In choosing the capitalist approach to allocating labour to its best alternative uses, the economists of Central and Eastern Europe were forced to accept its most unpleasant consequence: unemployment. This extensive restructuring and reallocation of human capital has rarely been seen in capitalist economies, and the process has not yet been completed. A sizeable fraction of the working age populations of these countries has changed industry, occupation, geographic residence, or labour force state, and more will do so in the future. Labour market institutions will play a large role in influencing the persistence of unemployment, and governments should avoid policies which inhibit structural change currently underway.

The potential for and the limitations of active labour market policies (ALMP) in dealing with the unemployment problem were examined by Michael Burda at a lunchtime talk in Sofia, organized by the 21 Century Foundation and CEPR, under the auspices of the Economic Policy Initiative. Michael Burda is a Professor of Economics at Humboldt Universität zu Berlin, and is a Research Fellow in CEPR’s Human Resources and Transition Economics programmes. His talk was based on CEPR Discussion Papers No. 1102 ‘The Impact of Active Labour Market Policies: A Closer Look at the Czech and Slovak Republics’ and No. 1302 ‘Active Labour Market Policies, Job Matching and the Czech Miracle’.

ALMPs are those policies which actively seek to integrate the unemployed into work rather than simply providing passive income support. Although under fire in many advanced welfare states, ALMPs deserve consideration as a substitute for passive income support in the CEE countries. ALMPs can be classified into several schemes. There is direct job creation, or ‘make-work’ which is often offered to the long-term unemployed as either a means of preserving good work habits or a work-test for benefit receipt. Another programme is that of direct government subsidies to employers to maintain staffing (wage subsidies), or grants for entrepreneurial start-ups. Third, re-training and re-integration programmes are offered to increase occupational and industrial mobility among the unemployed. Lastly, the efficiency of the job-matching process, including mobility grants, re-interview programmes, and efforts targeted to high unemployment groups, as well as resources devoted to staffing and equipping of labour offices.

ALMP programmes are criticized for a number of reasons. First, they are often associated with dead-weight losses to the extent that the individuals who benefit would have found jobs anyway. Second, they imply, almost by definition, displacement for some other group in the labour market. Third, programme participation may send the wrong signal to potential employers; studies in the US report negative effects on re-employment probability from participation in wage-subsidy programmes, as well as low take-up on the part of employers. Fourth, macroeconomic effects at the local labour market level may also undo the effects of ALMPs. For example, programmes which guarantee employment set a floor on workers' utility levels. Paradoxically, well-intentioned programmes could create upward pressure on wages and raise the natural rate of unemployment. Lastly, any ‘success’ of ALMP measures may reflect selection bias, because policy often favours the individuals most motivated to search actively, and with internal evaluation procedures often rewarding successful placement, participants may be chosen for their potential for success rather than need for treatment.

Despite sobering evidence from the West, ALMPs may have an important role to play given the extraordinary conditions of transformation. For example, the Czech Republic has proved to date an exception to the usual pattern, with unemployment remaining in the 3–4% range since 1990. With output growth now returning and unemployment declining in all Visegrád countries, it is clear that either the Czechs did something right, or had an enormous amount of good luck, or both. Even more puzzling is the fact that Slovakia, which shared institutions and policies with the Czech Republic for decades, experienced increases in unemployment resembling those in Bulgaria, Hungary, Poland and Romania. Informal evidence on the level and extent of ALMP implementation before and after the ‘velvet divorce’ suggests that these programmes made a difference in the Czech Republic and Slovakia.

How much difference did ALMPs make quantitatively? Using two panels of labour market stock and flow data from Czech and Slovak labour market districts, it is possible to detect a small, but statistically significant effect of ALMP expenditures on job finding. On the basis of these estimates, raising average Slovak district ALMP spending in 1993 to 1992 levels would have been associated with a 30% increase in monthly outflows out of unemployment into jobs, all other things equal. This effect is nevertheless modest, implying a 2% lower unemployment rate, ceteris paribus, or about half of the increase in Slovakia from 1992 to 1993. More detailed analysis of Czech data accounting for potential endogeneity of ALMP allocation reaches a similar conclusion. More significantly, cross-section staffing district data suggest that job-counselling staffing has an effect on outflows to jobs. In particular, a one percentage point increase in job-counselling staff is associated with about 0.2% more job outflows.

The economies of Central and Eastern Europe can learn from the mistakes of Western Europe, which chose in the 1970s to subsidize the status quo in ‘rust-belt’ industries and to pursue passive unemployment benefit systems with little incentive for adaptation to changing conditions, effectively leaving the unemployed alone to fend for themselves. Although ALMPs are almost certainty associated with deadweight losses, the Czech experience suggests that they can stimulate additional turnover among the unemployed. Fostering unemployment flows even at the cost of displacing other workers could be considered an objective in its own right, especially in transitional economies which are characterized by a stagnant unemployment pool. At the same time, it is important to stress that ALMPs are a flanking manoeuvre at best. The extreme distress currently experienced in Sweden should be a warning to all that ALMPs are not a substitute for sound, stable aggregate demand policies and, most importantly, an employment-friendly supply-side environment.