This article uses both linear and nonlinear causality tests to examine the causal relationships between changes in commodity prices and U.S. inflation. Prior to the Great Moderation, there is evidence that changes in commodity price indices linearly lead inflation. The stability of the causal relationship appears to vary over time with a stronger bivariate link established before the Great Moderation. Further, there is evidence of significant nonlinear causality from raw industrials and metals indices to inflation with most of this detected nonlinear relationship being captured using the Baba, Engle, Kraft, and Kroner asymmetric generalized autoregressive conditional heteroskedasticity model. This implies that the observed nonlinear Granger causality is largely driven by unanticipated shocks and volatility spillovers in the run‐up of commodity prices in late 2000. (JEL C18, C22, E31)