Liam Denning is a Bloomberg Gadfly columnist covering energy, mining and commodities. He previously was the editor of the Wall Street Journal's "Heard on the Street" column. Before that, he wrote for the Financial Times' Lex column. He has also worked as an investment banker and consultant.

I know, I know. What with tracking Canada's wildfires and Nigeria's unrest, you can barely tell where oil's going in the next 25 minutes, let alone 25 years.

Unpredictable, That's What You Are
WTI crude oil price
Source: Bloomberg
Note: Intraday pricing

Someone has to try, though. So on Wednesday, the Energy Information Administration released its annual International Energy Outlook, with projections through 2040. It's a report that oil bulls should love: The EIA's reference case has global oil consumption at just shy of 121 million barrels a day by the time I (hopefully) retire. That is a full 31 million barrels a day higher than its baseline year of 2012 (the EIA takes a 28-year view, rather than 25) or roughly equivalent to adding an extra U.S. and China combined.

Gulp, in every sense of the word.

Any number of reasons can be put forward as to why that number may be close to or way wide of the mark. Rising population and incomes, especially in the developing world, are the main arguments in favor. Political momentum toward capping carbon emissions and advances in batteries and autonomous driving are big counter-arguments.

But another question raised by that figure of 121 million barrels a day is: Where will the oil come from? Even over a span of 28 years, raising global oil production by 31 million barrels a day is no mean feat.

28 Years Later
Rolling 28-year additions to global oil production
Source: Energu Information Administration, BP

Reading that chart is a little tricky; sort of like saying "cast your mind back to 1993, and then cast that mind back to 1965." What you can see, though, is that by the early 1990s, we were coming off the wave of discoveries, in places such as Alaska and the North Sea, sparked by the oil crises of the 1970s. The bust of the late-1980s meant that, a decade later, the long-term trend of rising oil supply had been choked off -- setting up the bull market of the 2000s and the subsequent increase in supply from resources such as U.S. shale.

In contrast, the EIA's reference case points to a much more stable future. Its projections imply oil production will grow beyond 2015 by an average of just over 1 million barrels a day every year and, apart from the outlier year of 2017, won't deviate much from that. Compare that to the record of the past 25 years.

A Troubled Past
Annual changes in global oil supply over the past 25 years have varied considerably versus what is implied in the EIA's projections
Source: Energy Information Administration, BP

Of course, models are always smoother than reality, so this isn't a huge surprise. And who knows? What with the shorter time it takes to develop shale fields, maybe production will respond faster to price signals.

And, if anything, the world will need more shale output than even the EIA suggests if this scenario of 120 million barrels a day is ever to come to pass. That is because OPEC figures so large in meeting the extra demand. By 2040, its crude oil output, excluding natural gas liquids, is projected to be 47 million barrels a day, up from around 33 million now.

Now consider: A fifth of Nigeria's output is off-line due to attacks, Venezuela is reverting to a candle-based economy, Libya and Iraq are barely in control of their own territory, and Saudi Arabia is contemplating a less oily future. I am going to go out on a limb and suggest OPEC will really struggle to add more than the equivalent of another Saudi Arabia to its output over the next 25 years.

It is worth recalling that, 10 years ago this month, with oil demand -- and prices -- surging, OPEC itself was predicting that, by 2020, the world would be burning 106 million barrels a day and the cartel would be supplying 46 million of that, including natural gas liquids.

Today, OPEC produces 39 million, including NGLs, and there isn't much likelihood of it adding more than 2 or 3 million a day to that by the end of the decade -- and every possibility it could lose much of that as weaker members succumb to the pressures of low prices. OPEC's demand forecast for 2020 has also come down to 97 million barrels a day.

The implication from all this is that supply could tighten sharply, raising prices but also further encouraging conservation efforts. Alternatively, struggling OPEC members could be forced to open up reserves for faster development by foreign capital, or shale development may expand more quickly in North America and overseas. 

That is ultimately a deflationary oil world, in which OPEC largely ceases to have any meaning, competition rules, and factors eroding demand growth continue to mount. Here's guessing that demand never gets close to 120 million barrels a day.

This column does not necessarily reflect the opinion of Bloomberg LP and its owners.

To contact the author of this story:
Liam Denning in San Francisco at

To contact the editor responsible for this story:
Mark Gongloff at