College of LAS « Illinois


Predicting the Flow of Oil

Economist untangles the complexity behind one of the world’s most sought commodities.

Oil drop

Of all the commodities swapped on the global market, you’d be hard-pressed to find a more emotional subject than oil. An economist at the U of I, however, has developed a statistical model that he believes can predict and help understand the volatile market even as it’s punctuated by murky contracts, international politics, hurricanes, and conflict.

Hadi Salehi Esfahani, professor of economics and director of the Center for South Asian and Middle Eastern Studies, has gathered data from 125 countries to gauge a market that has thus far largely defied predictions. At the heart of his model is a system of equations and a data grid analyzing how each country trades oil.

He can crunch the numbers on any given situation (hypothetical or real) that affects the flow of oil. The recent oil spill in the Gulf of Mexico? Esfahani hasn’t applied his model on that yet, but by thinking in terms of shipping and demand he’s pretty certain the environmental disaster had little effect on oil prices. Tensions between the West and Iran could be a very different matter.

According to his calculations, the current situation—that is, sanctions on investing in Iranian oil by some Western countries, including the U.S.—have not significantly affected the worldwide price of oil, says Esfahani, himself a native of Iran.

If tensions escalate into widespread sanctions on Iranian oil, the effect will become “tangible,” Esfahani predicts, with prices rising 10-15 percent in the long term. If tensions escalate so that Persian Gulf shipping is threatened, he says insurance rates on oil shipping in the region would go “through the roof” and force other oil producers to find alternate routes, driving oil prices up some 30 percent.

If tensions and altered trade routes cause an overall 10 percent drop in oil production in countries around the Persian Gulf, Esfahani predicts oil prices will surge by 50 percent. If war breaks out, he predicts oil prices would triple or more in the short term.

Esfahani has also gauged how economic growth has affected oil prices. By analyzing historical data he determined that a quarter of the recent upward trend in oil prices (a 190 percent increase from 1997 to 2007, adjusted for inflation) stemmed from growth in China and India. Another quarter was due to “transition countries” that grew after the Soviet Union collapsed, and another quarter came from growth in developed countries.

His model—which he hopes to soon present in conferences and publications—is more detailed than previous ones, Esfahani says.


“Some other researchers have tried to do similar things, but they look at regions,” he says. “They look at East Asia, and not Japan, Korea, Taiwan, and China separately. These individual countries have different characteristics, different needs, and different market conditions.”

The model cannot account for the effects of sudden events such as Hurricane Katrina, which within days spiked the price of processed petroleum. Nor can it predict the effects of unknowns such as sweetheart deals and corruption. To Esfahani, however, those factors are random events that he has compensated for in statistical equations.

“Basically [my] contribution is to find a way or separate out and identify the impact of each variable, statistically, correcting for factors we don’t observe,” Esfahani says. “Secondly, the model tries to examine trade at the very micro-level and look at individual countries trading with each other.”

The model addresses many issues directly, including distance between trading partners, alternative trade routes, composition of oil, trade agreements, and effects on GDP. New oil discoveries in Brazil, for example, would increase that country’s own oil consumption.

He has also found that the amount of oil a country exports depends not only on the volume of its oil production, but also how efficiently and cheaply it can produce the oil.

The overarching theme of his work, however, is determining what shapes and hinders the oil trade.

“One of the things that is important in economics is barriers to trade,” he says. “If you remove barriers to trade, there are opportunities for improving everybody’s life, because unnecessary costs are going to be removed.”

By Dave Evensen
Winter 2011