Abstract:
Randomized experiments provide policy relevant treatment effects if there areno spillovers between participants and nonparticipants. We show that thisassumption is violated for a Danish activation program for unemployed workers.Using a difference-in-difference model we show that the nonparticipantsin the experiment regions find jobs slower after the introduction of the activationprogram (relative to workers in other regions). We then estimate anequilibrium search model. This model shows that a large scale role out of theactivation program decreases welfare, while a standard partial microeconometriccost-benefit analysis would conclude the opposite.