Printer-friendly version

Counseling Welfare Recipients in Hauts de Seine, France

Location: Hauts de Seine, France
Sample: 17,666 individuals
Timeline:
2006 to 2008
AEA RCT Registration Number: 
Partners:

In an effort to increase employment rates, the French government has expanded “active labor market policies,” which provide counseling to help find employment for people with limited ties to the labor market. Researchers sought to test whether intensive job-counseling with a private provider is an effective means of increasing employment rates, even for a population with such a low attachment to the labor market. Results suggested that job-counseling increased employment rates among counseled individuals, and somewhat reduced their dependence on welfare. However, take-up of the service was low, and the cost of the program exceeded the savings generated from lower welfare payments.

Policy Issue 

In an effort to increase employment rates, many governments have "active labor market policies" which provide counseling to help find employment for people with limited ties to the labor market. This counseling approach has been rapidly expanded in France, and quite often the counseling is provided by private firms. Private counseling firms receive significant compensation from the government, on the presumption that the reduction in unemployment will reduce government costs in other areas such as welfare and unemployment benefits. However, there is little evidence on whether the benefits of providing job counseling actually outweigh the costs.

Context of the Evaluation 

Job counseling was provided in a French suburban area near Paris, where unemployment is extremely high. At that time there were around 17,000 individuals in the area who had been on welfare for at least 2 years, and around half of those had been on welfare for more than 4 years. About half of them were below 40 years old, and only about 20 percent of them were holding a job before the treatment started.

Job counselor conducting a lesson in France
Details of the Intervention 

Researchers sought to test whether intensive job-counseling with a private provider is an effective means of increasing employment rates even for a population with such a low attachment to the labor market. Job-seekers in the "treatment group" had their names given to a private counseling firm, who would contact them to offer their services. The job counseling provided by this firm consisted of regular meetings with an individual advisor who would work with people on their confidence, advise on any social problems they might be having, and help them look for work or vocational trainings to increase their skills. These meetings took place at the offices of the private firm. From the sample population, researchers randomly selected 13,236 individuals to receive this job counseling, while another 4,430 served as the comparison group.

Data was collected from administrative records for the 17,666 people in the sample group, which provides monthly information on their employment status and whether or not they are receiving government benefits.

Results and Policy Lessons 

Take-up of the intervention was extremely low, at around 15 percent. Because of this low take-up, it is difficult to precisely estimate the impact it had on beneficiaries. The results suggest that job-counseling increased employment rates among treated individuals, and reduced somehow their dependence on welfare . However, because of the high costs of providing job counseling, the program clearly cost more to the welfare authority than they saved in lower welfare payments: in spite of low take up, this result is very significant. Although this is true for the policy as a whole, there is significant heterogeneity. In particular, the program is more efficient and more cost-effective for people who have spent less time on welfare.

Crpon, Bruno, Marc Gurgand, Thierry Kamionka, and Laurent Lequien. "Is Counseling Welfare Recipients Cost-Effective? Lessons from a Random Experiment." Working Paper, Paris School of Economics, 2011.