Do Flexible Labor Markets Indeed Reduce Unemployment? A Robustness Check

Abstract

Nickell et al. (2005) have frequently been cited as empirical evidence that labor market rigidities cause high unemployment. We find that their model is not robust. Leaving their database unchanged and changing three details in their estimation procedure, it turns out that several policy-relevant coefficients change sign or significance. We conclude that their claim from Non Accelerating Inflation Rate of Unemployment (NAIRU) theory that labor market rigidities cause unemployment is rather shaky. There is a remarkable discrepancy between weak empirical results and sweeping conclusions by policy practitioners with respect to the call for deregulation of labor markets.

Keywords:

• NAIRU
• labor market rigidities
• unemployment

ACKNOWLEDGEMENTS

This article is based on Robert Vergeer's PhD thesis, defended at TU Delft in August 2010 and sponsored by N.W.O., the Netherlands Council for Scientific Research. We thank the members of the PhD committee (Cees van Beers, John Groenewegen, Joan Muysken, Esther-Mirjam Sent, Ton van Schaik, and Servaas Storm) for their critical scrutiny. Earlier versions of this article benefited from comments by Ro Naastepad, Arthur van Soest, Federico Lucidi, Andreas Pyka, and Michel Dumont as well as from suggestions by two anonymous referees. The usual caveats apply.

Notes

According to Google Scholar, the article was mentioned in 590 sources between 2005 and June 2011.

We thank Luca Nunziata for making the data available and for his written comments.

This definition is not stated in the original article. We found it through trial and error. First, we calculated the interaction variables using various definitions. If the World mean is used to de-mean both variables interacted, a low correlation with the interaction variable used by Nickell et al. (2005) is obtained. If the country-specific means are used, the correlation with the original demeaned variables is above 99% for all interaction terms except union density interacted with coordination. For the latter, the correlation is above 90%. This may be because the union density variable provided in the original database is rounded. Furthermore, comparing two regressions using the Nickell et al. (2005) specification and altering only the two highly correlated interaction terms, did not alter the results of the regression.