We estimate a model of common and commodity-specific, high- and low-frequency factors, built on the spline-GARCH model of Engle and Rangel (2008) to explain the period of exceptionally high price volatility in commodity markets during 2006-2008. We find that decomposing realized volatility into high- and low-frequency components reveals the impact of slowly-evolving macroeconomic variables on the price volatility. Further, we find that while macroeconomic variables have similar effects within the same commodity category (e.g., storable agricultural), they have different effects across commodity groups (e.g., live stock versus energy).