ABSTRACT
The statistical examination of the links between labor market institutions and unemployment across developed countries is a relatively recent development, pioneered by Stephen Nickell and his colleagues (Layard et al. 1991; Nickell and Bell 1995; Nickell and Layard 1997). Chapter 5 by Blanchard and Wolfers in this volume, previously published as an NBER working paper in 1999 and in the Economic Journal in 2000, has been a hugely influential contribution in what might be termed the second generation of research on this question. It focuses on the interaction of macroeconomic shocks and institutions, expands the time period, and adds time-varying measures of institutions. But most importantly it is cautious—persistently checking the initial results with various robustness tests and questioning the meaningfulness of the regression results given the quality and potential endogeneity of the key institutional measures. I will begin with a brief overview of what may be termed the “first generation” studies, followed by short comments on the Blanchard-Wolfers contribution, and then will turn to more recent work that challenges the mainstream view that cross-country regression research confirms a central role for labor market institutions in the explanation for Europe’s high unemployment in the 1980s and 1990s.
