Measuring and Predicting Asset-Price Bubbles : Empirical Learnings from the Oil Market[s]

Ten years on from the collapse of mortgage markets in the United States and Britain and ensuing ‘Global Financial Crisis', this article examines the price dynamics of energy commodities, focusing on speculative bubbles, looking more particularly for clues as to when asset-price bubbles are formed and whether it's possible to predict their occurrence. Using bubble detection techniques in the context of a rational asset pricing model for commodities, we test for and date periods where prices are considerably higher than fundamentals. The price collapses in 2008 and the more recent one in 2014–2015 are of a particular interest. For crude oil analysis, in addition to the traditional convenience yield approach, we also utilize oil supply and demand data. We separately analyse three key crude oil benchmarks: WTI, Brent, and the lesser known Dubai Crude benchmark known as Dubai Fateh, and find important disparities. Overall, there are substantial differences in market bubble manifestations across supposedly ‘correlated' energy commodities. Speculative bubbles appear rife in the period from late 2007 to mid–2008, with no clear evidence of exuberance thereafter: in other words, it would have been difficult to predict the 2014–2016 oil glut using purely financial modeling and other statistical approaches, without incorporating ad hoc geopolitical and societal factors. Our research also suggests that the Dubai benchmark, an index intimately tied to Chinese demand, can be construed as a lead-indicator of speculative bubbles in crude oil and, by extension, other financial markets: a finding which should prove useful for future scholarly and institutional investment research