The coming decline of oil

Posted: 19 March 2002

Author: Gerald Leach

Amongst the billions of words brought forth by the climate debate over the past years, remarkably few have touched on an issue that ticks behind it like an unexploded time bomb. This is the probability that world oil production will reach a peak sometime during this decade and then start to fall, never to rise again. Gerald Leach reports.

An abundant and growing supply of cheap oil has been one of the fundamental drivers of increasing prosperity during the past century. At the same time, supply failure - or the fear of its failure - has been a principal source of economic hardship, geopolitical manoeuvring, and war. Equally, an imminent peak and decline of this industrial lifeblood promises severe economic difficulties and political tensions across the planet that will leave few untouched. These world-wide upheavals may outweigh in scale and gravity any impacts on climate-related questions such as greenhouse gas mitigation strategies and emissions reduction targets. But they will nevertheless upset these considerably. If oil "runs short" in the next few years there will be strong pressures to use more coal (including coal-based liquids for transport) and to step-up production of hard-to-extract "non-conventional" oils from shales and tar sands. Both moves would greatly increase carbon dioxide and other greenhouse gas emissions per unit of energy produced.chart© State of the World 1999, The Worldwatch Institute Pressures to save oil would also build up, for example, by more efficient use and faster deployment of renewable energy alternatives such as biofuels. The big question here is whether these greenhouse gas-reducing and oil-saving responses can do enough in time to compensate for oil's decline. With so much at stake, it may seem odd that so little has appeared in the media - let alone, the climate debate - about these concerns. Why not? And why are increasing numbers of oil experts now pointing to an imminent decline of global oil output? Are they just crying "wolf" or "limits to growth" again, as in the 1960s and 1970s? Or do they have convincing arguments up their sleeves this time? Half way to the ultimate

To start with the obvious, the planet has a finite endowment of oil. Petroleum experts call this quantity the "ultimately recoverable reserve," or simply the "Ultimate." At any time, the Ultimate equals Cumulative Production + Reserves (proved, probable, possible) + Yet-to-Find. One set of estimates for conventional oil at the end of 1999 was, in billions of barrels (Bb): Cumulative Production 820, Reserves 827, Yet-to-Find 153, with an Ultimate of 1800. On these estimates, cumulative production will be around 900, or half way to the Ultimate, in the year 2003. Ominously, in view of events since 11 September 2001, about half the Yet-to-Find lies in just five Middle East countries: Iran, Iraq, Kuwait, Saudi Arabia and the United Arab Emirates.

The halfway point to the Ultimate is a crucial milestone. Strong historical evidence for many oil basins shows that oil production peaks and starts to decline when about half the recoverable resource has been consumed. This is a matter of oil reservoir physics rather than production technology or economics. The idea was first proposed by the United States oil geologist M King Hubbert who predicted, in 1956, using two estimates of that country's Oil Ultimate, that United States oil production would peak in the early 1970s. Hubbert was widely ridiculed by the oil industry, but was proved right when United States oil production peaked and began to fall in 1971. Hubbert's model was taken up in the 1970s and early 1980s by competent authorities including Esso, Shell and the World Bank. Using a global Ultimate of around 2000 Bb they all independently predicted a conventional oil peak close to the year 2000. However, the forecasts overlooked the slowdown in global oil demand growth following the 1970s oil price shocks.

Correcting for this postpones their predicted peak to about 2010. In the mid-1990s, several independent analysts applied the same method to improved reserve data and broadly agreed that non-OPEC (Organization of Petroleum-Exporting Countries) oil production would peak about now, while global production would peak around 2005-2010.

This graph depicts world oil production to date. The vertical line indicates the probable midpoint of depletion as identified by Dr Colin Campbell, renowned oil geologist.©

Once oil production does peak, the analyses suggest that production will fall each year by about two million barrels per day. Daily production is now about 74 million barrels, so the fall will be some 2.7 per cent annually. That means a huge and highly disruptive shift from present-day growth in oil use. Although the current growth rate is running at only 1.1 per cent per year for the world, and 1.3 per cent for the OECD (Organization for Economic Cooperation and Development) industrialized countries, annual growth has been 2.1 per cent for the world less the former Soviet and East Europe block, 2.8 per cent for Latin America, 5.4 per cent for India and 7.5 per cent for China. Meanwhile, as non-OPEC producers lead the way over the oil peak and down the other side, the world will become even more dependent than now on the handful of Middle East suppliers. Towards the end of this decade or, if you like, during the first Kyoto Protocol commitment period of 2008-12, even they will be unable to turn up the taps - much, or at all - to meet almost inevitable supply-demand shortfalls.

Expert denials and support Many energy experts are blind to this looming problem - or deny it. There is virtually no mention of it, for example, in the summaries of the three-volume Third Assessment Report, published by the Intergovernmental Panel on Climate Change in 2001. The volume on Mitigation says nothing about oil supply in its summary of "Gaps in Knowledge" and concludes that "fossil fuel scarcity, at least at the global level, is not a significant factor in considering climate change." All of the Intergovernmental Panel's greenhouse gas emission scenarios have increasing global oil production to at least 2020, with steady oil prices. The latest (November 2001) projections by the United States Department of Energy take a similar line. From now to 2020, world oil consumption will rise by 56 per cent; OPEC production will nearly double; and more surprisingly, non-OPEC production will climb by 34 per cent. Other experts are more pessimistic - or at least uncertain. The influential, Paris-based International Energy Agency is a good example. It delayed adopting the Hubbert model until its 1998 World Energy Outlook.

The 1998 assessment used a generous United States Geological Service estimate of 2300 Bb for the world Ultimate, but even so predicted that non-OPEC oil production was more or less at peak, and that world production of conventional oil would decline from about 2015. This was a startling conclusion at the time and was published only after considerable internal pressure. For its June 2000 Outlook, the International Energy Agency changed tack. Using a much larger United States Geological Service Ultimate (which many experts consider spurious), it concluded that rising oil demand could be met out to the year 2020. A year later, the International Energy Agency altered course again. Its 2001 Outlook concludes that the oil reservoirs of OECD countries will soon peak and decline, throwing the world into increasing dependence on a small number of Middle East suppliers. Massive infrastructure additions will be needed to offset this outlook by bringing more natural gas to market and burning coal more cleanly. Why are there such differences amongst the experts? And why such a blanket of indifference or ignorance amongst politicians, the media and many economists? The answers are elusive but can probably be found hovering near some common misconceptions about oil production and reserves. Also to blame are profound differences between oil geologists who recognize physical constraints on reservoir production and traditional economists who assume that higher prices will solve all resource constraints: the dollar-inspired Cornucopia model of the Earth. The chief misconceptions seem, briefly, to be the following. What reserves?

Published oil reserve data indicate how much oil might be produced from known fields, with stated degrees of probability. These are "proved plus probable reserves." In contrast, "total recoverable oil" includes oil recoverable in fields that have not yet been discovered. Many people confuse the two quantities, saying for example that "30 years ago we had 30 years of supply remaining; now we have 40 years remaining." Estimates of 2000 Bb for the world's total recoverable (or Ultimate) have hardly changed over the past 40 years. It is this fixed endowment that makes the near-term decline of conventional oil inevitable. Proved + Probable Reserves.pie chartProved reserves of oilGenerally taken to be those quantities that geological and engineering informaton indicates with reasonable certainty can be recovered in the future from known reservoirs under existing economic and operating conditions.© BP plcOil companies are legally bound to report only their proved reserves. Many of them state in their annual reports that these have grown over the year because extractions have been more than replaced by discoveries and by expectations of greater oil recovery from known fields. This is good for the company's share price. However, the amount of oil in place - the proved + probable reserve - has not changed one bit. Reserves have simply been re-categorized as reported oil is "moved" from probable to proved reserves. Reserves to Production Ratio. An enticingly simple number used by oil analysts is the amount of oil in proved reserves divided by annual production. The global figure for 2001, according to British Petroleum's Statistical Review of World Energy, is 39.9 years. So what's the problem - if we have enough oil for 40 years at our present rate of use (and more will be found as we use it up)? But the Reserves/ Production ratio says nothing at all about physical constraints on the rate of oil production and hence about the Hubbert decline model. A simple calculation shows that the global reserves indicated by today's Reserves/ Production ratio of 40 would only be two-thirds used up over the next 40 years if production started falling today at 3 per cent a year. However, production in 40 year's time would be less than one-third of today's level. Non-conventional oil. Many analysts seem to believe that the "immense" reserves of non-conventional oil from the Canadian Athabaska tar sands and the Venezuelan Orinoco heavy oil deposits will be brought on stream and save the day. Their recoverable reserves are each about 300 Bb: large, yes, but taken together no more than 22 years of current global oil consumption. Severe extraction problems including lack of technological readiness, low energy content, high water needs and emissions of carbon dioxide and other pollutants make it improbable that these resources can come on-stream fast enough to compensate for the decline of conventional oil. The International Energy Agency's 1998 World Energy Outlook concluded that 19 Mb per day of non-conventional oil would be needed by 2020 to meet global oil demand but so much production was unlikely. Climate mitigation to the rescue?

In October 2001, the Royal Dutch/Shell company launched its long-awaited global energy scenarios. They contained some big surprises. At the launch, Shell's chairman Phil Watts told the world that it must prepare for the end of the Hydrocarbon Age - not because of oil's imminent decline (this was not mentioned) but because climate-friendly alternative energy technologies will "win over consumers in the coming decades." Oil which now provides 40 per cent of the world's primary energy will fall to barely 25 per cent by 2050. Natural gas, plus renewables and low-pollution, high-efficiency consumer technologies, will pick up the slack as oil's dominance dwindles.© World Resources Institute So with one stroke of the wand, the world's second largest energy company has conjured up the win-win solution of all time. Technology will take care of oil's decline while massively cutting greenhouse gas emissions. And, the real master stroke, no need to worry that oil's decline will create supply shortfalls, price surges and economic havoc - not least amongst oil company share prices.

With just a touch of cynicism, let alone bemusement, one can only hope and pray they are right.

Gerald Leach directed the energy programme at IIED from 1974 to 1988. He is now a Senior Research Fellow with the Stockholm Environment Institute, working on various aspects of energy, environment and development.

This article is reproduced with permission from the December 2001 issue of Tiempo, published jointly by the University of East Anglia, the Stockholm Environment Institute and the International Institute for Environment and Development (IIED). For more information on oil see: