Financial Frictions, the Phillips Curve and Monetary Policy

How does the presence of financial frictions alter the Phillips curve and the conduct of optimal monetary policy? I investigate this question in a tractable small-scale New Keynesian DSGE model with a financial accelerator. The accelerator amplifies shocks, decreases the slope of the Phillips curve and renders forward-looking behavior more relevant for current macroeconomic dynamics. I show analytically that these three factors imply an inflationary bias of discretionary monetary policy relative to the standard model and a stabilization bias relative to commitment policy. A conservative central banker who places a larger weight on inflation stabilization than society is able to reduce both biases and closely mimics the optimal policy under commitment. The required degree of inflation conservatism increases in the degree to which financial frictions are present.