EIA projects global energy through to 2040, and other forms of shamanism

Posted on Updated on

A simple deduction from the history of energy forecasting is that we are absolutely hopeless at it. If we are looking at energy consumption levels thirty years from now or even oil prices three years from now all we see is a long history of inaccurate and largely pointless projections. There is however a market for this kind of thing. An anecdote from Dan Gardner’s excellent book “Future Babble” sums the whole business up. A British general in World War II was once angrily demanding that he get a weather forecast for two weeks ahead, when a meteorologist told him that the forecasts were completely unreliable. He responded appropriately: “Of course I know they are bloody useless, but I need them for planning.” This is very much the same attitude that politicians take to long range energy forecasts. They need a number for the cost of oil and natural gas in the next couple of decades to help provide some legitimacy to their policy decisions.

These projections of course never cease. Today we have news of the US government’s Energy Information Agency telling us what the global energy mix will look like over the next 30 years. The graph below sums up their projections:

So, the world is going to be run on fossil fuels for a long time to come. But just look at what happens to coal between 2000 and 2010. Coal consumption soared. Zooming in, this is what has happened to global coal consumption in the last decade:

This huge increase was, as far as I know, not foreseen in any projections of future energy consumption. This simple example shows a clear need to have very large confidence limits on our future projections, but just look at the implied confidence in EIA’s projections.

And the civil servants at the UK’s Department of Energy and Climate Change evidently get the same silly requests from UK policy makers. Here are their latest projections for the future prices of fossil fuels

On oil prices they project a  linear increase over the next two decades.


A nice straight line, (though unlike EIA they at least hint at uncertainty). And the actual history of oil prices:


I’ll let the reader decade how confident we should be about DECC’s projections.

Their gas forecast however is even more ridiculous. The “central scenario” predict that gas prices will literally remain unchanged throughout the 2020s. This is an obviously nonsensical prediction, so why make it?


A more realistic interpretation of the above is that DECC has absolutely no idea what the price of gas or oil will be next decade. Things may go up in price or may go down, and we don’t know which way they will go.

And I’ll close with a little story.

Last September the International Energy Agency issued a report, which argued the following:

“Forecasts based on futures prices, surveys of analyst forecasts, forecasts based on a variety of simple time series regressions and other common forecasting techniques are generally inferior to the random-walk forecast, which implies that the best forecast of crude oil spot prices is simply the current price of oil”

Rather obvious and sensible conclusions. Yet, these conclusions naturally were ignored a month later when the IEA confidently told the world that oil prices were going to decline. My guess is that a coin toss was not involved in making this prediction.