终究并非飞跃式开局？一则评论

In a paper published in the Journal of Political Economy, Carneiro, Løken, and Salvanes (2015) reported large positive effects on offspring outcomes of a reform they described as introducing paid maternity leave in Norway. Causal identification rested on a discontinuity implying that only mothers giving birth after a specific cutoff date were entitled to paid leave. We show that the analysis relied on an incorrect description of the reform. The reform did not introduce paid maternity leave, but extended it by 5–6 weeks. The postulated discontinuity never existed as treatment and control groups had the same maternity leave conditions.