This project explores the economics of the term structure of interest rates, and how it relates to monetary policy. The questions addressed include: Can the Federal Reserve affect long-term interest rates? Why does Fed tightening not lower long-term interest rates? Does the Fed respond to interest rates, or do interest rates simply forecast Fed moves? How are Fed actions related to changes in interest rate risk premia? What are the macroeconomic risks that drive risk premia in the term structure? Are risks in the term structure primarily real or inflation? Has this changed over time? In their previous work, the investigators document even stronger predictability of bond expected excess returns than previously thought. They show strong, recession-related time-variation in bond risk premia. The investigators examine which kinds of interest rate shocks explain this fact, and find that covariance with a "level" shock to interest rates drives the risk premium. Inflation shocks and shocks to expected returns fail. The "level" shock can be interpreted as a monetary policy shock. The investigators also look at very high frequency data to isolate Fed policy shocks and find surprising output and price responses, and a surprisingly strong ability of the Fed to influence long-term interest rates. This work is extended in a number of ways. Short-term bonds as well as long-term bonds are included. The investigators have developed an underlying exactly identified affine term structure model, incidentally resolving the long standing puzzle whether such models exist that can capture term premia. This model is used to investigate a variety of reduced state variable representations. It is clear that bond data have measurement error, so the investigators develop a model that takes measurement error seriously. Over the longer term, the investigators plan to incorporate macroeconomic variables into the analysis and tie what they are learning about Fed policy from the high frequency identification to these term structure results.
