Time-Varying Phillips Curves
A growing theoretical literature argues that aggregate price flexibility and the inflation-output tradeoff faced by central banks should rise with microeconomic price change dispersion. However, there is little empirical work testing this prediction. I fill this gap by estimating time-varying forward looking New-Keynesian Phillips Curves (NKPC). I reject a NKPC with constant inflation-output tradeoff in favor of a slope that increases with microeconomic volatility. In contrast, there is no evidence that the inflation-output tradeoff varies with aggregate volatility or the business cycle more generally. Furthermore, I show that greater volatility does not affect price flexibility purely through increases in frequency.
I would like to thank Priyanka Anand, David Berger, Eduardo Engel, John Leahy, Alejandro Justiniano and Marty Eichenbaum for useful suggestions. This is a greatly expanded version of work which first appeared in the first chapter of my dissertation. The views expressed herein are those of the author and do not necessarily reflect the views of the National Bureau of Economic Research.