Difference-in-Differences

Policymakers typically use active labor market policies, such as training and job-search assistance, to help the unemployed find work. Nonetheless, these are often costly and have shown modest effects. In this research, I assess the employment effect of a low-cost and straightforward intervention: subsidized public transport for cash-constrained jobseekers. In particular, I exploit a natural experiment in Catalonia in 2012 that reduced transit costs for unemployment assistance recipients. Using three complementary empirical approaches (difference-in-differences, the synthetic control method, and the synthetic difference-in-differences method), I find that the transport subsidy offered in Catalonia brought meaningful employment gains concentrated on younger assistance recipients. These gains ranged from 18% to 25% of their estimated counterfactual outcome, three to twenty-four months after entering unemployment. Finally, I also find suggestive evidence that these employment gains did not come at the expense of lower earnings.

This paper evaluates the impact on the transition to work of a policy reform in Belgium that restricted the access to a specific unemployment insurance scheme for young labor market entrants. This scheme entitles youths with no or little labor market experience to unemployment benefits after a waiting period of one year. As of 2015, the Belgian government unexpectedly scrapped benefit eligibility for youths who start the waiting period at the age of 24 or older. The reform implied a change from an inclining to a flat rate (zero-level) benefit profile. We use a difference-in-differences approach to identify the causal impact of this reform on fresh university graduates. Our main finding is that this reform only increases the transition to very short-lived jobs.