Active Labour Market Policies for the Long-Term Unemployed: New Evidence from Causal Machine Learning

We investigate the effectiveness of three different job-search and training programmes for German long-term unemployed persons. On basis an extensive administrative data set, we evaluated effects those on various levels aggregation using Causal Machine Learning. found participants to benefit from investigated with placement services be most effective. Effects are realised quickly long-lasting any programme. While rather homogenous men, differential women in characteristics. Women particular when local labour market conditions improve. Regarding allocation mechanism programmes, observed as effective a random allocation. Therefore, propose data-driven rules respective that would improve status-quo.