Can Long-Run Restrictions Identify Technology Shocks?

Galí's innovative approach of imposing long-run restrictions on a vector autoregression (VAR) to identify the effects of a technology shock has become widely utilized. In this paper, we investigate its reliability through Monte Carlo simulations using calibrated business cycle models. We find it encouraging that the impulse responses derived from applying the Galí methodology to the artificial data generally have the same sign and qualitative pattern as the true responses. However, we find considerable estimation uncertainty about the quantitative impact of a technology shock on macroeconomic variables, and little precision in estimating the contribution of technology shocks to business cycle fluctuations. More generally, our analysis emphasizes that the conditions under which the methodology performs well appear considerably more restrictive than implied by the key identifying assumption, and depend on model structure, the nature of the underlying shocks, and variable selection in the VAR. This cautions against interpreting responses derived from this approach as model-independent stylized facts