This paper studies the relevance of social interactions among the unemployed. Identification is based on a salient and selective extension of the potential duration of unemployment benefits. If social interactions are important, this policy change affects entitled individuals not only directly, but also indirectly by altering the duration of unemployment in the reference group. Moreover, this spillover effect of the policy should also be observed in the non-entitled group. Results indicate that there are strong indirect effects on the entitled, strong positive spillovers on the non-entitled, and that social interactions are about as important as the direct effects of the policy change.
social interactions social multiplier unemployment quasi-experiment.