The control function in the semiparametric selection model is zero at infinity. This paper proposes additional restrictions of the same type and shows how to use them to test assumed exclusion restrictions necessary for root N estimation of the model. The test is based on the estimated control...

Traditional apprenticeships based on private arrangements are widespread in developing countries. Public interventions have attempted to address failures in the apprenticeship markets to expand access or improve training quality. Subsidized dual apprenticeships have the potential to address...

Active labor market policies are massively used with the objective being to improve labor market outcomes of individuals out of work. Many observational evaluation studies have been published. In this review, we critically assess policy effectiveness. We emphasize insights from recent randomized...

We use longitudinal individual wage, hours, and employment data to investigate the effect of the February 1, 1982 mandatory reduction of weekly working hours in France. Just after François Mitterrand's election in May 1981, the government decided to increase the minimum wage by 5%. Then, as...

The randomized trial literature has helped to renew the fields of microeconometric policy evaluation by emphasizing identification issues raised by endogenous program participation. Measurement and attrition issues have perhaps received less attention. This paper analyzes the dramatic impact of...

We propose a novel selectivity correction procedure to deal with survey attrition, at the crossroads of the "Heckit" and of the bounding approach of Lee (2009). As a substitute for the instrument needed in sample selectivity correction models, we use information on the number of attempts that...

Contracting out public services to private firms has ambiguous effects when quality is imperfectly observable. Using a randomized experiment over a national sample in France, we compare the efficiency of the public employment service (PES) vs. private providers in delivering very similar...

In this paper we discuss how the "encouragement design" used in randomized controlled trials can be extended to a setting with two treatments and one control group. Conditions to interpret the Two-Stage Least Squares (TSLS) estimates causally are stronger than in the case with only one treatment...

We evaluate an experimental program in which the French public employment service anonymized resumes for firms that were hiring. Firms were free to participate or not; participating firms were then randomly assigned to receive either anonymous resumes or name-bearing ones. We find that...

When treatments may occur at different points in time, most evaluation methods assume - implicitly or explicitly - that all the information used by subjects about the occurrence of a future treatment is available to the researcher. This is often called the "no anticipation" assumption. In...