the Board of Governors of the Federal Reserve System, jointly presented their paper on “The Performance of ForecastBased Monetary Policy Rules under Model uncertainty”. Their quantitative analysis compared forecast-based monetary policy rules with outcome

We investigate the role of expectations formation in an environment where agents have imperfect knowledge of the precise structure of the economy and expectations are governed by a perpetual learning technology that nests rational expectations as the limiting case of perfect knowledge. Learning raises the sensitivity of inflation expectations to economic shocks and gives rise to endogenous inflation scares. The ability of policy to control macroe-conomic fluctuations is more limited than appears under rational expectations and some stabilization policies that might appear optimal under rational expectations are, in fact, strictly inecient. Forecast- and outcome-based policies that are isomorphic under rational expectations yield very dierent outcomes when knowledge is imperfect. We show that the prevalence of inflation scares depends on policy design and nd that monitoring and responding to the public’s inflation expectations in addition to responding actual inflation improves macroeconomic performance