Is Counseling Welfare Recipients Cost-Effective ? Lessons from a Random Experiment

Job-search counseling is a potentially desirable labor market policy because it reduces market frictions, but it is strongly work-intensive as it requires repeated individual contact between job-seeker and case worker. Although it has become widely used, little is known about its cost-efficiency. This paper uses an experiment where individuals who have been on welfare for more than two years in a French district were randomly allocated to a counseling firm. We show that the policy causal impact is to increase employment and decrease the amounts of welfare transfers paid to the beneficiaries. However, the effects are small relative to the cost charged by the providing firm. Therefore, the net public cost, accounting for gains in welfare transfer payment, remains larger than reason- able social values that can be attached to having a former welfare recipient on a job. Although this is true for the policy as a whole, as implemented in this experiment, there is significant heterogeneity. In particular, it is more efficient and more cost-effective on a population of limited seniority on welfare