THE IMPACT OF ACTIVE LABOUR MARKET PROGRAMMES ON THE DURATION OF UNEMPLOYMENT IN SWITZERLAND* Rafael Lalive, Jan C. van Ours and Josef Zweim € uller This article evaluates the effects of Swiss active labour market programmes on the job chances of unemployed workers. The main innovation is a comparison of two important dynamic evaluation estimators: the ÔmatchingÕ estimator and the Ôtiming-of-eventsÕ estimator. We find that both estim- ators generate different treatment effects. According to the matching estimator temporary subsidised jobs shorten unemployment duration whereas training programmes and employment programmes do not. In contrast, the timing-of-events estimator suggests that none of the Swiss active labour market programmes shortens unemployment duration. The aim of the present article is to study the impact of active labour market policies (ALMPs) on the duration of unemployment in Switzerland. The new Swiss ALMPs reflect the increasing consensus among policy makers that actively assisting the unemployed in job search is preferable to simply providing them with passive income support. The danger is, so the argument goes, that reliance on passive income support may reduce work incentives and job-search activities and therefore increase the risk of long-term unemployment. ALMPs are seen by many as the key to minimise these risks. The question how participation in ALMP-measures affects labour market histories of individuals has been the subject of substantial debate. The main problem is that labour market outcomes for participants may be systematically different from non-participants for reasons that are unobservable to the researcher – the selection problem; see Heckman et al. (1999). In Switzerland, like in most European countries but unlike in the US, randomised social experiments are uncommon, so one has to deal with non- experimental data. In theory, several methods can be used to estimate the treatment effects of ALMPs. Each of these methods deals with the selection problem under different assumptions. In the case of unemployment duration as a variable of interest two methods are particularly useful. The first one is the method of ÔmatchingÕ, the second one is the Ôtiming-of-eventsÕ method. The main innovation of the present article is a direct comparison between the tim- ing-of-events approach and the matching approach in estimating the effect of ALMPs on the rate by which unemployed individuals find regular jobs. The method of matching is based on the conditional independence assumption. If many variables that influence both labour market outcomes and the selection process are observed, potential out- comes and selection are independent, conditional on the observables. The identifying assumption is that, after accounting for many observable variables (including individu- al’s past labour market performance), no unobserved heterogeneity correlated with * We thank without implicating Wolfgang Franz, Michael Gerfin, Lorenz Goette, Michael Lechner, Patrick Puhani, Frank Reize, Viktor Steiner, Gerard van den Berg, Rudolf Winter-Ebmer and two editors as well as three anonymous referees for comments on earlier versions of this article. We also thank M. Curti and J. Gast at the Swiss State Secretariat for Economic Affairs for help concerning the data and information on insti- tutional details, and Jacek Micuta at the centrale de compensation for access to social security data. The Economic Journal, 118 (January), 235–257. Ó The Author(s). Journal compilation Ó Royal Economic Society 2008. Published by Blackwell Publishing, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA. [ 235 ]