Do monitoring and counselling boost the job finding rates of the long-term unemployed ?

In July 2004, the Belgian government launched a new monitoring and counselling program targeted at the long-term unemployed. In this new program, people over 13 months unemployment are warned that their job-search efforts will be monitored 8 months later. At the same time, they are invited to participate actively in the support (job search, training) programs provided for by the Regional Employment Agencies, whose supply was expanded. This reform has been implemented gradually : it focused the unemployed under 30 years in 2004, then the 30-40 year-old unemployed in 2005, and finally the 40-50 year-old unemployed in 2006. This paper aims at evaluate the effect, in terms of transition from unemployment to employment and/or training, of this new program for the 30-40 years old in Wallonia. Using the fact that the reform was implemented gradually, the effect of the program is estimated by difference-in-differences, based on (single exit and two exits) discrete duration models. The results suggest that the program has had a very large relative effect, although limited in absolute terms, regarding transition to training for almost all the treated unemployed. They also show that the program has had a substantial relative effect, although limited in absolute terms as well, with regard to transition to employment for the majority of the treated unemployed.