14438 MnAac 'qq95 Labor Market Responses to a Change in Economic System Robert J. Flanagan The economic transitions in Eastern Europe have produced a contraction in the aggregate labor supply and the beginnings of a reallocation of labor from investment goods and beavy industry to consumer goods and services. The expansion of the pri- vate sector has reversed-if not eliminated-many of the labor market distortions created undsr central planning. Thus the first years of economic transition recorded increasing returns to human capital. There is some evidence that growlth in the pri- vate sector has led to more rational wage structures. Labor unions and minimun wage legislation have done little to inhibit labor market adjusments. Rather, labor market programs and incomes policies-with their emphasis on passive measures and wage subsides-have retarded the adjustment of wage structures and the restruur- ing of labor in state enterprises. But, the article observes, changing to active labor policies would probably not reverse the decline in employment. Unemployment in the transition economies is largely a matter of insufficient demand, and labor mar- ket policy is unlikely to have mucb impact until the mismatch between vocational training systems and modem production methods is solved. T rhe collapse of central planning systems has been dosely tied to a collapse of total factor productivity. In the early postwar decades labor force participa- otwn rates and capital resources increased in socialist countries, and their eco- nomic growth rates compared favorably with those of capitalist countries. But with the exhaustion of possibilities for extensive growth, these economies needed to use their resources more efficiently to achieve intensive growth, and failing this, their economic growth faltered. In command economies the increases in toral factor pro- ductivity growth needed to overcome slowing labor force and capital growth were not forthcoming. This sluggish total factor productivity growth, moreover, con- firmed market-oriented economists' warning that central planning processes would produce both allocative and X-inefficiency. Robert Flanagan is Konosuke Matsushita Professor of International Labor Economics and Policy Analysis, Graduate School of Business, at Stanford Universiq. The author wishes to thank John Earle for com- ments on carlier drafts Proceedings of the World Bank Annual Conference on Development Econonics 1994 @199S The International Bank for Reconstruction and Development / ThE woRID BANK 405 . .. t *. .f;,_,-. 406 Labor Market Responses to a Change in Economic System Today, however, the experience of economies in transition has turned this eco- nomic prediction on its head. Despite the strong presumption among Westem economists that transition from central planning to more market-oriented economies will eliminate the inefficiencies associated with central planning and produce higher growth, there is no clear vision of the processes through which this fundamental eco- nomic goal might be achieved. Gains in efficiency, moreover, might be offset initially by drastic declines in labor force participation. Macroeconomic stabilization has largely dominated the labor market research agenda during the early stages of Eastern Europe's transition. The importance of macroeconomic stability notwithstanding, there is a real danger that policies adopted for short-run purposes will conflict with fundamental long-run objecives. The damage such conflicts can do, moreover, is potentially greater for transition than for operating market economies. This artide focuses on the extent to which labor market policies, institutions, and behavior at the microeconomic level facilitate or impede economic restructuring. Any effort to tell a general story risks glossing over important differences among countries. It is equally risky to generalize from the experience of one country about develop- ments throughout a region. One must decde which variations really matter, or else there will be no story to telL The standard warning about data quality and availability also applies-particularly with regard to the private sector. Until better raw material is available, many issues of labor market adjustment easily explored in OECD countries must remain largely conjectural when speaking of former command economies. Aggregate Labor Market Adjustments The early phases of economic transition in Eastern Europe brought sharp declines in state sector output and real wages to virtually all the countries of the region. Table 1. Cbange in Labor Force Status of Working Age Popukltion, 1989-92 (percent distribution) Bzdgaria Czech Republic Hungar 1989 1992 Differene 1989 1992 Difference 1989 1992 Difference Men Employment 88.3 69.0 -19.3 81.2 753 -5.7 8S.5 70.9 -14.6 Unemployment 0.0 95 9.5 0.0 1.8 1.8 0.4 10.4 10.0 Inactive 11.7 215 9.8 18.8 22.7 3.9 14.1 18.7 4.6 Women Employment 85.1 62.1 -23.0 75.2 65.3 -9.9 78.1 64.3 -13.8 Unemployment 0.0 11.8 11.8 0.0 2.6 2.6 0.4 7.9 73 Inactive 14.9 26.1 11.2 24.8 32.1 7.3 21.6 27.8 6.2 Noat Figures are annual averages, except for the Czech Republic, where d=a are for the end of the year, and the Slovak Republic, where employnunt dam are for the end of the year. Flanagan 407 Declines in state sector employment ranged from about 33 percent in Hungary to about 20 percent in Bulgaria, the Czech and Slovak Republics, and Poland (Blanchard, Commander, and Coricelli 1993). In Romania employment in state enterprises dropped by about 23 percent, but that figure was closer to 40 percent when employment in cooperatives was included (Earle and Oprescu 1993). These large net declines in employment reflect the private sector's failure to expand its activity at a pace that absorbed the labor shed from the state sector. The different patterns of net employment change shown in table 1 reflect how much offsetting employment growth the private sector was able to generate. Among countries with intermediate declines in the state sector, the Czech Republic had the strongest private sector development and Bulgaria, Hungary, and the Slovak Republic the weakest. The allocation of net employment declines between unemployment (appearing for the first time since the onset of central planning) and departure from the labor force is determined in part by differences in incentives for market and nonmarket activity, influenced in part by labor market policy. Poliqc effects aside, however, the real wage declines in Eastern Europe should produce both substitution and income effects on aggregate labor supply. The substitution effect is visible in withdrawals from the labor force in response to deteriorating market rewards. The income effect is visible in the increased partcipation of individuals attempting to maintain family income as other family members lose their jobs. In virtually all OECD economies the substitution effect dominates, and labor supply elasticities are highest for women and for both sexes at the extremes of the age distribution (OECD 1983). Absent unusually large income effects on the labor supply in Eastern European countries, one would have expected the aggregate labor supply to -contract during the early phases of economic transition. Labor force participation rates did indeed decline in four of the countries surveyed. Consistent with Western experience, net withdrawals from the labor force were higher for women than for men in these four Poland Romania Sbvak Republic 1989 1992 Diff6ec 1989 1992 Difference 1989 1992 Differenca Men Employment 82.6 71.9 -10.7 87.S 81.0 -6.5 87.1 72.8 -14.3 Unemployment 0.0 9.4 9.4 0.0 S.3 5.3 0.0 8.8 8.8 inactivc 17.4 18.7 1.3 12. 13.6 1.1 12.9 18.5 5.6 Women Employment 68.6 58.3 -10.3 80.7 80.9 0.2 78.9 58.8 -20.1 Unemployment 0.0 10.6 10.6 0.0 9.3 9.3 0.0 9.7 9.7 Inactive 31.4 31.1 -0.3 19.3 9.8 -9.5 21.1 31.5 10.4 Sou;e.- Commission of thd European Communidies, 1993, p. 15. For Hungur, Poland, and the Czech and Slovak Republics the data are based on establishmen surveys and adminisative recors, cepr the 1992 data for Hungary are from a labor force survey. Bulgaian and Romanian da are from adminittive records (induding labor office reg- istions for unemployment) and estmates of privat employmenL 408 Labor Market Responses to a Cbange in Economic System countries. But in Poland labor force participation was close to a wash (with some increase for women), and in Romania it actually increased, mainly because of the large influx of women into the labor force. The labor force response of women in these countries is contrary to postwar experience in most market economies and recent experience in other Eastern European countries. It is, however, reminiscent of the old "additional worker effect" said to have influenced female labor force par- ticipation during the Great Depression. Relatively strong income effects are plausi- ble in Eastern European labor markets, even in the absence of labor market policy influences, given low asset levels and the rapid elimination of monetary overhang by early transition price shocks. The allocation of net employment dedines between unemployment and nonpar- ticipation was not uniform throughout the region. The comparative importance of increased nonparticipation stands out in the Czech Republic and-to a lesser extent-in Bulgaria and the Slovak Republic, while in Hungary, Poland, and Romania increased unemployment was more important. While policy-induced incentives appear to have affected each country's experience, the history is too short to establish the connection rigorously. Overall, unemployment insurance systems in transition economies are less generous than those in OECD countries (Scarpetta, Boeri, and Reutersward 1993).1 The Czech and Slovak Republics, although their eli- gibility requirements and replacement ratios are in the middle of the range, offer the briefest duration of benefits and (along with Bulgaria) impose the lowest maximum benefit payments, thereby discouraging long periods of joblessness. Preliminary evi- dence indicates that the generosity of the unemployment insurance schemes affects the allocation of net employment reductions between unemployment and labor force withdrawal in predictable ways. Unemployment Unemployment is influenced by labor market policies and institutions. North-American-style unemployment consists of relatively high inflows to and high outflows from joblessness, with most spells of unemployment being comparatively brief A scenario in which workers released from state enterprises queued for jobs for a brief period before moving into private sector jobs would produce such an unemployment pattern. In contrast, however, European-style joblessness is marked by low exit rates from unemployment-either because employers are reluctant to hire or workers are reluctant to accept jobs-and long durations of unemployment. Long unemployment itself, moreover, reduces the odds of getting a job. There is now considerable evidence that Eastern European countries have devel- oped super European-style unemployment (Blanchard, Commander, and Coricelli 1993; Boeri 1993). Inflow rates are lower than in most Western European coun- tries-in part reflecting the widespread use of early retirement incentives. But as the importance of this adjustment necessarily declines, future inflow rates should increase. While the majority of the newly unemployed in formerly command economies are workers released from state enterprises, there are also important Ftanagan 409 inflows from outside the labor force (particularly among inexperienced workers who have completed their schooling and cannot find a job) and from the private sec- tor. Outflow rates have been even lower than in Western Europe, and many spells of unemployment do not end with a job. While the low outflow rates reflect general demand problems indicated by extremely low ratios of job vacancies to unemploy- ment, there also appear to be important structural influences. In addition to the influence of unemployment compensation arrangements on the willingness to accept jobs, at least two structural policies can make employers in transition economies reluctant to hire. The former Czechoslovakia, Hungary, and Poland all have regulations governing collective dismissals that require consultation with employee representatives, advance notce of dismissals, and severance pay ranging from one to six months (Scarpetta, Boeri, and Reutersward 1993). While these costs may slow flows into unemployment, they also make employers reluctant to hire and thereby slow the movernent from unemployment to employment. Incomes policies that limit growth of the wage bill also discourage hiring. Since these regulations apply only to state enterprises, howeve&, their damage to restruc- turing may be slight. Labor Market Programs When unemployment is on the rise, a common recommendation is for greater spending on labor market programs. Eastem Europe is no exception, and the absence of rudimentary job-matching and safety-net institutions at the beginning of the economic transition added urgency to the recommendations. Some advocates of labor market programs suggest that such programs will substantially reduce unem- ployment in transition economies. Experience in the West, however, suggests that that is likely to be true only under special circumstances. The key questions, then, are whether current policies are likely to facilitate restructuring and whether labor market programs in general can reduce unemployment. By fiscal 1992-93 most transition economics had higher unemployment rates than most OECD countries yet spent less (as a percentage of GDP) on labor market policies that might improve the skill levels and the matching prospects of the unem- ployed. Moreover, public expenditure on labor market policy has concentrated on such passive measures as unemployment compensation and early retirement, mea- sures that may, in fact, extend joblessness and facilitate withdrawal from the labor force. Active measures, on the other hand, contribute directly to employment or the efficiency of future labor inputs (OECD 1993a). Acive labor market policies in transition economies have not focused on skill development. Expenditure is disproportionately weighted toward wage subsidies and public employment by comparison with such spending in Western economies. Although most wage subsidy expenditures in European countries do not go to state firms that should be shedding labor (OECD 1993a), much of the expenditure appears to support employees who would have been hired anyway. A case can be made for targeting wage subsidies to support new entrants into the labor force (to 410 LrOMaTrket Responses to a Change in Economic System provide work experience and on-the-job training) and those unemployed a long time (to prevent skill depreciation), but for reasons developed below, the case is not secure. Overall, however, current labor market policies throughout Eastern Europe do not appear to facilitate either restructuring or unemployment reduction. Nor would changing the mix of labor market policies be likely to alter substan- tially the region's unemployment experience. One difficulty is that active policies cannot solve the problem of insufficient demand. Training and related measures do not create vacancies but only help to fill them, and in Eastern European labor mar- kets the difficulty is not in filling vacancies but rather in their general absence. With limited demand, active labor market policies can change the order of workers in the job-seeking queue but not the length of the queue itself. Analysis of the effects of active labor market policies, moreover, rarely considers the extent to which the gains of targeted groups are obtained by displacing nontargeted groups (OECD 1993a). Policy measures, therefore, can seek to attain only such lirmited social gains as can be realized from changing the order in the unemployment queue. A second problem is that wage subsidies do not always deliver on their promise. In a full-information world wage subsidies should theoretically provide more jobs, yet the experience of some Western countries indicates that targeted policies can produce outcomes quite contrary to expectations. When information is unevenly distrib- uted-as when employers are imperfecdy informed about workers' productive potential, for instance-workers with superior abilities will need to signal their supe- riority in a credible manner, that is, in a way that less-qualified workers cannot imi- tate. Remaining outside labor market programs could be just such a signaL In the United States, for example, employer participation in targeted wage-subsidy pro- grams has been low. There is also evidence that employers prefer to hire unsubsidized workers, suggesting that they associate pubLic labor market programs with lower- quality workers. High-quality members of a targeted group have therefore effecively signaled their quality by remaining outside wage-subsidy programs-an option that is not available to low-quality workers, whose poor productivity would subject them to dismissal. Policies providing employers with general wage subsidies may mitigate this problem, but even when subsidies are paid to employers, targeting would raise the signaling issue. In Eastern Europe such ideas are far from theoretical. The private sector hires almost exclusively from state firms rather than from among the unemployed (Beleva, Jackman, and Nenova-Amar 1993; Commander and others 1993; Vecernilc 1993) and avoids using the new labor offices. While the signaling stigma appears to be less severe for training and other labor market programs designed to increase productivity, some taint may still apply. In summary, the labor force in Eastern European countries has contracted. Workers unemployed for long periods may become virtually unemployable, also reducing the effective future labor supply. New cohorts of school-leavers lack employment oppor- tunities, a particularly troublesome characteristic because the newest cohorts, who have the strongest incentive to make the major human capital investments required by change, are needed to accomplish major economic restmcturing. Flanagan 411 At this point unemployment in Eastern Europe is largely a demand problem. The exact nature of the structural problems that labor market policy might address will not become clear until there is a closer balance between job vacancies and unemployment. Yet the pattern of labor market programs in the region does not seem to be facilitating restructuring, and the emphasis on passive measures may even retard it. Sectoral Employment and Wage Adjustments Following the communist takeovers in Eastern Europe in the late 1940s, central planners sought to develop and expand investment goods and heavy industry at the expense of consumer goods and services. With some exceptions, job matching was left to workers and employers, while central planners developed national wage structures to encourage workers to follow planning preferences. National tariff wages were established for job categories under a national job evaluation that gave weight both to considerations familiar to Adam Smith, such as the qualifications required and onerousness of work, and to planning objectives, such as the social importance of the industry. Before long, industry wage structures throughout Eastern Europe had adapted to the tariff wage structre, with large increases in rel- ative wages in industry and construction and decreases in trade and services (Adam 1984). Tariff wage policies reduced both returns to human capital and interregional wage differentials and weakened the link between managerial performance and pay. Overall wage levels were kept low to permit high investment levels. Low wages produced high quit rates, but the range of labor mobility was severely circumscribed by housing shortages. To counter management efforts to attract and retain workers, furthermore, central planners imposed tax penalties for excessive increases in enter- prise wage bills-an early form of tax-based incomes policy. Labor unions in centrally planned economies had no bargaining role. Their main role was to challenge dismissals. Whether labor's weakness under central planning helped or hindered efficiency depends on whether "monopoly" or "voice" theories of union behavior were more applicable in Eastern Europe. Limitations on labor exit and voice opportunities, however, demonstrably reduced job satisfaction to well below Western levels (Blanchflower and Freeman 1993). To some degree, all Eastern European labor markets have moved away from the outcomes influenced by central planners. Since 1989 employment has shifted from industry and agriculture toward services. Within the service sector, employment has grown especially in trade, health, and education-sectors generally neglected under central planning. Private sector employment has grown in all countries (albeit at very different rates), but the private sector offers heterogeneous job opportunities that range from small service organizations with one or two employees to large manufacturing oper- ations. In most countries large-scale privatization has proceeded slowly, and state enterprises remain the main source of employment. The private sector's share of total employment is largest in Poland (45 percent) and Hungary (35 percent), both 412 Labor Market Respons to a Change in E&onomic System of which had significant private economic activity before the late 1980s, but throughout the rest of Eastern Europe it remains at less than 20 percent. To what extent, then, has private sector expansion helped to reverse the labor markets' vestiges of planners' preferences? To address this question, we must exam- ine private sector wage policies and the nature of labor market competition between the state and private sectors. Public-Private Pay Differentials The analysis of labor market competition between private and state firms in Eastern Europe is hampered by limited documentation of the private sector. The following discussion, therefore, considers three levels of information in an effort to ascertain the role of wages in competition between the state and private sectors: official data (uncontrolled for human capital and other compositional influences) on average wages in each sector, household surveys of individuals (with statstical controls for some influences on wages), and recent World Bank surveys of private firms in the region.2 For Bulgaria there is only survey evidence ro show that private firms tend to pay higher wages but lower benefits and bonuses (Beleva, Jackman, and Nenova-Amar 1993), and no reliable information has been reported for Romania (Earle and Oprescu 1993). Two studies of Czech data find substantial differentials. An analysis of a June 1991 government survey of wages in the Czech Republic found that men working in the private sector earned more than one and one-half timnes what men of the same education and age range earned in the state sector. The differential was smaller for men over forty-five years of age and negligible for women on the basis of average monthly wages (Flanagan 1993a). (At this early stage of Czech privatization, how- ever, the sample contained few private sector observations.) A year later, in June 1992, Vecerncik's (1993) analysis of the Survey of Economic Expectations and Attitudes found-after controlling for age, education, and broad occupational and industrial categories-that employees in private firms were earning about 13 per- cent more than employees in state enterprises and about 19 percent more than employees in cooperative organizations. In Hungary wage differentials between private and state firms emerged before the fall of communism, yet the Hungarian Household Panel data suggest that wage dif- ferences are somewhat smaller there thian in the Czech Republic. After controlling for age, schooling, occupation, and broad industrial category, Commander and oth- ers (1993) found that employees in the pardy private sector had monthly incomes (wages and earnings, including all cash payments) that were about 9 percent higher than those offered in the state sector-and 6 percent higher for the fully private sec- tor. The results are much weaker for hourly earnings, possibly reflecting the greater use of profit-sharing and other bonus arrangements in the private sector. Respondents to the World Bank survey of private manufacturing indicated that they tried to pay higher salaries than the state sector (Webster 1993a). Flanagan 413 Official Polish statistics indicate that average private sector wages exceed average state wages in construction and transportation, but that state wages (which do not dis- tinguish between the old cooperatives and the new private sector) are higher in indus- try and trade. There is some evidence that private sector wages tend to be higher for high-wage jobs and lower for low-wage jobs (Coricelli, Hagemejer, and Rybinski 1993). But for labor market behavior the key comparison is for jobs in which the pri- vate sector competes with the state sector. According to a World Bank survey of pri- vate manufacturing firms in Hungary and the Czech Republic, entrepreneurs reported that they typically paid unskilled workers 10 percent more than they could have made in state enterprises and skdlled workers what they could afford, which could be up to twice what state enterprises were paying (Webster 1993a; Webster and Swanson 1993). To some extent, however, higher wages in the private sector are offset by higher nonwage benefits paid by the state. It is widely believed-if not precisely docu- mented-that fringe benefits and services are greater in the state sector. The difference appears to be greatest for the largest state firms. The World Bank survey of private manufacturing in Hungary found that fringe benefits were indeed lower than in the public sector but not by enough to offset wage differences. Nonwage benefits within the private sector are no doubt smallest in the smaller service firms, which are not well covered by current statistical systems. Bank surveys also indicate that private sector employers are likely to use dismissals as disciplinary and motivational tools. Although there is some turnover among private firms, recent large-scale employment losses in state firms have sharply reduced differences in job security between the sectors In summary, where private sector firms compete directly with state firms for par- ticular skills, private firms appear to pay higher average wages, although the mar- gin and composition of the overall compensation package vary aaoss sectors and countries. Since the private sector appears to offer fewer nonwage benefits, some of the smaller wage differences may reflect compensating differentials. The larger wage differentials indicate two aspects of the labor reallocation process in transi- tion economies. First, the prevalence of a private sector wage premium in most transition economies implies that new firms favor the relative wage mechanism over the job vacancy mechanism as a strategy for attracting workers from state enterprises. Given current conditions in Eastern European labor markets, the existence of job vacan- aes in private firms should be sufficient attraction to the unemployed even in the absence of a wage differential. Throughout Eastern Europe, however, private sector employers are filling most vacancies by hiring employees directly from state jobs rather than from the pool of unemployed (Beleva, Jackman, and Nenova-Amar 1993; Commander and others 1993; Vecernik 1993). A large proportion of work- ers who lose state jobs and leave the unemployment rolls return to jobs in the state sector. Indeed in Hungary the private sector is a net contributor to unemployment. The composition of the private sector pay package is also consistent with a strat- egy for selecting out of the state sector workers who are generally younger and more motivated to acquire extra rewards for extra effort rather than to seek social benefits. The preliminary impression is therefore that private firms are using wage policy to 414 Labor Markn Responses to a Change in Economic System attract relatively qualified workers away from the state sector and that unemploy- ment is seen by employers as a signal of lower qualifications. Human Capital Returns Compared with rates of return in market economies, returns to schooling and expe- rience under central planning were positive but low. Average returns to schooling in 1988 for men in the Czech Republic were lower than in any of a comparison group of market economies (table 2).3 Returns for Hungarian men, which were about 1 percentage point higher, were well within the range of estimates for market economies with centralized bargaining systems (German Federal Republic, Norway, Sweden), which often share the socialist economic objective of narrowing the struc- ture of earnings across occupations. Rutkowski (1994) reports similar pretransition returns in Poland. Differences in returns are particularly notable at the university level. Relative to a high school education, a university degree provided a wage pre- mium of 65 peś.cent (in 1987) for U.S. males, 32 percent (in 1990) for Polish males, and 16 percent (in 1988) for Czech males (Goldin and Margo 1992; OECD 1993b; Flanagan 1993a). Pretransition wage structures also showed comparatively low returns on experience potential. The pretransition pattern of returns to education reflects the type of human cap- ital development encouraged under central planning. Central planners tended to value physical over mental work and productioni over service sectors. Young men were oriented toward vocational schools and apprenticeships, and-except for the nomenklaura-university-level education was devalued. In the late 1980s roughly 8 to 10 percent of the labor force in centrally planned economies had a university education, compared with 10 to 20 percent in most advanced market economies. Eastern European enrollment rates in secondary and higher education were also Tablc 2. Returns to Schooling, 1980s (percent) Country,year Men Women Czech Republic, 1928 3.8 5.4 Hunpry, 1928 4.7 6.1 Australia, 1936 4.S 4.1 Federal Republic of Germany, 1985-88 4.9 SA Norway, 1982 4.5 3.8 Sweden, 1929 4.3 4.3 Switzerland, 1987 5.5 7.4 United Kingdom, 198S-88 7.0 9.3 United States, 1985-88 7.0 8.1 Note: All estinates aze from a basic human capital earnings function specification in which the natural loaridun of dhe wage is a function of schooling. Porenial experience is specified as a quadratic. AUl estimates are significant at .01 level. S&ow Flanagan 1993b; unpubished regression resuls provided by Francine Blau and larry Kahn. Flanagan 415 lower than in industrial countries. Most enrollments in secondary education in Eastern Europe were in vocational programs, compared with 35 to 40 percent in continental European market economies and about 10 percent in the United Kingdom (Boeri and Keese 1992). This appears to be the major distortion in human capital formation under central planning. Central planning appears to have produced overinvestments in vocational and apprenticeship education and underinvestments in university education. To the extent that markets correct distortions associated with planner preferences, returns to university education should increase while those to vocational training should decrease during economic transition. In fact, when the tariff wage system was abandoned, overall wage variation increased and returns to schooling began to rise. By mid-1991-just eighteen months into the transition-the average rate of return to schooling for men in the Czech Republic had increased by 0.6 percentage point, reflecting rapidly declining returns to vocational education and rising returns to university education (Flanagan 1993a). (Returns for women were unchanged.) Returns in Hungary rose from 4.7 percent in 1988 to 7.5 percent in March 1992 (Commander and others 1993). In several countries, vocationally trained workers not only saw their wages fall relative to those of university graduates but also experienced relatively high rates of unem- ployment. In contrast, in Romania, where there has been little private sector devel- opment, the wage structure for different schooling levels barely changed in the early transition years. In 1992 a Romanian university degree provided only a 30 percent wage premium over high school completion. Contrary to the pattern in other tran- sition economies, post-transition movements in industrial relative wages in Romania appeared to reinforce the traditional preferences of central planners (Earle and Oprescu 1993). With the exception of Romania, then, the first years of economic transition have reversed the main pretransition distortions in human capital forma- tion brought about by central planning. What mechanisms produced these changes? In a well-functioning market, changes in the structure of wages in the state sector would mirror pay developments in the private sector, as state firms attempted to retain employees. For most countries in Eastern Europe except the Czech Republic, however, there is too little data on the private sector to test this hypothesis. In the Czech Republic the emerging private sec- tor appears to be competing primarily for middle-level educated workers rather than university-educated workers. Rising returns to schooling appear to be an artifact of the pattern of structural changes in the state sector rather than of increased demand in private business. Public enterprise restructuring has increased the relative impor- tance of industries that have always provided the highest returns to university educa- tion. Higher returns to education are available in the market, but not in most sectors. Institutional Influences From the perspective of labor in transition economies, the key questions are whether labor market institutions inhibit transfers to the private sector and the development 416 Labor MarkAt esponses to a Change in E&onomic System of more rational pay structures. In the case of unions there is also the concern over their potential for generating wage pressures. Labor Unions Eastern Europe has only recently introduced collective bargaining over wages and working conditions. The extent of union representation, moreover, is difficult to assess given that membership figures are provided by the unions-a dubious source even in Western countries. Early estimates place union densities across Eastern Europe in the range of 35 to 70 percent (Freeman 1992). In no country of the region is the domain of unions as broad as in the most unionized Western countries (such as Scandinavia) and in no country is it as limited as in the least unionized Western countries (France and the United States). Throughout Eastern Europe unionization is ersentially limited to the state sector and is far from complete even there. Most private firms are nonunion and have lit- tle expectation of becoming unionized. World Bank surveys of firms in the region found little evidence of unionization. In the former Czechoslovakia none of the sur- veyed firms had unions, although about 10 percent of firms had workers councils that mostly abided by earlier contracts made under state ownership (Webster and Swanson 1993). None of the firms in the Bulgarian survey were unionized (Beleva, Jackman, and Nenova-Amar 1993), nor were the few firms in Romania's private sec- tor (Earle and Oprescu 1993). In Hungary employers were amused that interview- ers asked whether their workers bargained through a union or council (Webster 1993a; Lado 1993 reports similar findings). In Poland neither unions nor workers councils represented workers in any of the establishments surveyed, and employers claimed that their workers did not see the need for unions (Webster 1993b).4 No respondent in any of the firms surveyed listed unions as a constraint on the firm's progress. It appears that private employe.-s can set pay levels, pay structures, and employment conditions unilaterally, subject only to state regulations. Given the distribution of union representation in Eastern Europe, differences between state and private wages provide tolerable estimates of union-nonunion wage differentials, since unions only have a life in the public sector. The evidence therefore indicates that the union wage 'premium" in Eastern Europe is largely negative-a result that is either rare or nonexistent in Westem countries. That wages in the declin- ing (public) sector are lower than in the expanding (private) sector is good news for the allocational objectives of the economic transitioni and should challenge the wide- spread presumption that incomes policies are needed to restrain union power in the state sector. The absence of unions in the private sector, moreover, has enabled employers to lay off workers as a response to performance and discpline proolems. Union power is notoriously difficult to measure for a single country, and cross-country comparisons are particularly hazardous where unions have such dif- ferent objectives as wage levels, security of employment, and income equality. Even within a single country, moreover, reports from the field are often conflicting. Many observers, for instance, stress the potential power of the Polish workers councils, Flanagan 417 which have had a highly variable effect according to enterprise surveys. Managers in Poland's state enterprises cite incomes policy regulations and not pressure from workers organizations as ahe major barrier to restructuring (Pinto, Belka, and Krajewski 1993). Similarly, Romanian unions cannot mobilize their members for widespread collective acton even though they represent about 60 percent of work- ers and, according to one survey, have negotiated agreements covering about 90 per- cent of employment (Earle and Oprescu 1993). Unions in Hungary and the Czech Republic have been particularly quiescent. Because of the confounding effect of incomes policy regulations in Eastern Europe, the effect of unions on the wage structure is difficult to assess. Nevertheless, the decentralization of some bargaining structures, such as those in Bulgaria and Romania, has contributed to pay dispersion among enterprises. In state enterprises unions continue to inhibit the use of dismissals for poor perfor- mance or disciplinary reasons. They may also be able to impose conditions on restructuring when large state enterprises are privatized. Although potentially sig- nificant, these union restrictions on operation are poorly documented. Unions can- not, however, thwart transfers to the private sector because workers can quit state enterprises with impunity. Minimum Wage Policies Throughout Eastern Europe minimum wage policies have been instituted to assist the least productive workers, with little consideration for any potentially adverse effects on employment At this stage of massive state employment reductions in Eastern Europe, howev.r, it is impossible to isolate the effects of minimum wage policies on employmenc. Attention has therefore focused on their effects on the wage strucure. Relatively high minimum wages produce (or maintain) compressed wage structures, which could inhibit the reallocation of labor from less productive to more productive tasks. Some Western countries index the minimum wage to consumer prices, which was also the initial intention in several Eastern European countries. But where other wages were not indexed (or were indexed less generously), some unions (notably in Czechoslovakia) negotiated tripartite agreements with explicit real wage reductions. In addition to affecting the wage structure, minimum wage poli- des also have direct implications for government budgets, since many social bene- fits (pensions, family allowances, unemployment benefits) are linked to the level of the minimum wage (UNECE 1992). Faced with these considerations Eastern European governments have moderated and even abandoned explicit indexation procedures, and minimum wages declined as a proportion of the average wage to levels well below those typical of industrial market economies (table 3). Thus far, there is little evidence that minimum wage policy has seriously interfered with wage structure adjustment during transition. To summarize, two distinct wage determination mechanisms have emerged during the economic transition in Eastern Europe. In the growing private sector, one mech- 418 Labor Market Responses to a Change in Economic System Table 3. Minimum Wage Relative to Average Gross Wage, Selected Countries, 1989-93 (percent) Czech Slouak Year Bulgania Republic Hungary Polnd Romania Republic 1989 S.l - 35.0 11.6 63.7 1990 44.6 - 42.0 21.4 73-0 1991 5.4 S3.0 39.0 34.7 60.3 52.0 1992 353 45.9 36.0 37.0 45.7 47.5 1993 First quarter - 41.1 39.0 41.1 - 46.4 Second quartcr - 36.6 35.0 39.3 - 42.5 - Not avaiable. Sow: Commission of the European Communities 1993. anism resembles the nonunion wage determination found in some advanced market economies. Because employers are subject to hard budget constraints, performance must uliimately support pay lelveEs. Unions are absent, and-subject to market conditions and incentive consideration-employers appear to have considerable discretion in wage-setting. In the large (but diminishing) state sector, enterprises are still subject to soft budget constraints. In many state enterprises unions or workers councils exist and have initiated collective bargaming, although at this point neither union activity nor government minimum wage legislation has substantially influenced relative wage structures in the state sector. Unions may, however, have slowed the downsizing of the state sector. O azatonal Efficiency during Transition The discussion so far has focused on improved allocation of labor as a source of pro- ductivity gain in transition economies. In a world of long job tenures and "career- relationships, productivity gains also depend on incentives in firms' internal labor markets. Training systems, performance evaluations, and payment systems that link pay to performance all contribute to internal efficiency, and all these mechanisms were generally absent from public enterprises under central planning. The problem of determining appropriate managerial pay mechanisms, therefore, is particularly acute in transition economies. Information on the role of the emerging private sector in introducing more sophis- ticated internal wage payment incentives and human resource management systems is regrettably scarce. The practices of foreign firms and joint ventures (which link wages to performance through merit systems, profit sharing, and other performance- related bonuses) contrast sharply with past practice in the state enterprises, where the large nonpecuniary component of compensation was unrelated to performance. It is not yet known to what extent market incentive practices have been adopted by state enterprises and the indigenous private sector in Eastern Europe. Sharper F1anhan 419 competition between the two sectors would hasten such transference, but private and state firms rarely compete direcly in product markets. Where competition does occur, incomes policies tend to impede the efforts of state enterprises to emulate the incentive systems of private firms (Pinto, Belka, and Krajewski 1993). Evidence is lirnited to Hungary, Poland, and the Czech Republic. Private manu- facturers responding to World Bank surveys noted that vocational schools and the organization of work in state enterprises had produced overspecialized workers who lacked the flexibility to adjust to new technologies (Webster and Swanson 1993; Webster 1993a,b). Many respondents also reported poor motivation among employ- ees who had worked for years in settngs where there was little connection between produceivity and profitability. Only in Hungary-where workers no longer looked to the state sector for employment security-did entrepreneurs comment favorably on worker motivation. When faced with nonproductive employees, most Hungarian employers were quick to lay off workers and establish acceptable norms (Webster 1993a). A World Bank survey of private manufacturers in Poland found that several employers had instituted profit-sharing schemes and escalating production incentives (Webster 1993b). From the limited evidence it appears that the changing wage signals observed in the aggregate data do not necessarily describe changes within organizations. This dis- junction is mainly an issue for state organizations. Training systems developed under central planning, moreover, are inadequate for modern jobs and work organization. Macroeconomic Policy and Micro economic Objectives While the government's role in pay deterniination has diminished throughout Eastern Europe, all countries in the region adopted incomes policies as part of their transitional macroeconomic policy packages. Such policies present an inherent ten- sion between macroeconomic and microeconomic objectives in any economy, since the policy rules advanced to contain upward wage pressure may distort wage struc- tures. This tension is particularly acute in Eastern Europe, where wage structures :re much furither from equilibrium than in the West. Incomes policies that thwart the development of incentive structures in labor markets can therefore inhibit labor reallocation and restructuring. The Case for Incomes Policy Two motivations underlie most applications of incomes policies in Western coun- tries. The first is a belief that market power, particularly that of labor unions, is a significant source of upward wage pressure and that postwar economic develop- ments were favorable to union bargaining power. A second motivation is to reduce inflationary expectations. Incomes policies have thus been used in efforts to prevent the emergence of inflation and to wind down existing inflation. Eastern European governments face a third consideration. As the residual claimants on the incomes of state enterprises, Eastern European governments have a distinct budgetary interest 420 Labor Markt Responses to a Change in Econcomic System in moderating wages in the state sector. Because government tax revenue depends on the difference between sales revenues and the wage bill of state enterprises, wage increases in the state sector implicitly lower tax revenues. The standard arguments for incomes policies appear weak in Easten Europe. The bargaining power of most unions in Eastern Europe is limited. Policies to restrain inflationary expectations would cover the entire economy, not just the state sector. Nor can resort to incomes policies rest on the demonstrable achievements of these policies or on careful analyses of compliance incentives and responses at the micrceconomic leveL In the West, policies based on explict guidelines for wage and price increases or social contracts (or multilateral negotiations among labor, man- agement, government, and other economnic interest groups) regarding the growth of incomes have foundered because of unresolved distributional conflicts and weak to nonexistent compliance incentives. Such policies have never achieved durable suc- cess, with the possible exception of Austria (Ulman and Flanagan 1971; Flanagan, Soskice, and Ulman 1983). Tax-based incomes policies, which have been more mlked about than used in the West, would give firms and workers tax incentives to moderate wage and price increases, thereby providing incentives for compliance and scope for allocational flexibility, since finns could exceed the guidelines and ratio- nally incur the tax if economic circumstances warranted. The inexperience of and lack of incentive for Eastern European managers in the state sector to bargain hard provides a stronger basis for concern about wage pres- sure. Soft budget constraints have permitted state-owned enterprises to be sloppy about costs and weak m resisting wage demands. Managers who expect to revert to employee status as their enterprises are privatized, furthermore, have even less incentive to resist union wage demands. Weak bargaining effectively reduces gov- ernment revenues from state enterprises. Through incomes policy governments in the region effectively avoid ceding authority for tax reductions to enterprise man- agers. The case for such policies in Eastem Europe thus rests more on the weakness of management than on the strength of workers organizations. It remains to be seen whether incomes policies actually stem inflation. Policy Effects Eastern Europe has adopted tax-based incomes policies-the policies advocated in the West-because they seem to offer greater allocational flexibility than other pol- icy designs. Yet the mechanism of these policies seems flawed when applied to the state enterprises of the region. Incomes policies are supposed to increase manage- mentes bargaining resistance by threatening the loss of a firm's marginal profits as the penalty for noncompliance. These policies are best suited for managers who are accustomed to seeking profits. Such policies should follow rather than precede the development of property rights and hard budget constraints, since management must face financial incentives in order to take the consequences of noncompliance seriously. Policies motivated by managerial weakness under soft budget constraints, therefore, cannot depend for their effectiveness on managerial responses to (nonex- Flanagan 421 istent) hard budget constraints. The willingness of many Eastern European managers to grant "excess" wage payments despite punitive taxes underscores this flaw, and governments are often unable to collect such taxes (Jackman and Pages 1993; Earle and Oprescu 1993). Even where successful, tax-based incomes policy designs in Eastern Europe do not allow the allocational flexibility permitted in poliqc formulations in the West. Tax-based incomes policies have a longer history in centrally planned economics than in the West because they were used by central planners to restrain the growth of enterprise wage bills to limit consumption demand. Such policies included puni- tive tax rates for wage increases in excess of the target-a feature that has been incorporated in the current incomes policies of transition economies. Thus Eastern European incomes policies have great potential to interfere with government efforts to achieve greater allocational efficiency through systemic reform. A policy that targets the average wage, for instance, can penalize companies that downsize to shed their least-efficient workers or those that expand and hire work- ers who are more highly skilled. In each case changes in work force composition could raise the average wage sufficiently to incur a penalty tax. Average wage tar- gets also create incentives to freeze or further compress the internal wage structures of organizations-the opposite of what is desired, in most cases, to provide incen- tives for training and additional effort. Policies that target the enterprise wage bill, by contrast, do not penalize the downsizing of the work force. Indeed, the down- sizing can be accompanied by an increased average wage to attract and retain supe- rior workers. However, a wage bill target taxes employment growth, making firms reluctant to hire, slowing outflows from unemployment, and creating incentives for further skill compression. Evidence from Eastern European labor markets increasingly indicates that these concerns are far from theoretical. Both wage bill and average wage targets have been applied to Polish enterprises, for instance, at different times. Under the average wage target, managers of state-owned enterprises complained of restraints on downsizing and restructring. They clearly preferred wage bill targets. They also reported that wage targets had induced pay compression and that policy discrimination between state and private employers inhibited public enterprise managers from transforming their enterprises and emulating the behavior of private managers (Pinto, Belka, and Krajewski 1993). Wage structure developments at the sectoral level in Bulgaria, the Czech Republic, and Hungary are consistent with these reports. Without violating the incomes policy wage rules and incurring penalty taxes, state enterprises in Bulgaria were unable to alter the compressed wage strucmre inherited from central planning (Beleva, Jackman, and Nenova-Amar 1993). In the Czech Republic, although the overall return on schooling-especially the return to university education in the state sector-increased rapidly during the first eighteen months following the 'vel- vet revolution," the relative wage of university graduates fell in virtualy all indus- trial sectors. The general increase, therefore, was an artifact of structural change (Flanagan 1993b). Similarly, in Hungary, although the relative wage of professional 422 Labor Market Responses to a Change in Exonomic System workers rose by 8 percent between 1989 and 1992 in industry as a whole, there was little change in the relative wages of professional workers or manual workers within sectors. Nor was much change in pay differentials for skill discernable among branches of Hungarian industry (Jackmran and Pages 1993). When wage policy is applied to sectors with a large nonwage component, it encourages the substitution of nonwage (and untaxed) benefits for wages, diluting the incentive effects of the compensation system. There is evidence that tax-based incomes policies in the USSR 'merely induced enterprises to shift further still to the use of bonuses and direct payrnents in the form of goods and services, often acquired on a barter basis from other enterprises.... At the macroeconomic level the ratio of nonwage to wage forms of remuneration rose steadily from the 1950s onward, at least until the advent of perestroika" (Standing 1991/92, p. 244). If incomes policy has not had a beneficial effect on wage structures in Eastern Europe, its use to achieve macroeconomic stability has had equally questionable results. In several of the countries where it was tried, wages were actually below incomes policy norms in the early months of the transition. The fear of a wage push from workers was strongest in Poland. Yet recent analyses have conduded that wage policy was not effective in Polandliquidity constraints at the beginning of 1990 made such policy redundant (Coriceili, Hagemejer, and Rybinski 1993). Other countries also stressed the role of liquidity constraints and concerns for job security as more important reasons for wage setting (Jackman and Pages 1993). No analysis of Eastern European labor markets in the World Bank country studies has claimed evidence of a substantial macroeconomic effect of the policies. In summary, incomes policies seem rather poorly suited for the transition envi- ronment. Such policies always present a tradeoff between macroeconomic and microeconomic objectives, but the tradeoff seems particularly acute in transition economies. To the extent that incomes policies are effective in their macroeconomic objectives, their potential for inhibiting the allocative functioning of wage structures is great. While this potential may be acceptable in market economies with strong allocational signals, in Eastern Europe today such policies impose delays on adjust- ment and restructuring that run counter to systemic reform. Even if incomes poli- cies were instrumental in stabilizing the macroeconomy (which has yet to be shown), the lack of change in wage structures in the region would be troublesome. Conclusion With few exceptions the aggregate supply of labor in Eastern Europe has contracted and threatens to contract still farther if the long duration of unemployment (implied by current flows in and out of unemployment) continues. The principal problem is insufficient demand. Until the demand problem is solved, moreover, labor market policy alone is unlikely to have much impact on unemployment. When job vacan- cies and unemployment are more closely aligned, a reorientation of labor market policies away from passive measures and wage subsidies toward training programs seems desirable. There is ample testimony from the emerging private sector regard- Flanagan 423 ing the limitations of traditional Eastern European vocational training for produc- ing workers skilled in modern production methods. And the problem can only worsen as the private sector expands. With respect to the allocation of labor the evidence is mixed. New private firms appear to have little trouble expanding (subject to the training problems noted above), but they take few risks in hiring the unemployed. Within the state sector there is evidence that some of the distortions in labor market signals under cen- tral planning have been reversed, if not eliminated, at the market level. Limited evidence indicates, however, that the private sector has not yet become the force driving these changes. Rather, the changes have resulted from the downsizing of state enterprises. In virtually all the transition economies in the region the major institutional bar- rier to adjustments of pay structures and to employment restructuring within the state sector has been incomes policy regulations. These policies seem poorly designed for the transition environment, and their use rests more on tradition than any demon- strated effectiveness. A better option might be to alter management bargaining incen- tives by altering management compensation schemes, which would give managers a personal stake in resisting demands for pay increases. Broader use of arrangements that link pay to enterprise surpluses could also provide incentives to resist general pay demands and create stronger internal performance incentives. Each country, further- more, could fix its exchange rate to a basket of Western currencies-much like the hard currency policy adopted by Austria for many years. While this solution does not directly address the problem of managerial bargaining incentives, it would under- score the fact that excessive wage increases could lead to job loss and would there- fore strengthen managers' bargaining hand. Notes 1. Atkinson and Micldewright (1991) warn chat the broad descriptive parameters of unemployment inrance systems may be a poor guide to the incentives actually at work. To discover their true effccts on job acceptance, detailed research into the implementation of these systems wilU be necessary. 2. Three surveys covered the manufacturing sector and included interviews at 121 firms in the former Czechoslovakia in January 1992, 106 Hungarian firms in September 1991, and 93 Polish firms in May 1991. In each country firms were drawn randomly from the population of registered, majority privately and domestically owned manufacturing firms with seven or more employees. Sec Webster (1993a,b) and Webster and Swanson (1993) for details. A fourth survey of 200 firms in the two largest cities in Bulgaria included respondents from industry, trade, construction, business services, and publishing (Beleva, Jackman, and Nenova-Amar 1993). 3. These conclusions would not be altered by comparing returns to schooling in centrally planned economics with returns in countries whose level of economic development is lower than that of advanced market economies, for returns to schooling red to be inversely related to stage of development. A sur- vey of the international literature on educational returns reports baUpark returns to schooling Cderived from the specifications used here): Africa, 13 percent; Asia, 11 percent; Latin America, 14 percent coun- tries with an intermediate level of economic development, 8 percent; and advanced industrial countries, 9 percent (Psacharopoulos 1985, table 3). 4. The firms surveyed, like most private firms in Eastern Europe at tks time, were relatively smal, "with 60 percent having fewer than twenty workers and almost 75 percent of firms reporting monthly sales of less than US$50,000." Most firms in the survey were limited-liability companies owned by groups of three to six people. Ninety percent had originated as private firms, mainly since 1988. The rest orig- inated in the state or cooperative sectors (Webster 1993b, pp. 2-3). . 424 Labor Market Responses to a Chan in Economic System References Adam, Jan. l984. Employment and Wage POlcies in Pblanit Czechos kia, and Hungary Since 19SO. New York: Sr. Martin's Press. Aikinson, Anthony B., and John Micklewright 1991. "Unemployment Compcnsation and Labor Marker Transitions: A Critical Review." Journal of EFconomic liteature 29(4): 1679-27. Bcleva, Iskra, Richard Jackman, and Maricla Nenova-Amaan 1993. "The Labour Market in Bulgaria.' Paper prepar-d for World Bank Confcrcnce on Unemployment, Restructuring, and the Labor Market in Eastern Europe and Russia, October 7-8, Washington, D.C. Blanchard, Olivier, Simon Comnmander, and Fabrizio Coricelli 1993. "Unemployment and Restructring in Eastem Europe." Paper prepared for World Bank Conference on Unemployment, Restructuring, and the Labor Market in Eastern Europe and Russia, October 7-8, Washington, D.C Blanchflower, David, and Richard B. Freeman. 1993. "The Legacy of Communist Labor Relations." Discussion Paper 180. Centre for Economic Policy Research, London. Boeri, Tito. 1993. "Labour Market Flows and the Persistence of Unemployment in Central and Eastern Europe." Paper prepared for Workshop on the Persistence of Unemployment in Central and Eastern Europe, OECD, Center for Cooperation with Economies in Transition, September 30- October 2, Paris. Boeri, Tito, and Mark Keese. 1992. "Labour Markets and the Transition in Central and Eastern Europe." OECD Economic Studies 18: 133-63. Commander, Simon, Janos Kollo, Cecilia Ugaz, and Balmz Vilagi. 1993. -Hungary." Paper prepared for World Bank Conference on Unemployment, Rcstructuring, and the Labor Market in Eastern Europe and Russia, October 7-8, Washington, D.C. Commissior of the European Communities. 1993. Emplo,ment Observatry: Central and Eastern Eumpe- Brussds. Coricelli, Fabrizio, Krzysztof Hagemejer, and Krzyszrof Rybinski. 1993. "Poland." Paper presented at World Bank Conference on Unemployment, Restructuring, and the Labor Market in Eastern Europe and Russia, October 7-8, Washington, D.C Earle, John S, and Gheorghe Oprescu. 1993. "Aggregate Labor Market Behavior in the Restuctuing of the Romanian Economy.' Paper presented at World Bank Conference on Unemployment, Restucturing, and the Labor Market in Eastern Europe and Russia, October 7-8, Washington, D.C Flanagn, Robert J. 1993a. "Labor Market Policy, Information, and Hiring Behavior? Paper presented * Conference on Institutional Frameworks and Labor Market Performance, November 27-28, Nuremberg, Germany. -.= 1993b. "Werc Communists Good Human Capitliss? The Case of the Czech Republic." Working paper. Stanford University, Graduate School of Business. Flanagan, Robert J., David Soskice, and Lloyd Ulman. 1983. Unionism, Economic Stabilizain, and Incomes Policies: European Experience. Washington, D.C.: Brookings Institution. Freeman, Richard B. 1992. 'What Direction for Labor Marker Institutions in Eastcrn and Central Europe?" NBER Working Paper 4209. National Bureau of Economic Rescarch, Cambridge, Mass. Goldin, Claudia, and Robert A. Margo. 1992. Ye Grcat Compression: The W'age Structure in the United States at Mid-Century? Quarterly Journal of Economics 107 (February): 1-34. Jackman, Rihard, and Carmen Pages. 1993. "Wagc Policy and Inflation in Eastern Europe." Paper pre- pared for World Bank Conference on Unemploymcnt, Restrucuring, and the Labor Market in Eastem Europe and Rnssia, October 7-8, Washington, D.C. lado, Maria. 1993. "Wages, Incomes Policy, and Industrial Relations." In G. Fischer and G. Standing, cds, Structral Cbange in Cenral and Eastern Europe: Labour Market and Social Poicy Implcatios Paris: OECD. OECD (Organization for Economic Cooperation and Dcvelopment). 1993a. Employment Outlook. Paris. - . 1993b. Str r Change in Centrl and Eastern Ewope Labor MarkEt and Social Iblicy Impllionm Paris. Pinto, B&, MC Bclca, and S. Krajewuld 1993. 'Transforming Statc Enterpriscs in Poland: Evideoce on Adjustment by Manufaturing Firma." Brookings Papers on Economic Activiy 1.