There is a common perception that the prices of unrelated commodities move together. This paper re-examines this notion, using a measure of comovement of economic time series called concordance. Concordance measures the proportion of time that the prices of two commodities are concurrently in the same boom period or same slump period. Using data on the prices of several unrelated commodities, the paper finds no evidence of comovement in commodity prices. The results carry an important policy implication, as the study provides no support for earlier claims of irrational trading behavior by participants in world commodity markets.

There is a common perception that the prices of unrelated commodities move together. This paper re-examines this notion, using a measure of co-movement of economic time series called concordance. Concordance measures the proportion of time that the prices of two commodities are concurrently in the same boom period or same slump period. Using data on the prices of several unrelated commodities, the paper finds no evidence of co-movement in commodity prices. The results carry an important policy implication, as the study provides no support for earlier claims of irrational trading behaviour by participants in world commodity markets.

Re-examines the notion that the prices of unrelated commodities move together, using a measure of co-movement of economic time series called concordance.

This paper tests and confirms the existence of a puzzling phenomenon - the prices of largely unrelated raw commodities have a persistent tendency to move together. We show that this comovement of prices is well in excess of anything that can be explained by the common effects of past, current, or expected future values of macroeconomic variables such as inflation, industrial production, interest rates, and exchange rates. These results are a rejection of the standard competitive model of commodity price formation with storage.

This paper provides a comprehensive analysis of the degree of co-movement among the nominal price returns of 11 major energy, agricultural, and food commodities using monthly data between 1970 and 2013. The authors study the extent and the time evolution of unconditional and conditional correlations using a uniform-spacings testing approach, a multivariate dynamic conditional correlation model and a rolling regression procedure.