Identication of Technology Shocks Using Misspecied VARs

Studies that seek to evaluate the e¤ects of technology shocks often employ structural VARs identi…ed with long-run restrictions. In the presence of a mismatch between the number of lags the data-generating process involves and the number of lags included in the VAR, estimates based on long-run restrictions can be largely biased. In this paper, I o¤er an alternative method that substantially reduces this bias associated with the standard longrun procedure. I show how the method can be employed to reduce misspeci…cation bias in a large class of maximum-share-type identi…cation problems. I assess the performance of the proposed approach using Monte Carlo simulations and demonstrate that it outperforms a range of methods including the standard long-run approach under most commonly adopted parametrizations of the real business cycle and new Keynesian models. Application of the procedure to the post-war U.S. data reveals that per-capita hours exhibit a positive humpshaped impulse response pro…le in the face of a technology shock. Keywords: Technology Shocks, Long-run Restrictions, Adjusted Maximum Share Identi…cation JEL classi…cation: E2, E3 University of Colorado at Boulder, Department of Economics, 256 UCB, Boulder, CO, 80309, USA, tel: (303) 492 2585, email:[email protected]. I would like to thank Mario Crucini, two anonymous reviewers, Elmar Mertens, Carlos Martins-Filho, Murat Iyigun and seminar and session participants at the University of ColoradoBoulder, 2011 meetings of the Society for Computational Economics, and Fall 2012 Midwest Macroeconomics Conference for useful comments and suggestions.