The impact of temporary employment subsidies on the exit rate from unemployment to regular employment: the case of Belgian older workers

The labour market participation rate of older Belgian workers is one of the lowest of the OECD. Early retirement, passive labour market measures and the wage bargaining system which makes older (white-collar) workers relatively more expensive are features that discourage employment of older workers. At the same time, population ageing is causing an increase in the proportion of older individuals in relation to the working population. Because these two factors together can generate important financial sustainability problems in the social security system, policy makers have tried to increase labour participation of older workers. First, positive incentives to delay the normal retirement age have been introduced. Second, active labour market measures aimed at keeping workers in the labour force have been extended to older workers. Temporary employment subsides have been used in Belgium for a long time to increase the employment rate of disadvantaged groups in the labour market such as young, long-term and low-qualified unemployed people. It is only recently that these measures have been extended to target specifically older workers. The ‘Activa plan’ (2002) allows employers to hire unemployed workers at a lower cost during a certain period. The amount of the reduction and its length depend on the characteristics of the unemployed. Compared to other target groups, this measure is particularly attractive for hiring older unemployed workers. The aim of our study is to evaluate the impact of this measure on the transition from unemployment to employment of older workers. Our data comes from the Belgian National Unemployment Office and covers the population of unemployment insurance beneficiaries during the period 2001-2010. We select the inflow of unemployed workers during the year 2001 and follow these individuals during the next nine years. The usual individual characteristics are available and because part of the “Activa” employment subsidy is paid by the unemployment office, the data tells us if someone has been hired using this measure and the length of his/her subsidised employment. Two methodological approaches are tested: a regression discontinuity and an event history analysis (multivariate mixed proportional hazard model).Preliminary results show that the impact of the “Activa” measure is quite different according to the target group. In particular, it seems to have a very limited impact on the reemployment probability of older workers.