This paper estimates the impact of different categories of employment policy interventions on subsequent outcomes for jobseekers. We generate a range of estimates to help us distinguish programme effects from selection effects. We also examine the robustness of our findings for a range of sub-populations.
Referrals to vacancies and job subsidies appear to be most effective in reducing the number of weeks of assistance or contact that jobseekers subsequently have with the public employment agency. The favourable estimated impact of subsidies is not evident until at least a year after the subsidy starts. There are only small differences in the estimated effectiveness across different ethnic groups. Interventions appear to be more effective for males than for females, and to a lesser extent more effective for younger than for older jobseekers. We find evidence to suggest that programme effectiveness is counter-cyclical.