Oil Price Forecasting: It’s a Mug’s Game

To paraphrase Ralph Waldo Emerson, foolish precision in the face of uncertainty is the hobgoblin of energy forecasters.  Much has been made recently about how energy forecasters—private and government—seriously underestimated the large increase in crude oil prices that has taken place in recent months.

What is surprising is that so many have been surprised and that so many so-called analysts continue to make point estimates. Crude oil prices have never been easy to forecast very far in the future, except under very stable conditions. And, that difficulty in forecasting also applies to oil companies who have a very strong economic incentive for greater accuracy.  The problem is that there is a great deal of uncertainty in the political, economic, and technical factors that can drive price.

In spite of sophisticated models, forecasters too often assume that the near term will be like the present at that influences how they treat economic and political factors.  Most important, point estimates for uncertain variables guarantees that an estimate will only be correct by luck.  How much the dollar will strengthen or weaken, how companies will respond to rising oil prices, whether shale producers will start drilling or maintain fiscal discipline, how much global growth will occur, whether OPEC producers begin to cheat and how much, how much Iranian and Venezuelan production will fall, and the impact of political risk are all factors that work against point estimates.

Bill Gilmer, Director of the Institute for Regional Forecasting, in a recent Forbes article made the obvious point that the world is dynamic; forecasts are static in that they are based on information available at the time they are made.  It is one thing to use available data to look ahead; it is something else to pretend to divine the future with relative precision.  Gilmer points out that “it doesn’t take big headlines to upset the forecast. The global crude oil market depends on the politics of dozens of producing countries, economic cycles in consumer countries and a vast infrastructure of pipes, ships and refineries. Even if we account for the known issues correctly, we could list 1,000 or more low-probability events that could push our forecast off course. … if these events are independent of each other, the chance that at least one will significantly and unexpectedly affect the oil market within a year is 1-(.999)1000 or 63.2%.”

Oil price forecasts would be more credible if they provided ranges reflecting uncertainty.  Otherwise, forecasting is a “Mugs Game,” an exercise in futility.