including the unemployed. Apart from these general schemes, some countries also offer means-tested assistance programs targeted at the long-term unemployed (e.g., Hungary). Social assistance benefits are generally flat rate (usually at guaranteed minimum income at uniform rates), and hence lower than earnings-related unemployment benefits. Minimum social assistance benefit levels as a percent of average wages in 1997-98 varied between a low of 10.6 percent in Romania and a high of 32.5 percent in the Slovak Republic. Benefits are provided indefinitely, subject to regular checks to determine continuing eligibility. When maximum durations are fixed as in Bulgaria and Slovenia (6 months), and Hungary (24 months), they tend to be renewable. If unemployed, social assistance claimants must also actively seek, be capable of, and available for work. Consequently, social assistance claimants have to report to employment offices periodically, although usually less frequently than unemployment benefit claimants. Boeri (1998) notes that the transition from earnings-related unemployment benefits to means-tested, flat-rate social assistance generally involves some compensation loss. However, since social assistance benefits take into account household characteristics (e.g., number of dependents), it is conceivable that low-income claimants with large families may receive higher payments than under unemployment benefits. In addition, the receipt of in- kind benefits may make social assistance more generous than unemployment benefits. 23 Notwithstanding differences in the nominal values in unemployment benefits and social assistance, high inflation in several countries have significantly reduced the generosity (in real terms) of income support for the unemployed as unemployment benefits (and social assistance to a lesser extent) have been poorly adjusted for inflation, or not at all. Consequently, in some countries, unemployment benefits have had to be "topped up" with social assistance in order to ensure that low-income claimants at least receive the social minima (e.g., the Czech and Slovak Republics). This concern aside, as prices stabilize, the transition from unemployment benefits to social assistance may result in unemployment traps as claimants (especially those with large families and no other eamers) may prefer social assistance benefits (often with in-kind provisions such as housing or free meals) to jobs, the majority of which tend to be low-wage (Boeri, 1997). Even if the adverse effects on work incentives can be addressed, the transition from unemployment benefits to social assistance necessarily involves an increased administrative burden as a result of the additional requirement of means-testing. 3.2 lmplem7aextad ss$$es The task of building administrative capacity to provide employment services to the unemployed - including unemployment benefits - has been quite challenging. With the rapid growth of benefit claimants, employment offices, often understaffed and underequipped, came under tremendous stress. Information systems on benefit delivery had to be introduced from scratch, and an integrated information system on delivery of all cash benefits still remains a remote goal in all transition economies. Furthermore, with employment offices handling both active and passive programs, the monitoring of claimant compliance suffered. The two roles - helping to find a job and monitoring eligibility - were often incompatible. Relatedly, in several countries unemployment benefit programs had to be amended as it became increasingly clear that they were too generous to be sustainable over the long-term - or that legislation did not provide an adequate basis for administering these benefits. There are numerous difficulties in making usual conditions for benefit eligibility "'operational" in any country. First, how should one monitor "availability for work?" A 24 recent attempt by Slovenia requires benefit recipients to make themselves available for contactirig by employment offices for three hours per day - but this arrangement, aimed at curbing informal employment, has not produced desired results (see below). Second, the requirement of "actively seeking employment" cannot be easily incorporated into legislation. What is reasonable to expect from the unemployed may well depend on individual circumstances (such as skills, qualifications, experience, and also the length of unemployment spell), as well as employment prospects in the local labor market. Third, additional problems are involved in defining a "suitable job", and with limiting the amount of work which may be undertaken without affecting the level or receipt of the benefit. The task of monitoring eligibility is even more difficult in transition economies. First, adjusting benefit legislation to suit local conditions and norms takes time (frequent changes in transition economies attest to that). Until legislative loopholes and deficiencies are fixed, fertile grounds exist for both "type I errors" (unjustified exclusions from benefits, for example, of workers whose employers did not pay benefit contributions), as well as "type I errors" (too easy access to benefits, for example, by persons who actually work). Second, transition economies offer lucrative employment opportunities in a thriving infornal economy, which raises the costs of monitoring. And third, many of the unemployed believe they are entitled to benefits - and this sentiment is sometimes shared with counselors at employment offices.'0 The task of monitoring eligibility is also hampered by the transitional economies' weak monitoring and enforcement capacity. These economies lack the technology, resources, and often also the political will to monitor and enforce existing laws." For example, Bardasi et al (2001) report that the proportion of benefit recipients who were actually searching for jobs is below 50 percent in Slovenia and ranges from 60 to 90 percent for the Czech Republic, Hungary, Poland and Slovakia. They report that these proportions are fall dramatically when one considers active job search only (defined as any method other than visiting the employment office): the proportion drops to about 25 percent for Slovakia, and exceeds 50 10 Vodopivec (1998) reports that benefit recipients in Slovenia were sometimes not invited to employment offices if it was known that they were not available for work (when they provided childcare, for example). 25 percent only for Poland and Hungary (men only).'2 Moreover, verifying the accuracy of self- reported eamings is often not done - in part because administrative information systems do not allow cross-checking of benefit receipt. It is thus not surprising that hlicklewright and Nagy (1996) report that in Hungary, disqualifications from unemployment insurance benefits occur rarely - for example, of the March 1992 cohort of unemployment insurance recipients, 4 percent of spells ended that way. The risk of disqualification was much higher for the young, the less-educated, blue-collar workers, and those living in the capital, Budapest. While conceivably these differences could occur with the same degree of enforcement of the rules, in all likelihood the severity with which the sanctions are imposed vary across offices within the country - as well as between countries. For example, the risk of benefit disqualification in Slovenia is much lower than in Hungary - in 1998, only one percent of spells ended with disqualification, and in 1999, only 0.65 percent, despite changes in legislation aimed at improving the monitoring of benefit eligibility. And in Estonia, the country with the most modest unenployment benefit, casual evidence suggest that employment offices sometimes side with the unemployed and let them collect benefits until exhaustion - precisely because the benefit is so low. The above considerations have an important bearing on the effects of income support systems, primarily on the decision to leave (formal) unemployment. For example, if monitoring of job search is lax, some of the unemployed may not make a genuine effort to search for jobs; or they may misuse the system by collecting benefits and performing undeclared paid work at the same time. We will return to these issues in the section on distributive and efficiency effects. 3.3 Coses f mnemploymaengpooge¢& Below we present the overall costs of unemployment benefit programs in transition economies, as well as summary measures of their generosity, using an accounting fiamework developed by Vroman (2002). The majority of transition economies kept their " Earle and Pauna (1998) report the average caseload of 668 unemployed per one unemployed in Romania in 1993, while only a in few OECD countries did the caseload exceeds 100. 12 For the sake of comparison, Bardasi et al (2001) also include the UK and Spain in their analysis of Central European countries. While the t)K has the highest proportion of benefit recipients who are actively searching, Spain's performance is simila to that of the Central European countries. 26 unemployment benefit expenditures below 1 percent of GDP, and Estonia and the Czech Republic even below 0.25 percent of GDP (Table 9, panel A), lower than in most OECD countries. The highest benefit expenditure shares were recorded by Poland and Hungary (in Hungary, this share rapidly decreased in the late 1990s). The two key determinants of the cost of unemployment protection are the income replacement rate (RRate) or benefit level, expressed as a fraction of average wage, and the ratio of benefit recipients and the number of unemployed as identified by labor force surveys (NBen/Unemp; note that the recipients are not necessarily a subset of the unemployed, and that this ratio can therefore exceed 1). The former factor reflects the relative value of benefits while the latter reflects the relative availability of benefits, both factors being outcomes of policy choices. Vroman (2002) posits that the product of the replacement rate and the share of compensated unemployed captures more inclusively the generosity of unemployment benefit programs. This product is termed the generosity index (G): Generosity index (G) = 1 00*RRate*(Nben/Unemp) Available evidence indicates that the replacement rate (RRate) has declined in all countries expect Slovenia (see Table 9, panel B). The story is more mixed for the share of unemployed who receive benefits (see Table 9, panel C) - Poland shows the most marked decline, from 79 percent in 1991 to 24 percent in 1999; other countries showed relatively slight or moderate deviations across time. Hungary represents an interesting example of relatively easy access to benefits, with the number of recipients even exceeding the number of unemployed in 1999. An examination of the generosity index of benefit programs clearly shows the effect of the sharp fall in the share of the recipients on the generosity of the Polish benefit program. Combining the two effects, the generosity index shows that the generosity of unemployment benefits is highest in Slovenia and Hungary, and lowest in Estonia, with the others clustered in the 5 to 15 percent range (see Table 9, panel D, and Figure 4). 27 Taible 9: Expelminfitunres nlmd Geiaerosflty of UepllRoymnea D3e1efKfts9 299@ 11991 11992 1 1993 1 1994 1 1995 1996 21997 1 199 T 19-99 A. Expenditures on unemyplo ment benefits (as a % of GIDP) Bulgaria 0.55 0.64 0.96 0.54 - _ _ Czech Republic 0.23 0.18 015 -- 0.13 0.14 0.21 0.24 _ Estonia -- -- - 0.11 0.07 0.07 0.08 - Hungary* 2.16 2.03 1.07 0.71 0.60 0.46 - Poland -- 1.71 1.72 1.77 1.88 1.77 -- - Slovak Republic 0.98 0.60 0.56 0.43 -- -- -- Slovenia* 0.57 0.82 1.22 1.13 0.75 0.71 0.90 0.89 0.79 B3. Unernlloyment benefit replacement rate Bulgaria 0.68 0.29 0.34 0.31 0.30 0.38 0.30 -- - Czech Republic 0.42 0.30 0.28 0.27 0.25 0.24 0.24 0.20 0.20 Estonia -- - 0.17 0.10 0.08 0.07 0.07 0.09 0.09 Hungary* -- - -- 0.25 0.23 0.21 0.20 0.20 0.22 Poland 0.34 0.38 0.36 0.37 0.37 0.33 0.32 0.30 0.24 Slovak Republic 0.49 0.32 0.30 0.27 0.25 0.23 0.30 - -- Slovenia* 0.32 0.29 0.33 0.34 0.33 0.31 0.36 0.37 0.37 C. Tlhe ratio of umem loyent benefit recipients and tIhe number of mom odl Bulgaria - 0.23 0.23 0.29 0.29 0.35 - - Czech Republic _ 0.33 0.37 0.34 0.37 0.45 0.45 0.43 Estonia - 0.30 0.20 0.24 0.26 0.25 0.31 Hungary* - 0.93 0.94 0.96 0.98 1.00 1.06 Poland 0.79 0.52 0.48 0.50 0.59 0.52 0.31 0.23 0.24 Slovak Republic - 0.40 0.37 0.26 0.32 0.33 - - Slovenia* 0.46 0.65 0.73 0.72 0.61 0.62 0.72 0.66 0.61 ID. Generosdty iindex Bulgaria - -- 7.7 7.1 8.8 10.9 10.6 - - Czech Republic - _ 9.5 9.9 8.6 8.9 10.8 9.0 8.5 Estonia - - - 3.2 1.5 1.9 1.8 1.8 2.8 Hungary* -- -- -- 23.5 22.1 20.4 19.9 20.4 22.7 Poland 27.0 19.8 17.4 18.6 21.6 17.3 9.8 6.9 5.6 SlovakRepublic - - 12.1 10.1 6.6 7.5 9.8 j - Slovenia* 14.6 18.9 24.2 24.2 20.5 20.1 26.6 24.6 22.8 Source: Employment Outlook, OECD, various issues, Czech Republic: Statistical Office and Ministry of Labor and Social Affairs; Estonia: National Labour Board; Hungary: Central Statstical Office and National Labour Centre; Slovenia: Yearly Work Report of Employment Office, various issues. * Payments/recipients under both unemployment insurance and unemployment assistance are included. The comparison of the generosity index of unemployment benefit programs in transition economies (averaged across time) with a sample of OECD countries shows that the programs in trasition economies are much less generous than in the majority of OECD 28 countries - they are comparable only to Portugal, Greece, and the US.13 The average generosity index across transition economies is 11.3 compared to 26.3 for the OECD sample. The most significant exceptions are Hungary and Slovenia, whose indices of generosity have been comparable to the average of the OECD sample. Figure 4: Generosity Index of Unemployment Benefit Systems, 1991-99 30 - 25 ~20- >' 15 - A 5 1991 1992 1993 1994 1995 1996 1997 1998 1999 - Bulgaria -_- Czech Republic Estonia I Hungazy )X Poland ISlovenia ---Slovak Republic Source: Table 9. 4. EVALUATION OF UNEMPLOYMENT BENEFIT SYSTEMS Above we discussed how the decline of output during the transition translated to reductions in employment and increases in unemployment, and described the formal unemployment benefit programs introduced to provide income support for the unemployed. This section seeks to evaluate the performance of these programs. Did they provide adequate and effective income support for the unemployed? Which groups of workers benefited the 13 The OECD sample comprises of Belgium, Denmark, France, Germany, Greece, Ireland, the Netherlands, Portugal, Spain, Sweden, the UK, and the US; data refer to 1992 (calculations of G for the OECD sample are from Vroman, 2002). 29 most and what were the effects of these programs on poverty? Was applying an OECD model - for example, benefit levels determined as a percentage of an individual's previous earnings as opposed to a flat benefit - a prudent choice? Moreover, what were the efficiency effects of these programs? Have they - as commonly found in developed economies - affected the work incentives of benefit recipients? Above all, have they reduced the intensity of job search and increased the reservation wage, thereby prolonging the duration of unemployment and contributing to higher unemployment? On the other hand, did the introduction of benefit programs ease the huge and painful task of enterprise restructuring? This section attempts to answer some of the above questions. We first discuss distributive and then efficiency effects. 401 Dismib'Vaye effecgs In the mid-1990s, in Hungary and Poland unemployment benefits represented a sizeable share of household income, they were received by the majority of households with unemployed workers, and they were largely directed to households in poverty or to those which were drawn out of poverty by such benefits. Although they were not designed to do so, unemployment benefits in these two countries strongly reduced poverty. In contrast, in some other countries in our sample, unemployment benefit programs had limited coverage, they provided a relatively small share of household income, and they did not reach many of the poor. In all countries, unemployment benefits in the mid 1990s were strongly progressive, thus bringing about the redistribution of income from the rich to the poor. Coverage. In the mid-1990s, the coverage of unemployment benefits varied tremendously among the countries in our sample (Table 10). In Hungary, 78 percent of households with at least one unemployed worker them received unemployment benefits, followed by 65 percent in Poland; in contrast, the coverage of households with an unemployed worker was only 17-19 percent in Estonia and Latvia. Understandably, from the standpoint of coverage of all households, the numbers are much lower, but variations are large as well. Interestingly, among households with an unemployed member, the receipt of unemployment benefits is less prevalent among poor households, with the exception of Estonia. 30 Table 10: Coverage, Targeting, Average Share, and Poverty Reduction of Unemployment Benefits' B_l_arla2 lEstonia Hun ary Latvia Poland Slovakia Slovenia "h. Hh. Hh. Hth. Hh. Hh. with with with with with with All hh. All hh. unemp. All hh.. unemp. All hh. unemp. All hh.. unemp. All hh.. unemp. All hh.. unemp. Coverage (in percent) The share of households that receive UB 4.22 1.94 18.74 16.20 77.53 2.14 17.42 9.31 64.73 7.83 24.35 8.08 42.69 The share of the poor households that receive UB3 10.57 5.08 20.76 32.49 64.45 5.22 13.01 15.60 51.56 17.75 32.10 13.84 34.64 Targeting - the share of UB budget received by the following groups: . Poor households 38.92 46.27 46.27 12.77 12.77 19.68 19.68 18.49 18.49 9.94 9.94 19.75 19.75 Households drawn from poverty by UB 12.16 11.60 11.60 38.24 38.24 33.24 33.24 34.35 34.35 25.59 25.59 25.14 25.14 Households above the poverty threshold even if they didn't receive UB 48.92 42.12 42.12 49.00 49.00 47.08 47.08 47.17 47.17 64.47 64.47 55.11 55.11 Average share of UB in total household income (in percent), among: . All households 0.55 0.43 4.11 4.17 19.94 0.78 6.31 3.21 22.34 0.67 2.07 2.13 11.28 Households receiving the benefit 13.04 21.92 21.92 25.72 25.72 36.21 36.21 34.51 34.51 8.51 8.51 26.43 26.43 Poverty reduction brought about by the UB receipt In percent of hypothetically poor (pre-. benefit headcount) 4 3.29 1.33 4.68 40.00 53.28 3.96 9.22 20.57 44.73 21.58 31.19 7.96 15.71 In percent of total population 0.60 0.25 1.84 5.23 19.72 0.59 3.79 3.53 19.21 1.05 2.97 1.31 5.84 Memorandum Items Poverty headcount 17.58 18.58 37.41 7.84 17.29 14.32 37.31 13.63 23.74 3.83 6.55 15.15 31.35 Percent of households that have an unemployed n.a. 10.37 100 20.90 100 12.31 100 14.38 100 32.14 100 18.92 100 Poverty Gap 5 3.36 3.43 0.94 1.95 1.89 0.17 2.56 Source: Own calculations from online HEIDE data (Household Expenditure and Income data for Transitional Economies), URL: hittp:llwww.worldbank.orelresearcliltransitionlhouse.hti); for Slovenia, data provided by the Slovenian Statistical Office. Survey year: Bulgaria, 1995; Hungary, 1993; Latvia, 1997; Poland, 1993; Slovak Republic, 1993; Slovenia, 1997-98. Sample Size: Bulgaria: 2,466; Hungary: 8,105; Latvia: 7,690; Poland: 16,051; Slovakia: 2,129; Slovenia: 2,577. Survey weights used where appropriate. Notes: Unemployment benefits include both payments of unemployment insurance and unemployment assistance. ' Bulgarian data do not allow the identification of unemployed household members. Poor are those households (and individuals living in them) whose income per equivalent adult is below the poverty line. Poverty line is defined as 60 percent of the median of the distribution of income per equivalent adult. The number of equivalent adults is obtained by using the OECD weighing scheme: I for the first adult in the household, 0.7 for each subsequent adult, and 0.5 for each child below 15 years. 4Hypothetically poor are households with pre-unemployment-bencil income per equivalent adult below the poverty line. Poverty gap is the share of income which, if distributed among the poor households, would bring them out of poverty, in total household income. 31 Average share of UB in household income. Countries also varied greatly by how sizeable was the average share of unemployment benefits in household income. Among all households, this share was the largest in Hungary (4.2 percent), followed by Poland (3.2); these two countries were also ahead of others when limiting households to those with at least one unemployed (in Hungary, the share within this set was 19.9 percent, and in Poland 22.3 percent). Comparable shares in other countries were smaller by several times, with the smallest being in Estonia (0.43 and 4.11 percent, in the sets of all households and households with at least one unemployed, respectively). Targeting. Another look at distributive properties of unemployment benefits is offered by viewing how benefits are spread over different groups. Among all countries, by far the largest share of unemployment benefits received by the poor was in Estonia (46 percent), but the share of individuals who were drawn out of poverty by unemployment benefits was the smallest. The small share of individuals drawn out of poverty in Estonia is quite likely due to the modest size of the benefit in the 1990s, as noted above. The highest share of individuals drawn out of poverty was recorded in Hungary, Poland, and Latvia. As mentioned above, in Hungary and Poland, the share of unemployment benefits in household income was highest among the countries included in our sample. The share of unemployment benefits received by households ranked above the poverty threshold by their pre-benefit income was highest in Slovakia (64 percent) and Slovenia (55 percent). Reduction of poverty. Although, admittedly, the objective of unemployment insurance programs is not poverty relief but rather consumption smoothing for the workers who lost their jobs, it is nonetheless interesting to observe to what extent unemployment benefits - by default, not by design - contributed to fighting poverty. In line with wide variations of the above-discussed distributive measures, the effects on poverty reduction also varied tremendously among countries. In Poland and Hungary, the poverty reduction effects were great. In Hungary, the reduction of poverty among the unemployed was over 50 percent (that is, in the absence of unemployment benefits, the number of poor unemployed would have more than doubled), and in Poland 45 percent. Even expressed as percentages of the total population the numbers are large: in Hungary, 4.1 percent of the total population was drawn out of poverty by unemployment benefits, and in Poland, 3.2 percent. Such favorable 32 results were produced by the attractive distributive properties discussed above - widespread coverage, the large share of benefits received by the poor, and the benefits representing a relatively large share of total household income, but are partly attributable also to a relatively small poverty gap in these two countries (0.94 and 1.95 percent in Hungary and Poland). The least reduction of poverty by benefits was recorded by Estonia, where 4.7 percent of the unemployed, and 0.25 percent of the total population, were drawn out of poverty by them. Small effects on poverty were recorded also by Bulgaria and Latvia, reflecting primarily the low coverage and small share of unemployment benefits in total income. Incidence of benefits by income quintiles. Further insights about the redistribution of income implied by unemployment benefits are obtained by the analysis of their incidence (Table 11). In all countries, in mid-1990s benefits were strongly progressive (poorer households received a larger share of benefits - Table 11, panel A). The most progressive system (with the bottom 40 percent of the households receiving 73 percent of unemployment benefits) was the Estonian one, which may be the consequence of the flat rate benefit regime. Interestingly, among the households with at least one unemployed, the incidence of benefits was regressive, suggesting that unemployment benefits were effective in moving recipient households up the post-benefit income distribution. The exceptions were Estonia and Slovakia, the countries with the smallest share of unemployment benefits in household incomes(see Table 10), where the small scale of transfers did not make a difference in the post-benefit distribution. The conclusion that unemployment benefits were effective in improving recipients' relative income positions is reinforced by the evidence that, within households with at least one unemployed, benefits ranked by pre-benefit income were progressive in all countries (Table 11, panel B). In sum, it is startling that although unemployment benefits are focused on job-losers, in transition countries they nonetheless contributed significantly to poverty reduction. Obviously, for many households, earnings represented the single most important household income, and unless it was compensated, the loss of a job pushed many of such households into poverty. The above evidence also suggests that it is more likely that unemployment benefits make a difference if coverage is large and the benefit amount is significant. While the fact that Estonia paid a flat benefit (that is, a benefit equal for all) contributed to the high 33 [iBnnI airla' E5s0DnIa IIunt any - LalvIa lPolanud Silovakia Silovenia LHh. Hh. Hh. Hh. Hh. Hh. with . with wilh..wth . with . with .__________________________________ j All hh. All hh. unemp. All hh.. unemp. All hh. unemp. All h __ it hh.. unemp. All b.. unemp. A. lHouseholnls ranked by post-heneflt income Highest quintile (the richest) 11.54 6.44 19.83 7.97 19.49 20.01 37.62 5.28 20.45 13.27 24.81 8.70 32.75 Second quintile 7.42 13.99 12.74 13.00 21.23 18.59 27.48 11.43 22.33 11.53 13.10 18.67 25.51 Third quintile 12.90 11.05 24.28 17.63 21.55 16.64 14.12 20.05 20.93 15.18 20.24 26.31 15.17 Fourth quintile 19.28 16.50 11.77 24.05 21.12 15.67 11.68 28.13 19.99 26.06 16.08 18.77 15.05 Lowest quintile (the poorest) 48.86 52.02 31.38 37.35 16.62 29.09 9.10 35.11 16.30 33.95 25.78 27.54 11.52 B. Houselnhlds rarnked by pre-beneft egneome Highest uintile (the richest) 6.29 5.81 16.73 2.92 12.82 5.11 13.55 2.60 | 14.34 7.46 15.98 3.85 18.62 Secondauintile 10.26 10.91 11.48 4.95 14.97 7.67 16.58 4.03 1 16.79 8.51 13.96 . 8.14 1 18.09 Third quintile 7.72 8.33 15.13 9.78 18.69 13.33 17.71 9.67 18.40 14.80 13.10 14.19 17.82 Fourth uintile 16.70 13.94 18.76 14.61 22.57 8.99 15.18 19.15 21.07 15.56 12.27 24.86 17.30 Lowest quintile (the poorest) 59.03 61.01 37.90 67.75 30.95 64.90 36.97 64.55 29.40 53.66 44.69 48.96 A8.17 Source: same as for Table 10. Notes: H-louseholds are ranked by income per equivalent adult (see Table 10 for the equivalency scale). Unemployment benefits include both payments of unemployment insurance and unemployment assis9ance. 2 Bulgarian data do not allow the identification of unemployed household members. 34 share of benefits received by the poor (the share of benefits received by the poor in an earnings-related scheme would be lower), the low amount of the benefit prevented a more significant effect on poverty reduction among recipients. Of course, the size of the poverty gap also determined the effectiveness of transfers in reducing poverty - for example, despite channeling quite a high share of benefits to the (pre-benefit) poor, Estonia's and Bulgaria's overall effect on reduction of poverty was modest not only because of the small level of the benefit, but also because their poverty was so deep. The evidence also shows that the income redistribution produced by unemployment benefits was strongly progressive, although it did not reach the extreme outcomes obtained, for example, in Chile, where about 60 percent of unemployment benefits is received by the poorest quintile of the population (Krumm et al, 1994). Because unemployment insurance contribution rates are earnings related, the incidence of net benefits, that is, the incidence of benefits once both the cost and benefits are considered, is even more progressive.14 The above evidence on distributive effects shows that unemployment benefits redistributed income from the rich to the poor and significantly reduced poverty. Have such effects been produced by worsening efficiency, for example, by producing work disincentives? This is the topic which we discuss next. 4.2 Efficiency effects Below we summarize the results of studies on how unemployment insurance schemes affected economic efficiency; we also present some new evidence on Estonia, the country with perhaps the most parsimonious unemployment benefit system of the countries studied here. Above all, we focus on incentive effects (on job-search effort and the duration of unemployment, on the restructuring of enterprises and on the overall reallocation of labor, and on the labor supply of other family members), and on aggregate effects. Effects on the duration of unemployment spells. Theoretical predictions about the effect of unemployment benefits on job-search effort are ambiguous. A stylized prediction from simple theoretical models is that an increase in unemployment benefit reduces the recipient's probability of transition from unemployment to employment, that is, it increases 14 The inpact of unemployment benefits is less progressive in developed countries. As shown by Forster (2000), the effects of benefits are progressive in about half of the OECD countries, and neutral in the other half. 35 the expected duration of unemployment. This follows from job-search models (because the reservation wage is assumed to rise initially with the benefit level), as well as from labor supply models (because less income is forgone by staying unemployed). However, once more complexity is introduced into the models (for example, recognizing that unemployment insurance is paid only for a finite period and that by taking employment, one re-qualifies for unemployment insurance) it can also be shown that an increase of the benefit rate makes the transition to employment more attractive, not less (see Atkinson and Micklewright, 1991). Or one can argue that unemployment benefits increases resources devoted to search and hence increases the probability of finding a job. In other words, the theoretical predictions on the effects of longer durations and higher replacement rates on the probability of transition from unemployment to employment are ambiguous. In contrast to the ambiguity of predictions arising from theoretical models, empirical evidence is much more clear-cut. Similar to the evidence on developed economies, empirical studies for transition economies show that unemployment benefits reduce the probability of leaving unemployment to take a job.15 Except for Romania, the negative effects of the potential benefit duration on the probability of exit from unemployment to employment have been confirned for all countries for which such studies were performed, although some studies also found little evidence of work disincentives (see Table 12 for a sunmmary of empirical findings). It is particularly interesting that the adverse incentive effects can be detected even in Estonia, a country with far the most parsimonious unemployment benefit program (see Figure 5).16 Most studies find that the exit rate from unemployment to employment significantly increases near benefit exhaustion (in some countries, the exit rate to inactivity also increases). The effects of the replacement ratio are less pronounced: Ham et al. (1999) find significant effects for the Czech Republic but not for Slovakia; Vodopivec (1995) also finds insignificant effects for Slovenia. As for the scale of these effects, Ham et is The vast majority of studies on developed economies find the elasticity of duration of unemployment with respect to replacement rate and particularly to duration is positive, and that there is a sharp increase in the probability of exit to enployment just before the benefit is exhausted (for a thorough review of the literature, see, for example, Devine and Keifer (1991), or Atkinson and Micklewright, 1991). 16 This phenomenon may be explained by the fact that employment offices - because of the low leel of benefits - are reluctant to take steps to disqualify benefit recipients. 36 al (1999) find the effects for the Czech Republic to be comparable to the ones in developed economies (few other studies provide estimates of the (elasticity of the duration of unemployment with respect to potential duration and level of benefits).17 Micklewright and Nagy (1996) estimate that about 8 percent of exits to jobs from unemployment occur at the point of exhaustion, Vodopivec (1995) provides an estimate of about 6 percent, and our calculations for Estonia produce an even higher figure, namely, 32 percent of such cases. Figure 5: Hazard of Exit from Unemployment to Employment, Estonia, 1991-1995 0.8 0.7 0.6 0.5 1.l ~0.4 ~0.3 0.2 0.1 0. 1 2 3 4 5 6 7 8 Spell length (months) Note: Based on 1998 observations (853 censored); data from the 1995 Estonian Labor Force Survey (see data description in Vodopivec 2002). The above results of analytical studies on the work disincentives created by unemployment benefits systems suggest that transition economies have certainly not escaped, them but that such disincentives are of similar in magnitude to those in developed economies. Moreover, the differences in the strength of these disincentive effects among countries is large, and both parsimonious and generous systems have generated such disincentives. Variations in the magnitude of these effects are produced by differences in the effectiveness of monitoring ofjob search and the enforcement of the work test among countries. 7 According to Layard et al (1991), the benefit elasticity ranges from 0.2 to 0.9, depending on the state of the labor market and the country concerned. According to Katz and Meyer (1990), the duration elasticity in the U.S. is in the range of 0.4 to 0.5. 37 Talmle 112: 1im vem Efffects of Ue lmploymo¢nte IBemeff½ts ½n T nsntnono Emomooes Sttudy I Mode/Metlnootlcu I Indhna BuEgar_a_____di Jones and Kotzeva (1998) Survivor fiunctions, binary logit Exit rate to employment increases between the 1 8U' and 260 month of the spell (social assistance benefit exhaustion). Study of the effects of the transition Survival functions indicate "waiting behavior" for social assistance recipients. to social assistance Cazes and Scarpetta (1998) Empirical hazard function, Exit probability toward the end of the entitlement period increased dramatically. piece-wise constant hazard Unemployment benefit recipients exit unemployment more slowly than non- function recipients, many leave to inactivity rather than to employment. Kotzeva, Mircheva, and A. Binomial logit Recipients of Ul are significantly less likely to take a job. Woergoetter (1996) Czech _Rerpubft_ Ham, Svejnar and Terrell (1998) Hazard model Elasticity of duration with respect to: - increase of replacement rate = 0.34 - increase in duration of benefit = 0.4 Est3nla Own calculation (see Figure 5) Empirical hazard function Exit to employment significantly increases around the point of benefit I __________________I________ exhaustion Iungmgry Micklewright and Nagy (1998) Non-parametric and parametric High proportion of Ul recipients remain until benefit exhaustion. proportional hazard, discrete Exit rates are characterized by a large spike in the period immediately afler time-duration model benefit exhaustion: job-exit hazard increases six- to eight-fold in comparison to the period prior to exhaustion. POIEnd Adamcehik (1999) Proportional hazard Negative effect of the receipt of benefits on probability of exit to a job, dramatic increase of the hazard as the benefit is about to expire. Puhani (1996) Hazard model Entitlement to unemployment benefits significantly prolongs duration of unemployment Steiner and Kwiatkowski (1995) Multinomial logit UR recipients had lower exit rates than non-UT recipients, particularly with (cited in Kwiatkowski, 1998) respect.to the exit rate from unemployment to inactivity. Boeri and Steiner (1996) Hazard model Exit rates increase as entitlement duration approaches exhaustion, especially in the capital (males: increased flow to employment; females: increased flow to inactivity). Exit rate to inactivity increased markedly in the month after benefit exhaustion. 38 Table 12: Incentives Effects of Unemployment Benefits in Transition Economies (cont'd.) Study | Model/Methodology Findings Poland _ _ _ _ _ _ _ _ _ _ _ _ _ _ Cazes and Scarpetta (1998) Empirical hazard function, Exit probability related to differentiated maximum lengths of UB entitlement. piece-wise constant hazard Unemployment benefit recipients exit unemployment much more slowly than function iion-recipients, but many leave to inactivity rather than to employment, .___ ___ ____ ___ ____ ___ ____ ___ ______ ____especially in backward areas. Romanla Earle and Pauna (1998) Hazard model Receipt of benefits increases probability of leaving unemployment (no ___________________________ _ | .disincentive effects). Slovakia Lubyova and Ours (1999) Proportional hazard Little evidence of disincentive effects. Hamn, Svejnar and Terrell (1998) Hazard model Elasticity of duration with respect to: - increase of replacement rate = 0.06 - increase in duration of benefit = 0.41 Slovenla Vodopivec (1995) Semi-parametric proportional Strong waiting effect - exit to employment significantly increases just before hazard benefit exhaustion 39 Enhancing restructuring of enterprises and labor reallocation. It may seem that, in order to speed up restructuring, the optimal level of insurace protection against unemployment is higher during the transition (for example, so as to overcome the reluctance of managers to lay-off workers). However, the theoretical grounds for increasing unemployment benefits to spur restructuring is weak. Blanchard (1997, pp. 113-4) shows that more generous benefits indeed add to the attractiveness of restructuring and to take unemployment risk, but at the same time hinder job creation. He concludes that "the case for increasing unemployment benefits on efficiency grounds is limited." The review of literature on transition economies did not reveal any rigorous investigations of this issue, although several studies asserted that the introduction of unemployment benefits helped the restructuring process. 18 From a political economy perspective, the connection between willingness to accept mass layoffs and the existence of a social safety net may seem quite plausible. Casual comparisons of policies and outcomes across countries, however, suggest that there are other, much more powerful forces at work which influence the pace of restucturing and labor reallocation. For example, in Slovenia, a country with one of the most generous unemployment benefit programs among the transition economies (see above), the pace of labor reallocation lagged much behind the pace of labor reallocation in Estonia, the country with the most meager benefits among the countries under consideration (see Figure 6 for the comparison of the intensity of restructuring among selected transition countries as measured by excess job reallocation rate). 18 Very interesting insights about the potential of unemployment benefits to stimulate restructuring are obtained by the simulation of the introduction of a modest unemployment benefit system in Russia (the system would provide one third of median cash eamings in the form unemployment assistance, see Klugman et al, 2002). By increasing the current wage floor, such an introduction would stimulate "high quality" mobility, as up to 14 percent of employed workers would become no worse off than unemployed. Of course, the effectiveness of unemployment benefit to stimulate mobility would depend on many other factors - job creation capacity of the economy certainly being one of the most important ones. 40 Figure 6: Intensity of Restructuring: Excess Job Reallocation Rate, Mid-1990s 16 14 12 10 - 8 - 6 -_ 4 4 Source: Faggio, G. and J. Konings (1999) Note that some other institutions used to keep unemployment low or to preempt inflows into unemployment stand directly in the way of labor reallocation: employment protection legislation (including severance pay provisions) is clearly on the top of this list (see the model of Blanchard, 1998, showing that larger job security leads to more "sclerotic" labor markets).'9 Note also that there are theoretical arguments for and empirical support of claims that tight job security provisions reduce labor force participation. For example, OECD (1999) finds a strong link between stricter EPL and (i) higher unemployment rates for younger workers, (ii) higher rates of self-employment, and (iii) lower employment rates for prime-age women, youths and older workers; Heckman and Pages (2000) also confirm the link between job security and lower employment and attribute 5 percentage points of employment reduction in Latin America to job security provisions. Effects on labor supply of other family members. Because unemployment assistance requires means-testing, one can expect that this will create disincentives for other family members to take a job. There is empirical evidence supporting such predictions for the Slovak Republic and for Poland. For example, Terrell et al (1996) report that the presence of an unemployed spouse lowered the hazard of exit to employment of social assistance l9 Vodopivec (2000) shows that worker and job flows in Estonia's transition have surpassed those in most other transition economies, thereby contnbuting to the efficient reallocation of labor, and attributes that to deliberate policies aimed at stimulating job creation and employment, chief among them being low employment protection. 41 recipients by 72 percent for females and by 82 percent for males. Similar effects are reported for Poland by Boeri (1997). Aggregate effects. The survey of the literature revealed only two studies explicitly addressing the link between the generosity of the unemployment benefit system and aggregate effects.20 Investigating the link between unemployment and labor market institutions in 8 transition economies in the late 1990s, Cazes (2002) finds that overall and long-term unemployment rates were not significantly affected by the replacement rate and duration of unemployment benefit programs, but that - somewhat counterintuitively - a longer benefit duration had a significant positive association with the youth unemployment rate. Similarly, Boeri (1996) reports that in the early 1990s, no significant changes in the escape rate to jobs were associated with the reductions of the generosity of unemployment benefits in Hungary, Poland, and the Slovak Republic, but he also finds that in the Czech Republic (where more job opportunities existed) the policy shift did increase the intensity of escape rates. It also remains to be seen what are the long-term consequences of the introduction of traditional unemployment insurance in an environment of relatively strong labor unions and centralized wage setting - in particular, whether such arrangements will produce labor market outcomes similar to those in the countries of continental European Community ('Eurosclerosis").21 §. CONCLUSHJ$OMS AND IFOLIHCY HILI[CAl'WHCN$ To help the swollen ranks of the unemployed, all transition economies introduced new labor market programs, both income support programs tailored according to traditional unemployment insurance programs, as well as active measures. Because transition economies were ill-prepared for the emergence of large-scale unemployment, there is litle 20 Riboud et al (2001) discuss the labor market outcomes of trasition economies in the framework of policies and institutions, but stop short of econometric testing of the relationship. 21 Theoretical models on unemployment persistence show that economies with UI systems experience larger and more prolonged unemployment following transient shocks. For example, Ljungqvist and Sargent (1997) develop a model to study the dynamics of two economies, one with an UI system and one without, when a transient economic shock is introduced. The non-UI economy recovers more rapidly as reservation wages adjust quicker and job search intensity is higher than in the UI economy. Nickell and Layard (1999) also show that unemployment benefits raise equilibriun unemployment. 42 doubt that the introduction of public cash benefit programs for the unemployed was warranted. Due to the unexpected and sudden occurrence of unemployment, individuals were unable to prepare for this event by self-insuring (for example, by saving) and self- protecting (for example, by choosing a more stable job); undeveloped financial markets made things worse. Moreover, the magnitude and the covariance of the shock rendered other private copying mechanisms (such as private transfers) inadequate. The above review of their performance shows quite a positive experience with unemployment benefit programs. The evidence shows that in all transition economies included in our sample, unemployment benefits helped to redistribute income from the rich to the poor. Moreover, although not designed for this task, in some economies unemployment benefits strongly reduced poverty. Specifically, their impact on poverty was strong in countries where programs had broad coverage and where benefits constituted a sizeable share of household income. On the other hand, there is also evidence - as it is for developed economies - that unemployment benefits created work disincentives. (It is possible that the availability of unemployment benefits, by increasing labor costs and wage pressures, contributed to higher unemployment, and that it generated more persistent unemployment - but no such imnpact has been confirmed in the literature.) How can these programs be improved? The above analysis suggests that - at least for those transition economies with weak administrative capacity - a flat benefit level (equal for all recipients) may be more appropriate than earnings-related benefit (equal to a certain percent of an individual's previous earnings). Flat benefits would not only foster redistribution of income from the rich to the poor and mitigate work disincentive effects, but also simplify the administration of benefits and reduce the scope for misuses. The above results on disincentives, together with the description of implementation problems faced by these programs, also suggest that transition economies should improve the administration of their benefit programs. This includes better monitoring of conditions for keeping benefits - including checking availability for work and concealed employment, enforcing job search (for example, by requiring proofs ofjob search) and providing adequate job-search assistance 43 through counseling, workshops, and job clubs. 22 To deter misuses of the program, the approach used by Westem countries could be followed by (1) requiring recipients to regularly declare their casual earnings; (2) taking away a portion of the benefit beyond a certain threshold, in proportion to declared earnings; and (3) performing checks on the accuracy of declared eamings through cross-referencing administrative databases (for example, tax returns and work history data bases). With the changing nature of unemployment - in the l990s, the share of long-term unemployed increased in all transition economies - the problem of the limited benefit duration of unemployment insurance programs has become more prominent. Instead of increasing the maximum duration of unemployment insurance eligibility, one way to address this problem is by offering unemployment assistance to claimants who have exhausted their eligibility for unemployment insurance benefits, a practice which already exists in several transition countries. Although unemployment assistance does not necessarily create less work disincentives than unemployment insurance (Vroman, 2002), better targeting under unemployment assistance can generate savings, and its a istration would not create excessive demands on increasingly sophisticated delivery systems of Central and Eastem European transition countries. 22 OECD (2000) reports several successful cases among OECD countries where more stringent enforcement of benefit eligibility criteria and stricter imposition of sanctions contributed to the reduction of unemployment 44 REFERENCES Adamchik, V. (1999). "The Effect of Unemployment Benefits on the Probability of Re- employment in Poland," Oxford Bulletin of Economics and Statistics 61, 95-108. Atkinson, A.B. and J. Micklewright (1991). "Unemployment Compensation and Labor Market Transitions: A Critical Review." Journal of Economic Literature 29: 1679- 1727. Bardasi, E., A. Lasaosa, J. Micklewright, and G. Nagy (2001). "Measuring the Generosity of Unemployment Benefit Systems: Evidence from Hungary and elsewhere in Central Europe," Acta Oeconomica. Blanchard, 0. (1998). "Employment Protection and Unemployment," MIT, Paper available at: http://web.mit.edu/blanchar/www/. 16pp. Blanchard, 0. (1997). The Economics of Post-Communist Transition. Clarendon Lectures in Economics. Oxford: Oxford University Press. Blanchard, O., and P. Portugal (1998). "What Hides Behind An Unemployment Rate: Comparing Portuguese and U.S. Unemployment." MIT, processed. 39pp. Bilsen, V., and J. Konings (1998). "Job Creation, Job Destruction Growth of Newly Established, Privatized, and State-Owned Enterprises in Transition Economies: Survey Evidence from Bulgaria, Hungary, and Romania," Journal of Comparative Economics 26, 429-445. Boeri, T. (1998). "Labor market flows in the midst of structural change," in S. Commander, ed., Enterprise Restructuring and Unemployment Models of Transition," EDI Development Studies, Washington, D.C.: World Bank. Boeri, T. (1997). "Labor Market Reforms in Transition Economies," Oxford Review of Economic Policy 13(2),126-135. Boeri, T. (1996). Unemployment Outflows and the Scope of Labour Market Policies in Central and Eastem Europe, in OECD, "Lessons from Labour Market Policies in the Transition Countries," OECD Proceedings, Paris: OECD. Boeri, T. and V. Steiner (1996). "Wait Unemployment in Economies in Transition: the Case of Poland," Zentrum for Europaishe Wirtschaftsforschung, processed, 28pp. Bojnec, S., and J. Konings (1998). "Job Creation, Job Destruction and Labour Demand in Slovenia." Leuven Institute for Central and East European Studies, WP No. 74. 45 Caballero, R., and M. Hammour (2000). "Institutions, Restructuring, and Macroeconomic Performance." MIT, processed, 39pp. Cazes, S. (2002). "Do Labor Market Institutions Matter in Transition Economies? An Analysis of Labour Market Flexibility in the Late Nineties," ILO, International Institute for Labour Studies, Discussion Paper 140/2002. Cazes, S., and S. Scarpetta (1998). "Labour Market Transitions and Unemployment Duration: Evidence from Bulgarian and Polish Micro-data," Economics of Transition 6, 113-144. Davis, S., J., Haltiwanger, and S. Schuh (1996). Job Creation and Job Destruction. Cambridge: MIT Press. Devine, T. J. and N. M. Kiefer (1991). Empirical Labor Economics: The Search Approach. Oxford: Oxford University Press. Earle, J. and C. Pauna (1998). "Long-Term Unemployment, Social Assistance and Labor Market Policies in Romania," Empirical Economics 23: 203-35. Faggio, G. and J. Konings (1999). "Gross Job Flows and Firm Growth in Transition Countries: Evidence using Firm Level Data on Five Countries." CEPR Discussion Paper No. 2261. Ham, J.C., J. Svejnar and K. Terrell (1998). "Unemployment and the Social Safety net During Transitions to a Market Economy: Evidence from the Czech and Slovak Republics," American Economic Review, 8 8(5), 1117-42. Haltiwanger, J., and M. Vodopivec (2002). "Gross Worker and Job Flows in a Transition Economy: An Analysis of Estonia." Labour Economics. 9(5), pp. 601-30. Heckman, J. J., and C. Pages (2000). "The Cost of Job Security Regulation: Evidence from Latin American Labor Markets." NBER Working Paper No. 7773. Holzmann, R., J. Gacs, and W. Georg (1995). Output Decline in Eastern Europe: Unavoidable, External Influence or Homemade? International Studies in Economics and Econometrics, Vol.34. Dordrecht, Boston and London: Kluwer Academic. Jones, D.C. and M. Kotzeva (1998). "Work Incentives and Other Effects of the Transition to Social Assistance in the Transition Economies: Evidence from Bulgaria," in OECD, Lessons from Labour Market Policies in the Transition Countries," OECD Proceedings, Paris: OECD. 46 Klugman, J, J. Micklewright and G. Rewdmond (2002). "Poverty in the Transition: Social Expenditures and the Working-Age Poor." EBRD, processed Konings, J., H. Lehmann, and M.E. Schaffer (1996). "Job Creation and Destruction in a Transition Economy: Ownership, Firm Size, and Gross Job Flows in Polish Manufacturing," Labour Economics, 3(3), 299-317. Kotzeva, M., D. Mircheva, and A. W6rg6tter (1996). "Evaluation of Active and Passive Labour Market Policy in Bulgaria,"in OECD, Lessons from Labour Market Policies in the Transition Countries, OECD Proceedings, Paris: OECD. Krurnm, K., B. Milanovic, and M. Walton (1994). "Transfers and the Transition from Socialism," The World Bank Policy Research Working Paper No. 1380. Kwiatkowski, E., (1998). "Unemployment Benefits and Labor Market Developments in Poland in Transition," University of Lodz, mimeo. Layard R. and Richter (1994). "Labour Market Adjustment in Russia." Stockholm Institute for East European Economics, Working Paper No. 98. Lubyova M. and J.C. van Ours (1999). "Unemployment Duration Of Job Losers In A Labour Market In Transition," Economics of Transition 7(3), 665-86. Mencinger, J. (2000). "Generating Employment in Economies in Transition: The Case of Slovenia," in B. Marin, D. Meulders, and D. J. Snower (Eds). Innovative Employment Initiatives, Vienna: Ashgate. Micklewright, J. and G. Nagy (1998). "The Implications of Exhausting Unemployment Entitlement in Hungary," Budapest University of Economics Working Paper on the Labour Market, No. 1998/2. Micklewright, J. and G. Nagy (1996). "Evaluating Labor Market Policy in Hungary", in OECD, Lessons from Labour Market Policies in the Transition Countries, OECD Proceedings, Paris: OECD. Ljungqvist, L. and T. Sargent (1997). "Taxes and Subsidies in Swedish Unemployment," in R. Freeman, R. Topel and B. Swedenborg, eds., The Welfare State in Transition: Reforming Swedish Model, SNS Forlag, Stockholm. Nickell S. and R. Layard (1999). "Labor Market Institutions and Economic Performance," in Ashenfelter and Card, eds., Handbook of Labor Economics, Vol. 3. North Holland. OECD (2000). Employment Outlook, Paris: OECD. 47 OECD (1999). Employment Outlook, Paris: OECD. Grazem P.and M. Vodopivec (1995). "Winners and Losers in Transition: Retbus to Education, Experience, and Gender." World Bank Economic Review, 9(2), pp. 210- 230. Puhani, P. A (1996). "TPoland on the Dole: Unemployment Benefits, Training and Long-Term Unemployment during the Transition," Zentrum for Europaishe Wirtschaftsforschung Discussion Paper No.96-30. Riboud, M. Silva-Jauregui, C. and Sanchez-Paramo, C. (2001). "Does Eurosclerosis Matter? Institutional Reform and Labor Market Performance in Central and Eastern European Countries in the 1990s," in B. Funck and L. Pizzati, eds., Labor, Employment, and Social Policies in the EU Enlargement Process: Changing Perspectives and Policy Options, The World Bank. Scarpetta S., and A. Reutersward (1994). "Unemployment Benefit System and Active Labour Market Policies in Central and Eastem Europe: An Overview." In Unemployment in Transition Countries: Transient or Persistent, Paris: OECD. Terrell, K. and D. Munich (1996). "Labour Market Policies in the Czech Republic," in OECD, Lessons from Labour Market Policies in the Transition Countries, OECD Proceedings, Paris: OECD. Terrell, K., M. Lubyova, and M. Strapec, (1996). "Evidence on the Implementation and Effectiveness of Passive and Active Labour Market Policies in the Slovak Republic," in OECD, Lessons from Labour Market Policies in the Transition Counties, OECD Proceedings, Paris: OECD. Vodopivec, M. (2002). "Worker Reallocation During Estonian Transition To Market," International Journal of Manpower 23(1), 77-97. Vodopivec, M. (1995). "Unemnployment Insurance and Duration of Unemployment," World Bank Policy Research Worling Paper No.1552. Vroman, W. (2002). "Unemployment Insurance and Unemployment Assistance: A Comparison." World Bank Social Protection Discussion Paper No. 0203. 48 Social Protection Discussion Paper Series Titles No. Title 0309 Gender and Social Funds: Challenges and Opportunities By Kathleen Kuehnast 0308 Waivers and Exemptions for Health Services in Developing Countries Ricardo Bitran and Ursula Giedion 0307 Ageing and Pensions in the Euro Area: Survey and Projection Results by P. C. Rother, M. Catenaro and G. Schwab 0306 Safety Nets in Transition Economies: A Primer by Louise Fox 0305 Child Work: An Expository Framework of Altruistic and Non-Altruistic Models by Furio Camillo Rosati and Zafiris Tzannatos 0304 Pension Reform in Croatia by Zoran Anusic, Philip O'Keefe, and Sanja Madzarevic-Sujster 0303 Joblessness and Precarious Work in Bulgaria: Addressing the Multiple Aspects of Vulnerability in the Labour Market by Alexandre Kolev 0302 Systemic Shocks and Social Protection: Role and Effectiveness of Public Works Programs by Kalanidhi Subbarao 0301 What Role for Safety Net Transfers in Very Low Income Countries? by W. James Smith and Kalanidhi Subbarao 0233 Public Attitudes Matter: A Conceptual Frame for Accounting for Political Economy in Safety Nets and Social Assistance Policies by Carol Graham 0232 Ensuring Access to Essential Services: Demand-Side Housing Subsidies by Harold M. Katsura and Clare T. Romanik 0231 Gender and Risk in the Design of Social Protection Interventions by Kene Ezemenari, Nazmul Chaudhury, and Janet Owens 0230 Vulnerability and Poverty Measurement Issues for Public Policy by Jean-Yves Duclos Social Protecnonm lDfiscussionm Paper Series T1Iles coimdinued No. Title 0229 Emerging from Ethnic Conflict: Challenges for Social Protection Design in Transition Countries by Christian Bodewig 0228 Child Work in Zambia: A Comparative Study of Survey Instruments by Niels-Hugo Blunch, Amit Dar, Lorenzo Guarcello, Scott Lyon, Amy Ritualo and Furio Rosati 0227 Strengthening Public Safety Nets from the Bottom Up by Jonathan Morduch and Manohar Sharma 0226 Incentives and the Role of Institutions in Provision of Social Safety Nets by Chris de Neubourg 0225 Food-Based Safety Nets and Related Programs by Beatrice Lorge Rogers, Ph.D. and Jennifer Coates, M.S. 0224 Subsidies as a Social Safety Net: Effectiveness and Challenges by Harold Alderman 0223 Assisting the Poor with Cash: Design and Implementation of Social Transfer Programs by Steven R. Tabor 0222 Effectiveness of Lending for Vocational Education and Training: Lessons from World Bank Experience by S. Canagarajah, A. Dar, R. Nording and D. Raju 0221 Participation of Children in Schooling and Labor Activities: A Review of Empirical Studies by Amit Dar, Niels-Hugo Blunch, Bona Kim and Masaru Sasaki 0220 The World Bank and Children: A Review of Activities by Iqbal Kaur and Zafiris Tzanratos 0219 Managing Public Pension Reserves Part II: Lessons from Five Recent OECD Initiatives by Robert Palacios 0218 Guidelines for Assessing the Sources of Risk and Vulnerability by Karin Heitzmann, R. Sudharshan Canagarajah and Paul B. Siegel Social Protection Discussion Paper Series Titles continued No. Title 0217 Czech Pension System: Challenges and Reform Options by Esperanza Lasagabaster, Roberto Rocha and Patrick Wiese 0216 Extending Social Protection to Informal Workers in the Horticulture Global Value Chain by Arrnando Barrientos and Stephanie Ware Barrientos 0215 Social Fund Support of Microfmance: A Review of Implementation Experience by Alexandra Gross and Samantha de Silva 0214 Income Support Systems for the Unemployed: Issues and Options by Milan Vodopivec and Dhushyanth Raju 0213 Social Protection @ Your Fingertips. Using Information & Communications Technologies in Social Protection by Knut Leipold 0212 Short-and Long-term Impacts of Economic Policies on Child Labor and Schooling in Ghana by Niels-Hugo Blunch, Sudharshan Canagarajah and Sangeeta Goyal 0211 Supporting and Expanding Cormmunity-Based HIV/AIDS Prevention and Care Responses: A Report on Save the Children (US) Malawi COPE Project by Susan S. Hunter 0210 World Vision's Experience Working with HIV/AIDS Orphans in Uganda - 1990-1995 by Joe Muwonge 0209 The Reformed Pension Systems in Latin America by Jose E. Devesa-Carpio and Carlos Vidal-Melia 0208 Mandatory Annuity Design in Developing Economies by Suzanne Doyle and John Piggott 0207 Long-Term Welfare and Investment Impact of AIDS-Related Changes in Family Composition: Evidence from Uganda by Klaus Deininger, Anja Crommelynck and Gloria Kempaka Social FProtecidonm Discunssion mPaper Series T'iMies conid mued No. Title 0206 Child Labor Handbook by Alessandro Cigno, Furio C. Rosati and Zafiris Tzannatos 0205 An Overview of Labor Markets World-Wide: Key Trends and Major Policy Issues by Gordon Betchermnan 0204 Options of Public Income Support for the Unemployed in the Philippines and Social Protection by Jude H. Esguerra, Makoto Ogawa, and Milan Vodopivec 0203 Unemployment Insurance and Unemployment Assistance: A Comparison by Wayne Vroman 0202 Does Eurosclerosis Matter? Institutional Reform and Labor Market Perforrnance in Central and Eastem European Countries in the 1990s. by Michelle Riboud, Carolina Sanchez-Paramo and Carlos Silva-Jauregui 0201 Pension Reform and Capital Markets: Are There Any (Hard) Links? by Eduardo Walker and Fernando Lefort 0131 Child Labor, Nutrition and Education in Rural India: An Economic Analysis of Parental Choice and Policy Options by Alessandro Cigno, Furio Camillo Rosati and Zafiris Tzannatos 0130 Social Protection and the Infornal Sector in Developing Countries: Challenges and Opportunities by Sudharshan Canagarajah and S.V. Sethuraman 0129 Chile's Pension Reform After 20 Years by Rodrigo Acufna R. and Augusto Iglesias P. 0128 Labor Market Regulation: International Experience in Promoting Employment and Social Protection by Gordon Betcherman, Amy Luinstra, and Makoto Ogawa 0127 Generational Accounting and Hungarian Pension Reform by R6bert I. Gal, Andras Simonovits and Geza Tarcali 0126 Orphans and Other Vulnerable Children: What role for social protection? edited by Anthony Levine Social Protection Discussion Paper Series Titles continued No. Title 0125 Child Farm Labour: The Wealth Paradox by Sonia Bhalotra and Christopher Heady 0124 What Can Be Done about Child Labor? An Overview of Recent Research and Its Implications for Designing Progrars to Reduce Child Labor by Bj0rne Grimsrud 0123 Measuring and Analyzing Child Labor: Methodological Issues by Bj0rne Grimsrud 0122 Farnily-Controlled Child Labor in Sub-Saharan Africa-A Survey of Research by Jens Christopher Andvig 0121 Is Child Work Necessary? by Sonia Bhalotra 0120 The Cost and Benefits of Collective Bargaining: A Survey by Toke Aidt and Zafiris Tzannatos 0119 The Informal Sector Revisited: A Synthesis Across Space and Time by Niels-Hugo Blunch, Sudharshan Canagarajah and Dhushyanth Raju 0118 Social Services Delivery through Community-Based Projects by Dinah McLeod and Maurizia Tovo (available in Spanish) 0117 Earnings Inequality in Transition Economies of Central Europe Trends and Patterns During the 1990s by Jan J. Rutkowski 0116 Viewing Microinsurance as a Social Risk Management Instrument by Paul B. Siegel, Jeffrey Alwang and Sudharshan Canagarajah 0115 Vulnerability: A View from Different Disciplines by Jeffrey Alwang, Paul B. Siegel and Steen L. Jorgensen 0114 Individual Accounts as Social Insurance: A World Bank Perspective by Robert Holzmann and Robert Palacios 0113 Regulating Private Pension Funds' Structure, Performance and Investments: Cross-country Evidence by P.S. Srinivas, Edward Whitehouse and Juan Yermo Social Protectkomi DAscussion Palpeir neries Thiles contniuned No. Title 0112 The World Bank and the Provision of Assistance to Redundant Workers: Experience with Enterprise Restructuring and Future Directions by Yi Chen 0111 Labor Markets in Transition Economies: Recent Developments and Future Challenges by Mansoora Rashid and Jan Rutkowski 0110 Operating Instruction Included: A Review of Social Investnent Fund Operations Manuals by Juliana Weissman 0109 Risk and Vulnerability: The Forward Looking Role of Social Protection in a Globalizing World by Robert Holzmann 0108 Australia's Mandatory Retirement Saving Policy: A View from the New Millennium by Hazel Bateman and John Piggott 0107 Annuity Markets and Benefit Design in Multipillar Pension Schemes: Experience and Lessons from Four Latin American Countries by Robert Palacios and Rafael Rofinan 0106 Guide for Task Teams on Procurement Procedures Used in Social Funds by Jorge A. Cavero Uriona 0105 Programmes Actifs Pour Le Marche Du Travail: Un Aper9u General Des Evidences Resultant Des Evaluations by Zafiris Tzannatos and Amit Dar 0104 Kazakhstan: An Ambitious Pension Reform by Emily S. Andrews 0103 Long-term Consequences of an Innovative Redundancy-retraining Project: The Austrian Steel Foundation by Rudolf Winter-Ebmer 0102 Community Based Targeting Mechanisms for Social Safety Nets by Jonathan Conning and Michael Kevane (available in Spanish) Social Protection Discussion Paper Series Titles continued No. Title 0101 Disability and Work in Poland by Tom Hoopengardner 0024 Do Market Wages Influence Child Labor and Child Schooling? by Jackline Wahba 0023 Including the Most Vulnerable: Social Funds and People with Disabilities by Pamela Dudzik and Dinah McLeod 0022 Promoting Good Local Governance through Social Funds and Decentralization by Andrew Parker and Rodrigo Serrano 0021 Creating Partnerships with Working Children and Youth by Per Miljeteig 0020 Contractual Savings or Stock Market Development. Which Leads? by Mario Catalan, Gregorio Impavido and Alberto R. Musalem 0019 Pension Reform and Public Information in Poland by Agnieszka Chlon 0018 Worker Reallocation During Estonia's Transition to Market: How Efficient and How Equitable? by Milan Vodopivec 0017 How Poor are the Old? A Survey of Evidence from 44 Countries by Edward Whitehouse 0016 Administrative Charges for Funded Pensions: An Intemational Comparison and Assessment by Edward Whitehouse 0015 The Pension System in Argentina: Six years after the Reform by Rafael Rofinan 0014 Pension Systems in East Asia and the Pacific: Challenges and Opportunities by Robert Holzmann, Ian W. Mac Arthur and Yvonne Sin 0013 Survey of Disability Projects. The Experience of SHIA, Swedish International Aid for Solidarity and Humanity by Kaj Nordquist Sociall Protection Discussion Paper Series TnLdes cointmued No. Title 0012 The Swedish Pension Reform Model: Framework and Issues by Edward Palmer 0011 Ratcheting Labor Standards: Regulation for continuous Improvement in the Global Workplace by Charles Sabel, Dara O'Rourke and Archon Fung 0010 Can Investments in Emerging Markets Help to Solve the Aging problem? by Robert Holzmann 0009 International Pattems of Pension Provision by Robert Palacios and Montserrat Pallares-Miralles 0008 Regulation of Withdrawals in Individual Account Systems by Jan Walliser 0007 Disability Issues, Trends and Recommendations for the World Bank by Robert L. Metts 0006 Social Risk Management: A New Conceptual Framework for Social Protection and Beyond by Robert Holzmann and Steen J0rgensen 0005 Active Labor Market Programs: Policy Issues for East Asia by Gordon Betcherman, Amit Dar, Amy Luinstra, and Makoto Ogawa 0004 Pension Reform, Financial Literacy and Public Information: A Case Study of the United Kingdom by Edward Whitehouse 0003 Managing Public Pension Reserves Part I: Evidence from the International Experience by Augusto Iglesias and Robert J. Palacios 0002 Extending Coverage in Multi-Pillar Pension Systems: Constraints and Hypotheses, Preliminary Evidence and Future Research Agenda by Robert Holzmann, Truman Packard and Jose Cuesta 0001 Contribution pour une Strategie de Protection Sociale au Benin by Maurizia Tovo and Regina Bendokat * The papers below (No. 9801-9818 and 9901-9934) are no longer being printed, but are available for download from our website at www.worldbank.org/sp Social Protection Discussion Paper Series Titles continued No. Title 9934 Helping the Poor Manage Risk Better: The Role of Social Funds by Steen J0rgensen and Julie Van Domelen 9933 Coordinating Poverty Alleviation Programs with Regional and Local Governments: The Experience of the Chilean Social Fund - FOSIS by Jorge C. Barrientos 9932 Poverty and Disability: A Survey of the Literature by Ann Elwan 9931 Uncertainty About Children's Survival and Fertility: A Test Using Indian Microdata by Vincenzo Atella and Furio Camillo Rosati 9930 Beneficiary Assessment Manual for Social Funds by Lawrence F. Salmen 9929 Improving the Regulation and Supervision of Pension Funds: Are There Lessons from the Banking Sector? by Roberto Rocha, Richard Hinz, and Joaquin Gutierrez 9928 Notional Accounts as a Pension Reform Strategy: An Evaluation By Richard Disney 9927 Reform Options for Pay-As-You-Go Public Pension Systems by Sheetal K. Chand and Albert Jaeger 9926 An Asset-Based Approach to Social Risk Management: A Conceptual Framework by Paul Siegel and Jeffrey Alwang 9925 Migration from the Russian North During the Transition Period by Timothy Heleniak 9924 Pension Plans and Retirement Incentives by Richard Disney and Edward Whitehouse 9923 Shaping Pension Reform in Poland: Security Through Diversity by Agnieszka Chlon, Marek G6ra and Michal Rutkowski Social Protecftion Discussion lIaper Sernes T1Illes coilimuned No. Title 9922 Latvian Pension Reform by Louise Fox and Edward Palmer 9921 OECD Public Pension Programmes in Crisis: An Evaluation of the Reform Options by Richard Disney 9920 A Social Protection Strategy for Togo by Regina Bendokat and Maurizia Tovo 9919 The Pension System in Singapore by Mukul G. Asher 9918 Labor Markets and Poverty in Bulgaria by Jan J. Rutkowski 9917 Taking Stock of Pension Reformns Around the World by Anita M. Schwarz and Asli Demirguc-Kunt 9916 Child Labor and Schooling in Africa: A Comparative Study by Sudharshan Canagarajah and Helena Skyt Nielsen 9915 Evaluating the Impact of Active Labor Programs: Results of Cross Country Studies in Europe and Central Asia by David H. Fretwell, Jacob Benus, and Christopher J. O'Leary 9914 Safety Nets in Transition Economies: Toward a Refonn Strategy by Emily S. Andrews and Dena Ringold 9913 Public Service Employment: A Review of Programs in Selected OECD Countries and Transition Economies by Sandra Wilson and David Fretwell 9912 The Role of NPOs in Policies to Combat Social Exclusion by Christoph Badelt 9911 Unemployment and Unemployment Protection in Three Groups of Countries by Wayne Vroman 9910 The Tax Treatment of Funded Pensions by Edward Whitehouse Social Protection Discussion Paper Series Titles continued No. Title 9909 Russia's Social Protection Malaise: Key Reform Priorities as a Response to the Present Crisis edited by Michal Rutkowski 9908 Causalities Between Social Capital and Social Funds by Jesper Kammersgaard 9907 Collecting and Transferring Pension Contributions by Rafael Rofman and Gustavo Demarco 9906 Optimal Unemployment Insurance: A Guide to the Literature by Edi Karni 9905 The Effects of Legislative Change on Female Labour Supply: Marriage and Divorce, Child and Spousal Support, Property Division and Pension Splitting by Antony Dnes 9904 Social Protection as Social Risk Management: Conceptual Underpinnings for the Social Protection Sector Strategy Paper by Robert Holzmann and Steen Jorgensen (available in Russian) 9903 A Bundle of Joy or an Expensive Luxury: A Comparative Analysis of the Economic Environment for Family Formation in Westem Europe by Pierella Paci 9902 World Bank Lending for Labor Markets: 1991 to 1998 by Amit Dar and Zafiris Tzannatos 9901 Active Labor Market Programs: A Review of the Evidence from Evaluations by Amit Dar and Zafiris Tzannatos 9818 Child Labor and School Enrollment in Thailand in the 1 990s By Zafiris Tzannatos 9817 Supervising Mandatory Funded Pension Systems: Issues and Challenges by Gustavo Demarco and Rafael Rofman 9816 Getting an Earful: A Review of Beneficiary Assessments of Social Funds by Daniel Owen and Julie Van Domelen 994-5 This paper has been revised, see Discussion Paper No. 9923 Soeal )Pirotecticni Discunssiomn Papeir ernies TItRes coimtnned No. Title 9814 Family Allowances by Suzanne Roddis and Zafiris Tzannatos 9813 Unemployment Benefits by Zafiris Tzannatos and Suzanne Roddis 9812 The Role of Choice in the Transition to a Funded Pension System by Robert Palacios and Edward Whitehouse 9811 An Alternative Technical Education System: A Case Study of Mexico by Kye Woo Lee 9810 Pension Refonn in Britain by Edward Whitehouse 9809 Financing the Transition to Multipillar by Robert Holzmann 9808 Women and Labor Market Changes in the Global Economy: Growth Helps, Inequalities Hurt and Public Policy Matters by Zafiris Tzannatos 9807 The World Bank Approach to Pension Reform by Robert Holzmann 9806 Govermnent Guarantees on Pension Fund Retums by George Pennacchi 9805 The Hungarian Pension System in Transition by Robert Palacios and Roberto Rocha 9804 Risks in Pensions and Annuities: Efficient Designs by Salvador Valdes-Prieto 9803 Building an Environment for Pension Reform in Developing Countries by Olivia S. Mitchell 9802 Export Processing Zones: A Review in Need of Update by Takayoshi Kusago and Zafiris Tzannatos 9801 World Bank Lending for Labor Markets: 1991 to 1996 by Amit Dar and Zafiris Tzannatos Summary Findangs The paper reviews unemployment benefit systems in Central and Eastern Europe in the 1990s. It describes them and analyzes their costs by studying replacement rates, the shares of recipients among the unemployed, and a summary measure of benefit generosity. Moreover, it evaluates their distributive effects (via analyzing data from household income and expenditures surveys) and efficiency effects (via literature review). The evidence shows that unemployment benefits were progressive and that - in countries with broad coverage and sizeable share of benefits in household incomes - they also strongly reduced poverty. The paper also summarizes evidence about work disincentives created by unemployment benefits. HUMAN DEVELOPMENT NETWORKI: Abotut this series... Papers in this series are not formal publications of the World Bank. they present preliminary and unpolishedThe World Bank Employment Policy Primer aims to provide a comprehenisive, up-to-date resource on labor market policy issues. This resource is based on the lessons from research and operational experience in designing and implementing labor market policies and interventions. The primer series includes two types of products: primer papers thiat present, in some detail, new research results or up-to-date assessments of the literature and experience and primer notes that summarize "best practice" on policy-relevant topics in a very concise and accessible manner. Together, these products are intended to provide a practical tool- kit for people who make or are concerned about employment policy. A flexible format with products published both on our website and in published form ensures that the laiest developments are reflected. The Employment Policy Primer, including all papers and notes, can be accessed at the labor markets website at www.worldbank.org/labormarkets. For free copies of this paper, please contact the Social Protection Advisory Service, The World Bank, 1818 11 St., N.W., MSN 8-802, Washington, D.C. 20433-0011. Telephone (202) 458-5267. Fax (202) 614-0471. E-mail: socialprotection@worldbank.org.