Why don’t oil shocks cause inflation? Evidence from disaggregate inflation data

Abstract

This paper uses disaggregate U.S. inflation data to evaluate explanations for the breakdown of the relationship between oil price shocks and consumer price inflation. A data set with measures of inflation, energy intensity, labor intensity, and sensitivity to monetary policy is constructed for 97 sectors that make up core CPI inflation. A comparison of the 1973–85 and 1986–2006 time periods reveals that substitution away from energy use in production and monetary policy were both important, with approximately two-thirds of the change in response of inflation to oil shocks being due to reduced energy usage, and one-third to monetary policy. We find no evidence that other factors, such as changes in wage rigidities or changes in the persistence of oil shocks, played a role.