Introduction

The transition from school to work is of primary interest to researchers in labour economics. Because of the high rate of youth unemployment in Europe, researchers now tend to evaluate the effects of unemployment on early labour market experience. Young people tend to be unemployed for shorter periods than adults but the frequency of their unemployment is higher. Because young people encounter successive spells of short unemployment during their first years in the labour market, it is of interest to measure the impact of this recurrence longitudinally. Specifically, one might ask if unemployment causes unemployment or if selection effects cause this pattern of recurrence (Roed et al, 1999). Are young people unemployed because they have difficulty escaping the unemployment spiral, because of specific individual characteristics, or because of the macroeconomic environment?

Beyond this lies the question of how young people accumulate human capital and how they manage their returns on investment in education and training (Bratberg and Nilsen, 2000). Level of education may also affect the impact of recurring unemployment on patterns of labour market mobility. An individual with a higher level of education is likely to have better job prospects and be less affected by jobless periods. However, specificities of national labour markets, educational systems and the economic context may have an impact on the role of unemployment during the transition from school to work (Raffe, 2001). Thus, youth unemployment policies may differ according to the origin of the persistence of unemployment (Fougère et al, 2000).

This chapter is devoted to the analysis of recurring unemployment among young people who have experienced at least one period of unemployment. The empirical findings are generated from several national surveys that were conducted using identical questionnaires, and the analysis was based on comparisons of France, Germany and Sweden. A Poisson model has been used for count data in order to model the process of unemployment recurrence. Two sets of factors were included in the model: micro-level data such as age, gender and education, and macro-level data such as youth unemployment rate, adult unemployment rate and the proportion of long-term unemployed in the country. In addition, unobserved heterogeneity in the count model is controlled. Because young people can be selected into recurring unemployment by unobserved characteristics, state dependence is controlled for when analysing this process (Van den Berg and Van Ours, 1996).