Files
Abstract
The nature of commodity price behavior remains in substantial dispute. Previous empirical tests of the standard storage model have concluded that storage arbitrage cannot explain the high serial correlation of prices. We introduce a Maximum Likelihood estimator that, unlike available approaches, makes full use of the predictions of the model about the most striking feature of commodity prices, the skewness revealed in their asymmetric dynamics, displaying occasional spikes. Our results for sugar establish the empirical relevance of storage in determining its price behavior. The dynamics of commodity prices can be quite different from those of the shocks that drive them.