Markov-Switching Models with Evolving Regime-Specific Parameters: Are Post-War Booms or Recessions All Alike?

Abstract

In this paper, we relax the assumption of constant regime-specific mean growth rates in Hamilton’s (1989) two-state Markov-switching model of the business cycle. We introduce a random walk hierarchical prior for each regime-specific mean growth rate and impose a cointegrating relationship between the mean growth rates in recessionary and expansionary periods. By applying the proposed model to postwar U.S. real GDP growth (1947:Q4-2011:Q3), we uncover the evolving nature of the regime-specific mean growth rates of real output in the U.S. business cycle. Additional features of the postwar U.S. business cycle that we uncover include: i) a steady decline in the long-run mean growth rate of real output over the postwar sample and ii) an asymmetric error-correction mechanism when the economy deviates from its long-run equilibrium.

Keywords: Business Cycle, Evolving Regime-Specific Parameters, Hierarchical Prior, Markov Switching, Error-Correction Dynamics, MCMC, State-Space Model

JEL Classification: C11, C22, C51, E32