Room : P02

We use a (donut) regression discontinuity design to estimate the impact of a generous temporary hiring subsidy targeted at unemployed youth during the early recovery from the Great Recession in Belgium. The subsidy raises the job finding rate by about 10 percentage points within one year after entry in unemployment. Five years later, eligible high-school graduates have accumulated 2.7 quarters more employment, but for high-school drop-outs no persistent gains emerge. However, these long-run positive employment effects are observed in the private sector only: Overall employment remains unaffected. Moreover, close to the country border the hiring subsidy is a complete waste.