Nonlinear cointegration in the food-ethanol-oil system
Evidence from smooth threshold vector error correction models
We examine the nature of relationship between prices of crude oil, ethanol and grains (maize, wheat and rice). Our working hypothesis is that profit maximization, the US biofuel policies and automotive engine technology give rise to a nonlinear relationship between oil and ethanol prices, and by extension between oil and grains prices. While legislation sets a floor in the ethanol market, engine technology, which shapes fuel substitution, sets a ceiling. We explore price relationships in the food-ethanol-oil nexus by applying both discrete and smooth threshold error correction models. First, we find that oil prices are the long run drivers of ethanol and grains prices. Second, ethanol prices co-move with oil prices in the long run. However, in the short run, oil and ethanol prices are linked in a nonlinear manner. Ethanol prices appear to drift apart from the path as this is determined by oil prices, due to policy changes. Adjustment back to the long run equilibrium the path is rapid, less than two months, when the deviations are small. Large deviations take more time to be corrected.
C32, Q11, Q41
Permanent HTML link