Non-stationary inflation and panel estimates of United States short and long-run Phillips curves

This paper argues that because United States inflation has been non-stationary over the past five decades the vast body of empirical research that proceeds without adequately accounting for the non-stationarity in the data is invalid. Using 50 years of United States inflation data the standard results in the Phillips curve literature are shown to be due to unaccounted shifts in the mean rates of inflation over the period. Short and long-run Phillips curves for the United States are then estimated using time series panel data techniques that account for these shifts in mean.