The treatment effect of a Swedish employment subsidy is estimated using exact covariate-matching and instrumental variables methods. Our estimates suggest that the programme had a positive treatment effect for the participants.We also show how non-parametric methods can be used to estimate the time profile of treatment effects as well as how to estimate the effect of entering the programme at different points in time in the unemployment spell. Our main results are derived using matching methods. However, as a sensitivity test, we apply instrumental variables difference-indifference methods. These estimates indicate that our matching results are robust. Url or DOI http://www.ifau.se/upload/pdf/se/2004/wp04-18.pdf Authors Anders Forslund Per Johansson Linus Lindqvist Country Sweden Publication Year 2004 Ranges Intervention Intervention Start Year 1998 Intervention End Year 2002 Evaluation Evaluation Start Year 1998 Evaluation End Year 2000 Policy field Employment incentives Private sector employment incentives Target group Labour market status Long term unemployed Unemployed (All cat.) Details Funding Source Other Outcome Variable Employment status Income/wages Data Source Administrative Evaluation Method DID PSM
