brown bag Travail et Économie Publique

Du 09/02/2023 de 12:30 à 13:30

PSE- 48 boulevard Jourdan, 74014 Paris, salle R1-09

HEIM Arthur (PSE)

Can intensive welfare-to-work programs help single parents on long-term welfare ? Experimental evidence from France

Since the beginning of the 1990s, welfare States in OECD countries have shifted towards active labor market policies (ALMPs) to (re)engage a part of the inactive population in employment and to promote their economic autonomy. Single-parent households have been an important target group for ALMPs, especially through welfare-to-work programs. This shift towards activation has been analysed through the lens of the social investment paradigm, which promotes employment as the main protection against poverty. However, disappointing poverty trends question the effectiveness of ALMPs in mitigating social risks for vulnerable target groups. In this paper, I evaluate the effects of an intensive welfare-to-work program targeting single-parents - 96% of whom are women - on welfare for more than two years to foster employment and overall participation in society. The program consists of a year-long intensive social support, including face-to-face and group sessions with highly-trained social workers, several times per week. It costs approximately €2,900 per participant and has been rolled-out each year from 2018 to 2022 in France using a randomized encouragement design to select participants and controls. Using the random encouragement as an instrument and a panel of monthly administrative records for the first three cohorts (± 1200 households) up to 36 months since randomization, I find no average effect of the program on the probability of being employed or receiving welfare. However, the program increases the average amount of aids received from the Family Allowance Funds (CAF) and reduces labour income for the upper-tail of the income distribution. Participants are also more likely to keep their children's custody and receive public-funded alimonies, explaining most of the increased welfare payments. Heterogeneity analysis shows evidence of substitution bias, where families who were initially registered at the unemployment agency undergo negative effects from the program, while we find evidence of positive or null effects for those unregistered. These results are among the few large-scale randomized experiments in France that estimate the causal effects of anti-poverty measures, and cast doubt on the efficacy of welfare-to-work programs for single-parent households.