Give it Another Try: What are the Effects of a Job Creation Scheme Especially Designed for Hard-to-Place Workers?

Job creation schemes (JCSs) have for a long time been considered to function as stepping stone for unemployed workers into the labour market. However, previous research has shown that public job creation schemes could at best be effective for persons with little or no employment chances. This paper contributes to this discussion by studying the effect of a very large and innovative JCS which employs an extraordinary selection mechanism to identify this group. Relying on a combination of administrative data and survey data, we employ radius-matching with regression adjustment to estimate treatment effects. Our results indicate that the effects are still negative but weaker than the ones reported in previous studies. We argue that this is due to a principal-agent problem at the very last step of the selection process which undermines the intention of the programme. Furthermore, we point to effect heterogeneity with regard to programme duration which suggests that the results are more optimistic for shorter contracts. Finally, we show that the results are sensitive to the inclusion of survey variables, even if high quality administrative data are available.