Abstract. Vector autoregressions with Markov-switching parameters (MS-VARs) offer substantial gains in data fit over VARs with constant parameters. However, Bayesian inference for MS-VARs has remained challenging, impeding their uptake for empirical applications. We show that Sequential Monte Carlo (SMC) estimators can accurately estimate MS-VAR posteriors. Relative to multi-step, model-specific MCMC routines, SMC has the advantages of generality, parallelizability, and freedom from reliance on particular analytical relationships between prior and likelihood. We use SMC’s flexibility to demonstrate that model selection among MS-VARs can be highly sensitive to the choice of prior.
Abstract. Debates among policy makers about the appropriate response of fiscal policy to the Great Recession centered on the size of the fiscal multiplier, defined as the number of dollars that output increases in response to a dollar of fiscal stimulus.Mostly using the structure of micro-founded Dynamic Stochastic General Equilibrium models, macroeconomists have argued that fiscal multipliers may vary over time and be particularly large in liquidity traps or during recessions. I extend existing techniques for the Bayesian estimation of vector autoregressions with Markov-switching in selected coefficients to empirically investigate both the extent of time-variation in fiscal multipliers and what factors cause the variation. In contradiction to recent results in the literature, my estimates suggest that the value of the government spending multiplier is likely smaller in recessions than in expansions, while tax cuts have a greater effect in recessions than in expansions. I find little evidence that regime change in monetary policy rules and fiscal policy rules have caused time variation in the value of the fiscal multiplier.