The Monetary Policy Risk Premium and Expected Bond Returns

This brief note builds on Sabol (2015) by describing ways to account for forecasting errors made about the expected path of short-term interest rates in a model of expected bond returns. I consider the Cieslak and Povala (2014) model of monetary policy expectations frictions as one such measure of unexpected returns. I conduct a real time out-of-sample forecasting exercise and provide figures to easily show the validity of these models. Adding the predictable changes in Fed Policy, or the monetary policy risk premium, to measures of expected returns leads to improved forecasts. Much of this gain accrues to forecasts of shorter duration bonds.