Monetary Policy and Term Structure of Interest Rate in A Continuous-Time DSGE Model: Theory and Estimation

Description

In this project, we first develop a continuous-time New Keynesian, dynamic stochasticgeneral equilibrium (DSGE) model with regime-switching monetary policy. Compared toexisting DSGE models, our model provides a rich and yet tractable characterization ofthe term structure of interest rates using continuous-time techniques. We then providea rigorous estimation of the model based on the Treasury yield curve using BayesianMCMC methods. Our project contributes to both the macroeconomic literature onmonetary policy and the finance literature on term structure.The New Keynesian DSGE models have become the dominant approach used by bothacademics and central bankers around the world for monetary policy analysis (see, e.g.,Gali and Gertler (2007)). Under the sticky price equilibrium of these models, monetarypolicy is not neutral and has important impacts on the real activities of the economy anddirect implications for the term structure. However, the existing macroeconomicliterature on DSGE models has mainly focused on the real activities and ignored theterm structure implications. Since the yield curve contains market expectations forfuture economic activities, we can better identify monetary policy by incorporating termstructure data in the estimation of DSGE models. The term structure also provides analternative perspective to examine potential shortcomings of these models: If the DSGEmodels make counter factual predictions on term structure dynamics, then one should becareful in using them for policy analysis.While traditional term structure models in finance have relied on latent variables toexplain bond yields, the fast growing macro term structure literature (e.g., Ang andPiazzesi (2003)) typically introduces macro variables into term structure models usingthe no-arbitrage Taylor (1993) rule approach. Recently, Li, Li, and Yu (2010) documentregime-switching Taylor rules and stable macro variables under the aggressive inflationcontrol regime for the US economy. Without a well specified DSGE model, however, it isnot clear that the exogenously specified pricing kernels in these “reduced-form” modelsare consistent with general equilibrium. It is also difficult to identify any causal relationsbetween monetary policy and the volatility of the macro variables despite the apparentcorrelations between the two. Our regime-switching DSGE model captures importantnew empirical facts documented in Li, Li, and Yu (2010). By endogenizing the dynamicsof the macro variables, it allows us to answer fundamental questions such as whethermonetary policy has contributed to the “Great Moderation.”