We analyse the duration of unemployment spells in Poland using data from the Polish Labour Force Survey of August 1994. The effects on the duration of unemployment of important socio-economic and demographic characteristics are explored besides the impacts of the unemployment benefit system and training schemes. Finally, we investigate whether prior unemployment influences one's chances to find a job. Entitlements to unemployment benefits prolong unemployment spell durations significantly. This effect is roughly of the same magnitude under the two benefit regimes that existed between 1990 and 1994, although the generosity of the unemployment benefit system has been reduced drastically in 1992. The results give credence to the view that the unlimited entitlement period of the old regime was not the main culprit for the widespread incidence of long-term unemployment. Training programmes organised by labour offices should not be regarded as a panacea for the problems of the long-term unemployed. The results suggest that active labour market policies should perhaps be seen more as a tool for social rather than economic policy. People with previous unemployment spells must expect to stay unemployed far longer than people who become unemployed for the first time. On the other hand, controlling for unobserved individual heterogeneity, we find that the probability of finding a job increases, especially for men, with the duration of unemployment.
unemployment duration incentives training Poland transition