Inflation Forecast

Core CPI Inflation: Persistence and Expectations

Two independent models of core CPI inflation, updated as new data arrives.

The models ask how much inflation’s past behavior and households’ inflation expectations tell us about where inflation is headed. Persistence model — a Bayesian Phillips curve combining a prior estimated from 1958–99 inflation dynamics with 2000–present data; a heavier weight on the prior anchors the forecast toward that pre‑2000 relationship, a lighter weight lets recent data dominate. Expectations model — a Phillips curve using University of Michigan survey expectations in place of (pure) or alongside (hybrid) lagged inflation.

Persistence model based on M.T. Kiley (2022), “Anchored or Not: How Much Information Does 21st Century Data Contain on Inflation Dynamics?,” FEDS Working Paper 2022‑016, and M.T. Kiley (2023), “A (Bayesian) Update on Inflation and Inflation Persistence,” FEDS Notes.

Expectations model based on M.T. Kiley (2015), “Low Inflation in the United States: A Summary of Recent Research,” FEDS Notes, and J.M. Roberts (1995), “New Keynesian Economics and the Phillips Curve,” Journal of Money, Credit and Banking, 27(4), 975–984.

Q4/Q4 core CPI inflation, this year and next
Computed in real time: for a year already partly realized, actual months are used where known and each model's own forecast fills in the rest.
12-month percent change, history and forecast
Since 2010. Dashed vertical line marks the last actual data point; scenarios diverge from there.