Testing for speculative bubbles in agricultural commodity prices: a regime switching approach

Purpose – The authors' paper aims to deal with the question whether speculative bubbles are present in agricultural commodity prices. Design/methodology/approach – The authors apply a regime switching regression model to test the hypothesis that agricultural prices contain periodically collapsing bubbles. Using daily futures prices for six agricultural commodities, the authors calculate net convenience yields from which price fundamentals are derived. Findings – The authors discover pronounced deviations between observed prices and their fundamental values. However, they do not find evidence for the presence of periodically and partially collapsing speculative bubbles for five of six commodities. Except for soybeans, the signs and the significance of the estimated coefficients are not entirely in line with the predictions of the theoretical model. Originality/value – The authors' study adds to the heated discussion on the impact of speculative behavior on agricultural commodity prices. So far, most contributions in the literature either use theoretical arguments for the (non-) existence of bubbles or apply indirect tests which are plagued by low statistical reliability. In contrast, the authors apply a direct test. They find that the outcome of empirical bubble tests depends on the considered bubble type and on the testing procedure. In view of these ambiguities, definite statements on the presence of speculative bubbles as well as demands for limitations of speculative positions in commodity futures markets should be carefully reconsidered.