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...

In this paper, we specify and estimate a structurally dependent competing risks model for the transitions out of unemployment into either new job or recall. The recall probability is allowed to affect the search intensity for new jobs.

This paper compares the effectiveness of public and private providers of employment services. Reporting from a randomized field experiment conducted in Denmark we assess empirically the case for contracting out employment services for a well-defined group of highly educated job-seekers...