Randomized experiments provide policy relevant treatment effects if there are no spillovers between participants and nonparticipants. We show that this assumption is violated for a Danish activation program for unemployed workers. Using a difference-in-difference model we show that the...

Workers acquire skills through formal schooling, through training provided by governments, and through training provided by firms. This chapter reviews, synthesizes, and augments the literature on the last of these, which has languished in recent years despite the sizable contribution of firm...