Economics 101: Best predictions of oil price are the ones you never hear

Opec and non-Opec oil producers are likely to extend their agreement to limit supplies beyond the June expiry, some Opec delegates say. Alexander Klein / AFP

Kuwait, Qatar, Saudi Arabia and the UAE are currently preparing for the Opec meeting on May 25. What the group intends to do depends upon its forecast for oil prices, but as usual, observers are split between those expecting higher oil prices versus those expecting lower oil prices. What makes oil prices so hard to predict?

Oil analysts are representative of economists in general in that they are paid handsomely to produce forecasts that are generally inaccurate – often missing booms and busts. How do economic forecasters continue to “con” people into paying them so much to apparently deliver so little? Two factors in combination play an important role.

First, oil prices are an application of the efficient markets hypothesis (EMH), in that the best prediction of a future oil price is approximately the prevailing one. This theory earned Eugene Fama the 2013 economics Nobel Prize, but is widely misinterpreted as a claim that markets are perfect, a misconception that is partially attributable to the theory’s poor naming.

Understand the EMH requires a quick epistemological detour.

In natural sciences, such as physics and chemistry, the phenomenon being studied does not modify its behaviour in response to our gaining a better understanding of its behaviour. For example, developing an accurate model of how much weight steel can bear before deforming does not alter steel’s strength.

Social sciences, including economics, are different, because humans have the ability to – and an incentive to – alter their behaviour in response to the development and publication of an accurate model of their behaviour. If a car salesperson explains to you how she predicts your income based on your appearance, or a cross-examining lawyer outlines to you his strategy for portraying you as a liar, you will most likely change the way in which you behave, a choice unavailable to a volcano being studied by seismologists.

In the context of commodity markets, including oil, if it becomes publicly known that oil prices next week are likely to be $10 higher than their current level, people will not sit still – there is money at stake! Instead, they will begin to buy oil, forcing the price up by $10 well in advance of next week, and invalidating the original prediction. Conversely, if people are confident that oil prices will crash six months from today, the selling will start immediately, bringing the crash forward, rendering the forecast completely inaccurate.

Thus, the EMH gives us a very clear explanation of why public predictions are doomed to failure. Economists and oil analysts are not intellectually deficient – they are studying phenomena that mutate in response to being studied, an inescapable catch-22.

There is a backdoor, however, which constitutes the second factor underlying the persistently high remuneration of commodity forecasters – the EMH depends upon our discoveries about human behaviour being made public. If they are kept confidential and exploited by traders with limited capital, and hence a limited capacity to influence the market, then forecasters can justify their six-figure salaries. Good forecasters charge big fees to their clients, who then go about their business quietly without alerting the market to the key insights, lest they be instantly eroded by the EMH. If a for-profit forecaster is publicly talking about oil markets, then you are only hearing half of the story, or they are just engaging in some public relations outreach. Most public statements come from those paid to attract attention, such as journalists, or those seeking to influence markets, such as oil ministers.

Thus, while a meteorologist has every incentive to disclose his new, super-accurate rain-forecasting model, as clouds do not read the newspapers, an oil forecaster has every incentive to keep her mouth shut, because oil producers do read the newspapers. Accurate oil predictions exist – you just have to pay to hear them!

Omar Al Ubaydli is programme director for international and geopolitical studies at the Bahrain Centre for Strategic, International and Energy Studies, and an affiliated associate professor of economics at George Mason University. He welcomes economics questions from readers via email ([e-mail]) or tweet (@omareconomics).

Follow The National's Business section on Twitter