A recent report about undue pressure by the United States on the Paris-based International Energy Agency, IEA, to distort crude oil projection figures is rather unsettling. The report, citing a whistleblower at IEA, indicated that the United States influenced the agency to underplay data, which indicated imminent, crude oil supply shortages while overplaying prospects of new discoveries. A U.S. motive for this, however, has not reallly been articulated. While the report has been denied and described as groundless by the IEA’s executive director, Nobuo Tanaka, it does raise some issues about IEA’s forecasts in particular and crude oil forecasts in general.

First is the issue of credibility. If (and really if) the claim of such undue influence were true, just how often has that occurred in the past? To what degree have the forecast figures been underplayed or overplayed? What degree of reliability can subsequent forecasts command? Since IEA data are often employed in their official planning, what harm might have been done to member-nations’ development plans by such underplay of forecast figures? Moreover, if the U.S. has exerted such influence on the IEA which membership derives from OECD countries, might she have exerted any such influence on her own Washington-based Energy Information Administration, EIA and to what effect?

Secondly, there is the issue of underlying criteria employed in forecasting. Different models employ different criteria, which often lead to different forecast figures. Some of these criteria however, may be unrealistic. For example, in IEA’s 2008 production forecast to the year 2030, more than half of projected supply by that year derives from oilfields yet to be found or developed. This, and the production sequencing of these oilfields, may therefore lend credence to the charge of distortion of forecast figures leveled against the agency. While it may be standard practice with some models, factoring in yet-to-be-discovered oil in an oil supply forecast is, to paraphrase one accountant, akin to preparing next year’s income profile and including proceeds from a yet-to-be-drawn lottery.

Even within a professional community, there may hardly be any agreement. For example, IEA’s survey of the rates of production for about 800 major oilfields around the world concluded that global oil supplies are declining at 6.7% per year which is about twice the previously given rate, but IHS CERA’s survey held that “sixty percent of the more than 1,000 fields examined in detail for the study were found to have production levels that were either steady or climbing”. One is not certain how many oilfields were common to both surveys but the conclusions are prima facie, incongruent.

Thirdly, various analyses on the accuracy of forecasts (forecast versus observed figures) have also led to various interpretations and many are not encouraging. There are far too many uncertainties to forecasting that projection figures carry no ironclad guarantees (and this has necessitated frequent revisions). For example, very few projections envisioned the oil price shock of 2008 as well as the consequent crash in demand. Political and other vagaries especially in the more prone or even in least expected producer-zones, can affect actual supply. In addition, global policies as they pertain to curbing deleterious (fossil fuel) emissions, can also affect demand. Then there is the uncertainty of technology: yet-to-be-discovered technology, whether in the oil industry (for accessibility to geologically intractable reserves) or in renewable energy (for easier, cheaper, more efficient production, storage etc). Significant technological advancement in any form of energy can tilt interest in its favor. For example, the hydraulic fracturing process has led to increased access to shale gas especialliy in the U.S. (though environmental concerns and the rapid losses often encountered due perhaps to associated steep pressure gradients, mean it is “not yet Uhuru”).

Finally, two quick points on estimates (whether of crude oil reserves, price, demand, supply and the likes):