Federal Reserve Credibility and the Term Structure

Michigan State University, East Lansing MI

Investigators

Abstract

The award funds research on the relationship between monetary policy and the term structure of interest rates (how interest rates with the length/term of a loan). Understanding this relationship is important for central bankers, who need to understand how their actions affect long-term interest rates and therefore the economy as a whole. Investors also want to understand this relationship, because it is important for making the best possible investment decisions. Previous research has demonstrated that the credibility of a central bank is an important factor in this relationship. If a central bank has no credibility, then its current decisions do not affect investors' expectations about the future and likely future interest rates. Alternative, a bank might have "full commitment", where it chooses a monetary policy rule and does not make any changes in that rule over long time periods. This project defines commitment possibilities between these two options, in which a central bank has the ability to commit to future plans but may sometimes revise these plans. The research team will use data and a model of the US economy to identify past periods where the Federal Reserve made changes in previous commitments and will then explore how these episodes affected term structures. The results of the project will give policymakers and investors new insight into the pros and cons of central bank commitments. The project considers a general framework of optimal monetary policy where credibility is defined as the ability to commit to previously announced plans; this nests the commonly used cases in the literature of full commitment and no commitment. The definition of credibility is based on the theoretical literature on time-inconsistency and estimated within a fully optimizing model. The estimation framework uses recent Bayesian econometric techniques in the context of regime-switching models which allows the identification of historical episodes where Fed re-optimizations were more likely to have occurred. The project will evaluate the quantitative effects of re-optimization episodes on the term premium.