We would like to thank an anonymous referee and the editor P. Evans, as well as J. Campbell, F. Collard, M. Dupaigne, M. Eichenbaum, J. Galí, L. Gambetti, S. Grégoir, A. Kurman, J. Matheron, F. Pelgrin, L. Phaneuf, F. Portier, H. Uhlig, R. Vigfusson, and E. Wasmer for valuable remarks. This paper has benefited from helpful discussions during presentations at CIRANO Workshop on Structural VARs, UQAM seminar, Macroeconomic Workshop (Aix/Marseille), Laser Seminar (Montpellier), HEC-Lausanne seminar, HEC-Montréal seminar, Université Laval seminar, and AMeN workshop (Barcelona). The traditional disclaimer applies. The views expressed herein are those of the authors and not necessary those of the Banque de France.
The Response of Hours to a Technology Shock: A Two-Step Structural VAR Approach
Article first published online: 16 JUL 2009
© 2009 The Ohio State University No claim to original US government works
Journal of Money, Credit and Banking
Volume 41, Issue 5, pages 987–1013, August 2009
How to Cite
FÈVE, P. and GUAY, A. (2009), The Response of Hours to a Technology Shock: A Two-Step Structural VAR Approach. Journal of Money, Credit and Banking, 41: 987–1013. doi: 10.1111/j.1538-4616.2009.00241.x
- Issue published online: 16 JUL 2009
- Article first published online: 16 JUL 2009
- Received February 19, 2008; and accepted in revised form January 2, 2009.
- long-run restriction;
- technology shocks;
- hours worked
The response of hours to a technology shock is a controversial issue in macroeconomics. Part of the difficulty lies in that the estimated response is sensitive to the specification of hours in structural vector autoregressions (SVARs). This paper uses a simple two-step approach to consistently estimate the response of hours. The first step considers a SVAR model with a relevant stationary variable, but excluding hours. Given a consistent estimate of technology shocks in the first step, the response of hours to this shock is estimated in a second step. Simulation experiments from an estimated dynamic stochastic general equilibrium (DSGE) model show that this approach outperforms standard SVARs. When applied to U.S. data, the two-step approach predicts a short-run decrease followed by a hump-shaped positive response. This result is robust to other specifications and data.