Non-profit provision of job training and mediation services; an empirical analysis using contract data of job training service providers

This paper analyses the relative performance and selection behaviour of not-for-profit (NFP) job training service providers, using contract data from the Dutch social benefit administration. Our analysis takes full account of selection effects, both ex ante (before the contracting process) as well as ex post (at the start of the program). First, for each cohort type of unemployed clients, cohorts that are contracted are ex ante equivalent for providers that are procured. Thus, within cohort type variation in performance outcomes suffices to obtain consistent estimates of performance differentials. Second, ex post selection of clients by providers, at the start of programs, is measured explicitly in our data. Our estimation results show that FPs are more active in selecting clients, both by sending back more of them, and indirectly, by encouraging clients to start a program, so as to receive additional (fixed) payments by the social benefit administration (per client at the start of a program). Regarding the estimation results for the job placement rates, we find NFP job training service providers only to outperform FPs slightly in the durability of job contracts. This effect is however too small to lead to overall better placement rates.