It was the wildest of wild rides. In mid-January of 2007, crude oil futures on the NYMEX Exchange closed at $52 a barrel. Eighteen months later, a relentless climb had pushed prices to an all-time, inflation-adjusted high of $145 a barrel. World markets panicked, and analysts predicted that we'd never see $100 oil again.
Five months later, the price had plummeted by nearly 77% to $34 a barrel. Even so, analysts and many academics stuck to their guns. The culprit, they said, was not speculation, but the fundamentals: A growing demand for oil in rapidly modernizing countries like China and India, plus the long-standing appetite for energy in the West. And all this was happening as the ability to develop new sources of oil seemed dim. Fluctuations in supply and demand, they said, were the drivers of the wild price swings.
But was that really the case? A new analysis of the 2008 oil shock by Kenneth Singleton of Stanford GSB casts doubt on the conventional wisdom. He argues that market participants had imperfect information about many of the key drivers of prices, including supplies, demands, and inventories. For demands and inventories, these informational problems were particularly acute for emerging economies. As a result, "there was an economically and statistically significant effect of investor flows on futures prices," he wrote. It's not yet clear, though, what the price of oil might have been without the flow of investor money.
Confronted with imperfect measurement, investors naturally attempted to learn about fundamentals from past market prices or returns. The associated trading activities of the major players (index investors, hedge funds, broker/dealers in commodity markets, and commercial hedgers), along with changes in supply and demand, are what ultimately determine prices in commodity spot and futures markets, says Singleton.
Although the paper has yet to be published, Singleton's presentation in August to a forum held by the Commodity Futures Trading Commission in Washington, D.C., has already stirred controversy. His conclusions are in opposition to those of many Wall Street analysts and to the CFTC itself, at a time when regulators in the United States and Europe are considering rules to limit trading by banks, hedge funds, and funds tracking commodity indices, including oil. A more developed structural model of price determination that incorporates the informational challenges and limits to arbitrage highlighted in Singleton's analysis is needed to quantify the degree to which investor flows affected the level of oil prices. He hopes to explore these issues in future research.
Singleton believes that the investor-driven swings in prices were costly to society as a whole, but he does not agree with the populist assumption that the phenomenon was, at bottom, about excessive or irrational speculation. "Occasional booms and busts in asset prices are likely to be an inherent feature of market settings in which there are structural information and funding frictions," he said in an interview.
Singleton, the Adams Distinguished Professor of Management, was recently named editor of the prestigious Journal of Finance. His working paper, "Investor Flows and the 2008 Boom/Bust in Oil Prices," is an outgrowth of a survey paper he prepared for the Air Transport Association of America.
Many experts who theorize about oil pricing assume that information in the marketplace is reasonably accurate and investors fully understand the structure of the global economy. Singleton disagrees, saying that traders may find it optimal to learn about the global economy and oil-market-relevant fundamentals from past returns.
"There is likely to be some disagreement about virtually every source of fundamental risk, including the future of global demands, the prospects for supply, future financing costs, and so on," says Singleton. One particularly knotty area is forecasting economic growth in the developing world, a key driver of demand for oil.
Indeed, when he plotted the cross-sectional dispersion of the price forecasts of professionals (a measure of the extent of their disagreement) versus the actual price of crude oil, Singleton found that as prices increased, the professional forecasters were disagreeing much more with each other. Simply put, they were more uncertain about what was happening in the market.
Extensive disagreement and learning from past returns tends to generate momentum in prices as it can lead investors to become more optimistic about future oil prices precisely when they are positively surprised by recent returns. Such optimism as prices rise, and similarly pessimism following price declines, may amplify price movements.
"If index investors are just slightly too optimistic [in market rallies] or pessimistic [in market downturns] relative to the true state of the world, then their errors, while inconsequential for their own welfare, may be material for society as a whole."
"These phenomena are entirely absent, essentially by assumption, from the models of oil price determination that focus on representative suppliers, consumers, and hedgers."
Singleton's analysis is part of a growing literature documenting the "financialization" of commodity markets: as retail, institutional, and hedge-fund investors have increased the weights on commodities in their portfolio allocations, there have been significant changes in the behavior of commodity prices. For instance, equity returns in emerging markets have become more highly correlated with changes in oil prices.
This literature is not yet sufficiently developed to yield clear policy prescriptions. Nevertheless, "it is clear that greater transparency in the oil market - about commodity-related positions of investor groups, the spare capacities of suppliers, and global inventories - would serve to mitigate some of the informational frictions underlying the occasional booms and busts in commodity markets," says Singleton.