Commodity prices redux: A global factor story

Abstract

In this study, we novelly employ a Bayesian Dynamic Factor Model (BDFM) and a Factor-Augmented Vector Autoregressive (FAVAR) model to investigate the relatively under-explored phenomenon of cross-commodity price synchronisation and the factors driving comovement in commodity prices. Our findings indicate that macroeconomic and financial factors are key determinants of world commodity prices, whereas uncertainty plays a comparatively minor role, accounting for less than 10% initially and remaining below 20% at longer horizons. At the level of commodity group prices, however, uncertainty becomes significantly more important, with its contribution exceeding one-third of the variance at longer horizons across all commodity groups. These results highlight a clear distinction between aggregate and disaggregated dynamics: global commodity prices are largely driven by macroeconomic and financial conditions, implying that policymakers retain meaningful scope to influence them, whereas commodity-group prices are more sensitive to shocks and uncertainty.