Is less more? A look at nonparticipation in Swiss active labour market programmes

This paper addresses the question of how the evaluation results for the effectiveness of active labour market policies are sensitive to the definition and interpretation of the "nonparticipation" state. Swiss active labour market programmes are re-evaluated by means of matching estimation with two different ways to handle nonparticipation. The first definition takes into account persons in the sample with no programme at all, and it leads to negative effects of almost all programmes compared to "nonparticipation". Another definition of nonparticipation is based on a "risk set" of persons who have not participated in a programme yet and search more intensely for a job instead of participating in a programme at that point of time. Nonparticipation as defined in this way loses superiority compared to other programmes. The two definitions lead to reversed findings in terms of employment. Effectiveness in this paper is furthermore measured in terms of a couple of outcomes that have not been considered so far, in order to give a clearer picture of the effects.