What Drives Inflation? Lessons from Disaggregated Price Data

The COVID pandemic triggered high inflation alongside large relative price movements. I document that this comovement is not unique to the COVID episode, and follows a systematic pattern: the relative price of industries with flexible prices and inelastic factor supply rises with the aggregate output gap, and, controlling for the output gap, aggregate inflation rises with the relative price of those same industries. I show that both facts arise naturally in a multi-sector New Keynesian model where industries have heterogeneous price adjustment frequencies and factor supply elasticities. Through the model, I compute the share of consumer price inflation due to aggregate demand misalignment versus relative price adjustments, and construct price indices which forecast inflation at different horizons. Applying this framework to US data, relative price adjustments account for 49% of inflation fluctuations between 2000 and 2019, but only for 31% over the COVID period (2020-2023).