Abstract
In this paper, we use panel data to test whether Federal Open Market Committee (FOMC) board members' forecasts are rational. Rationality is rejected in the sense that forecasts by members are heavily dependent on previous own forecasts and last consensus made in FOMC. Furthermore, we reveal the strategic behavior of FOMC board members. Forecasts by governors, who always have voting rights, agree much with the previous consensus of FOMC members' forecasts. In contrast, non-governors, who rotate voting rights, exaggerate their forecasts: they aggressively deviate their forecasts from previous consensus. The former is herding behavior and the latter is anti-herding behavior. Our results imply that individual members behave strategically; governors want to present policy-consistent forecasts to the Congress and non-governors utilize their forecasts to influence decision making in FOMC.
