Using a Projection Method to Analyze Inflation Bias in a Micro-Founded Model∗

Since Kydland and Prescott (1977) and Barro and Gordon (1983), a quadratic one-period loss function has been used widely in most of existing literature to analyze inflation bias induced by discretion. In this paper, instead of following the conventional linear-quadratic approach, we use a projection method to investigate the size of inflation bias in a mirco-founded non-linear model where firms follow the Calvo pricing-setting rule. The size of inflation bias is between 1% and 4%, depending on parameter values. These results on inflation bias based on a projection method are then compared with those based on the linear-quadratic approximation as in Woodford (2003): The linearquadratic approximation underestimates the inflation bias by up to about a tenth. ∗The authors appreciate comments by Andrew Levin, David Lopez-Salido, Alex Wolman, and participants at 2008 SCE Conference, as well as encouragements from Mike Woodford. The views in this paper are solely the responsibility of the authors and should not be interpreted as reflecting the views of the Board of Governors of the Federal Reserve System or any other person associated with the Federal Reserve System.