A Markov-Switching Model of Business Cycle Dynamics with a Post-Recession “Bounce-Back” Effect

This paper presents a nonlinear model of U.S. GDP growth dynamics that allows for a post-recession “bounce-back” effect in the level of GDP. While a number of studies have attempted to capture such an effect using ad hoc recession-based dummy variable methods, we endogenously estimate this business cycle asymmetry using an extended version of Hamilton's (1989) Markov-switching model. Like Hamilton, we find model regimes that correspond closely to NBER-dated recession and expansions. We also find a large “bounce-back” effect that, according to our Monte Carlo analysis, is statistically significant and implies a relatively small permanent effect of recessions. ∗ We would like to thank Mrinalini Lhila for providing research assistance. Responsibility for all errors is our own. Morley acknowledges support from the Weidenbaum Center on the Economy, Government, and Public Policy. The views expressed in this paper should not be interpreted as those of the Weidenbaum Center, the Federal Reserve Bank of St. Louis, or the Federal Reserve System. 1