A dynamic factor analysis of the response of U. S. interest rates to news

This paper uses a dynamic factor model recently studied by Forni, Hallin, Lippi and Reichlin (2000) and Forni, Giannone, Lippi and Reichlin (2004) to analyze the response of 21 U.S. interest rates to news. Using daily data, we find that the news that affects interest rates daily can be summarized by two common factors. This finding is robust to both the sample period and time aggregation. Each rate has an important idiosyncratic component; however, the relative importance of the idiosyncratic component declines as the frequency of the observations is reduced, and nearly vanishes when rates are observed at the monthly frequency. Using an identification scheme that allows for the fact that when policy actions are unknown to the market the funds rate should respond first to policy actions, we are unable to identify a unique effect of monetary policy in the funds rate at the daily frequency.