In the not-yet-named era in which we now live beyond the Cold War, a philosophically edged disagreement has arisen among our literati and social seers as to just how dangerous the world really is. Much of the discussion seems to turn on temperament or historic intuition rather than evidence. But of all the dangers cited by the more pessimistic among us--unconventional weapons proliferation, clashes of civilizations, global warming and environmental despoliation, rampant ethno-nationalism, and the rest--one almost never hears of the possibility of a major, world-wrenching energy crisis. And yet such a difficulty is at least as likely as any of these other would-be terrors. Given the false alarms that have been raised in the past, this assertion is certain to be received with considerable skepticism. It is an assertion I intend to justify in what follows.

The world is using up its geological endowment of oil at a prodigious rate, and that rate will increase as newly wealthy countries, particularly in Asia, enter the industrial phase of economic growth--as indeed they will, recent hiccoughs notwithstanding. At the same time, and despite astounding advances in the science of geology and in techniques of finding fossil fuel deposits, discovery rates of new oil reserves are falling sharply. For every four barrels used, only one is found. The lines of discovery, consumption, and extraction are bearing down on one another and will inevitably cross, probably in the year 2003; at that point, the world will pass its peak production of oil, meaning that more than half of the world's finite supply of conventional oil will have been extracted and consumed. When that happens--and 2003 is only five years away--both the developed and developing countries of the world will increasingly be faced with politically difficult and, ultimately, hugely expensive problems of adjustment. The international political problems facing the United States will be shaped in particular by the fact that as the world proceeds down the slope of oil production, five Middle Eastern countries will increasingly gain relative share. The last time that happened, in the 1970s when their share was about 36 percent, there was, in point of fact, hell to pay. Soon--in all probability by the year 2000--that share will stand at around 30 percent.

Before proceeding to show why we will soon come to and pass peak production and to suggest what the implications may be, we might ask why it is that so few people seem concerned about the energy balance. Two mutually reinforcing reasons come to mind: one broadly historical, the other having to do with political psychology.

The Déjà Vu Factor

The first reason not to worry is that we are thought to have already "been there and done that." The oil shocks of the 1970s, which had profound if temporary economic consequences and alerted the world to the issue of resource depletion--as embodied, for example, in the pronouncements of the Club of Rome--are widely considered to have been assimilated and put to rest. All things considered, the energy crisis turned out to be short-lived and, as Americans even remotely educated in the matter know, the price of gasoline at the pump is no higher today, adjusted for inflation, than it was before 1973.

Common wisdom has it, too, that the shocks were largely artificial, by which most people mean politically induced. There is much truth in this. In 1973 certain Arab oil-producing countries restricted the sale of oil for a few weeks as an economic weapon in the Yom Kippur War, and prices escalated five-fold. During 1978-79 the fall of the Shah prompted panic buying that drove prices briefly to almost $50 a barrel. But as politics can turn sour, so can they turn sweet. Saudi Arabia's economic and political interests have largely coincided with those of the United States since the late 1970s. As the major swing producer, with large reserves and excess capacity, the Saudis have helped push the price down and keep it down for most of the last fifteen years. Meanwhile, the likelihood of another Arab-Israeli war on the scale of 1973--providing a pretext for another embargo--has been rendered remote by the Egyptian-Israeli peace treaty of 1979 and further achievements in the peace process since then. Even the 1980-88 Iran-Iraq War, it is often noted, failed to have a major influence on oil production and export despite the fact that the battlefield was literally only an artillery barrage away from major oil terminals and facilities.

A form of economic common knowledge has reinforced the complacency afforded by these broadly political observations. As everyone knows, new industrial efficiencies and efforts at conservation have helped ease the energy problem for the longer term, and many believe that a shift away from old-fashioned industrialism in much of the world will reduce even further the ratio of energy input to productive output. There have been significant new discoveries of oil since 1973, too. Many prominent economists have therefore come to act as though the world's oil endowment is effectively infinite. When the world runs low on "conventional" oil--that is to say, easily recoverable oil--it is assumed that price signals will stimulate efforts to recover more expensive reserves. Those same price signals, in turn, will stimulate research and development into other energy sources and into technologies that can lower the price of non-conventional fossil fuel recovery--for example, the process of extracting oil from tarsands and shale. The alarmist view aired during the 1970s, that the world was running out of oil, is now dismissed by most economists and many laymen alike as having been quite silly--an almost perfect example of creating a scare by crying wolf. And in a way it was; in the 1970s the world had used no more than about 20 percent of its conventional oil endowment.

Prevention is Harder than Cure

Human psychology mixes with the logic of democratic politics to produce the second reason for downplaying energy issues: namely, that it is much easier to react to a diplomatic crisis or a socio-economic discontinuity than it is to prepare for one. If government analysts were to predict an oil crisis or a market convulsion leading to a doubling of price within three or four years, then the governments for whom those analysts labor would be forced either to reject expert advice or to make painful choices about how to cope. It is much easier to ignore the advice, or, better still, to ensure that one never hears it. One may then treat the crisis, when it comes, as an act of God or some other unpreventable occurrence, and--a crucial determinant of behavior in democratic politics--with any luck at all, it will all come crashing down on someone else's watch.

Western governments have, in fact, gone out of their way not to know what the position is regarding the future of oil supplies. Let an anecdote illustrate the point.

I live in a remote village in France called Milhac. It is so remote that most Frenchmen have never heard of it, let alone visited it. And yet last May I had a visitation from the International Energy Agency (IEA), an arm of the OECD set up in 1974 in the wake of the first oil shock. It seems that the IEA had taken an interest in my writings about oil depletion and, not having anyone on their 150-strong staff who has addressed the issue of the fundamental resource constraints, the mountain, so to speak, came to me in Milhac. It soon became evident that, in the two decades of their existence, they had not even begun to study the related issues of oil endowment, consumption, and discovery trends in any depth. Only in a transnational bureaucracy completely protected from the pressures of reality could such a thing happen. For more than twenty years they have instead simply accepted published reports on existing and probable reserves at face value, even though, as I show below, there are plentiful reasons for doubting their accuracy. Made up mostly of economists, the IEA relies for intelligence on short-term forecasts of supply and demand, and does little to evaluate the underlying resource base. The United States and most West European governments, in turn, rely on the IEA for many of their basic estimates. As William A. Weingarten, director of the Office of International Energy and Commodities at the U.S. Department of State, explained to me, "We believe . . . that the price signals generated by an efficient world energy market will be important in making the transition and we are working with other members of the IEA to try to improve the information available to market participants."

Yes, the oracle of the market does speak, but what it says is of limited use for longer term planning. The oil market tends to fluctuate in response to short-term cues; the current fall in prices, for example, reflects short-term oil stocks' movements and hedging based on warm weather, the Kyoto conference, and the Asian economic collapse. The fall is excessive. And since the market is highly efficient but short term, it will probably overreact in the opposite direction when true shortages recur, thereby exacerbating the resultant price escalation, just as was the case in 1974 and 1978-79.

The general complacency that flows from these two elements of the situation--earlier unjustified panic of the 1970s and the democratic propensity not to hear unwanted news--is reinforced by yet another: namely, that, on the surface and in the relative short-run, the conventional factors that determine the world energy balance look relatively benign. The oil-rich Arab states are running short of cash and will increase their production. A slowdown in Asian growth may dampen demand and also prices. The economic growth that has been occurring in much of East Asia and that may be anticipated in Russia and other former European communist countries is concentrated in industries that are not particularly energy-intensive--and much of Eastern Europe's old energy-wasting industry is shutting down. It may be, too, that the fears of global warming (whether justified or not is a separate matter) will translate into reduced fossil fuel consumption and, therefore, lower demand pressure on price. As with Joseph's interpretation of Pharaoh's dream, the world may indeed have seven or so good years ahead. If so, this will only make the bad years that will follow seem all the more shocking and difficult to bear.

Not Production, But Extraction

For most of this century, the world has relied primarily upon an abundant supply of cheap oil to drive its economy. It is difficult to think of any aspect of modern life, from transport to agriculture, that does not depend upon it. Its depletion is therefore an immensely important subject. The beginning of wisdom concerning this subject is that, despite standard usage, we do not produce oil, we extract it. This circumstance alone implies certain immutable facts about the situation at hand and, given the reluctance to face those facts, it is necessary to repeat them in brief.

Oil comes mainly from the decomposition of algae and is preserved at the bottom of stagnant lakes and marine troughs under certain conditions of pressure and heat. Gas comes from plant remains. As the organic material is buried, it is at a critical point converted by chemical reactions into oil and gas. It then migrates through the rocks to fill such geological reservoirs and traps as may be physically reached. The conditions for the formation and preservation of oil occurred only very rarely in time and place in the earth's geological history. Much oil that was once trapped has since leaked away. One of the factors responsible for the abundance of oil in the Middle East is the widespread occurrence of salt, an excellent seal, above the reservoirs.

Almost all the oil produced to date, as well as what waits to be produced over the next twenty years or so, can be described as conventional. Conventional oil is that which flows at high rates and in good quality, much of it from giant fields found long ago. Ninety percent of oil produced today comes from fields more than twenty years old, and seventy percent from fields more than thirty years old. This is the kind of oil we know most about, and it has been responsible for the supply to which we have grown accustomed.

There is also non-conventional oil. It comprises the heavy oil and tarsand deposits of western Canada, Venezuela, and Siberia; oil from enhanced recovery achieved by changing the characteristics of the oil in the reservoir through steam injection and other methods; oil in hostile environments; oil in small accumulations; and oil of poor quality. The essential characteristic of all these types of non-conventional oil is that they are difficult and slow to produce. They may also be expensive to transport or refine.

We need to know how much conventional oil there is left and when its production will peak. Oil being a finite resource, production starts at zero and ends at zero. It reaches peak somewhere in between, more or less coinciding with the midpoint of depletion, when half the total endowment has been extracted. To assess the total endowment, we need to know how much has been produced, how much remains to be produced from known fields, and how many new fields await discovery.

Oil & Gas Journal, the most widely used publication on the subject, reports that 807 billion barrels had been produced to the end of 1997, almost half of that since the shocks of the 1970s. Reported reserves in known fields are 1.020 trillion barrels, but, as shown below, there is good reason to reduce this estimate to about 830 billion barrels of median probability. If these numbers are reasonably accurate, it means that approximately 1.635 trillion barrels have been discovered to date (produced + reserves). The graph on the following page, which documents the number of gallons of oil discovered per decade, shows that peak discovery occurred during the 1960s. Extrapolating this historical discovery pattern suggests a total endowment of about 1.8 trillion barrels, which is in line with the trend of some twenty-five published estimates over as many years, as well as with other analytical techniques. That means, in turn, that about 165 billion barrels remain to be found, and about 995 billion barrels remain to be produced.

This discovery plot makes a very compelling argument about the resource base of the earth's conventional oil. The discovery rate over the past thirty years has fallen, despite all the technological advances and the huge increase in knowledge gained from worldwide exploration. First came onshore exploration, which found the larger basins and most of the giant fields, which together hold some 70 percent of all known oil. Then offshore development was opened, effectively by the development of the semi-submersible rig in the 1960s. Discovery there, too, has now peaked. Attention is now turning to deep offshore deposits, but there are only a few deepwater areas that are geologically promising or indeed possible. To date, only about 25 billion barrels have been found in deep water--namely, about one year's world supply, and that from the more promising tracts. More can be expected, but not enough to defer peak by any significant period.

Some people have thought that great new possibilities for discovery would flow from the opening of the former Soviet Union after the fall of communism. It turns out, however, that Soviet explorers were as intelligent as their Western counterparts and were able to find the larger basins and most of the giant fields. Although they were working with rather primitive technology, it was not more primitive than that available to the West in those earlier years when most of its oil was found; after all, reserves in the Caspian area were discovered even before Colonel Drake drilled his famous well in Titusville, Pennsylvania in 1859.

In fact, too, the offshore extensions of this geologically productive area are well known and have been for many years. On the shores of the northern Caspian lies the giant but very difficult Tengiz Field, with reserves of about 8 billion barrels. There is some possibility that other similar finds may be made offshore, although there are doubts whether the critical salt seal is intact. There are also fears that the structures may contain gas and not oil. Additionally, there remain large, poorly known, and barely explored basins in the Siberian Arctic and offshore, but the extreme environment places such oil as they may contain in the non-conventional category. Apart from the resource constraints, doing business in the former Soviet Union has proved difficult for most foreign companies, so that even if its potential is greater than estimated, not much new oil is likely to find its way onto world markets for a long time to come. At best, Caspian oil will be of North Sea proportions--three years' world supply--and while this will help, it does not change the longer term picture very much.

In short, the world has now produced almost half of its geological endowment of conventional oil. The earth has now been so extensively explored that virtually all the prolific producing trends have been identified. Geophysical techniques have also improved radically so that it is possible to map in very great detail the traps for oil that occur within the productive areas. There is, it so happens, a natural geological polarity about oil, reflecting the rare conditions of its formation: it is either there in abundance or it is not there at all.

Measuring Up

Unfortunately, these advances have not been matched by accuracy in reporting the results. World statistics on production and reserves are becoming increasingly unreliable, and this contributes much to the capacity for complacency exercised by government officials, economists, and laymen alike. For example, published reports of production in 1995 (simply a matter of reading the meter) range from 60.48 (OPEC Bulletin) to 69.89 (World Oil) million barrels a day. The situation with respect to reserves is even worse; there are no universal definitions and there is no reliable audit. Huge unsubstantiated increases were announced in the late 1980s as various OPEC countries vied to increase their share of the agreed total OPEC global limit, this because national quotas are based partly on reserves. The size of oil accumulations in traps far underground cannot be measured directly, and the reserves they hold are no more than estimates subject to varying degrees of probability. What is reported ranges widely. Some is so-called proved, with, say, a 90 percent probability of accuracy; some is proved and probable with a 50 percent probability; and some is proved and probable and possible, with, say, a 10 percent probability. Summing numbers with different probability rankings gives statistically flawed results. If one takes the trouble to convert the reported numbers to a 50 percent probability value, namely, the best estimate of what will actually be produced, then one has a firm basis for developing a depletion profile with which to determine peak. Such an estimate must also take into account the impact of foreseeable technological advances. When one does all this and does it properly, the numbers look much less robust than advertised.

There is nothing we can do to change the earth's natural endowment; there is something we can do about how that endowment is measured and reported. But a subject of such importance is a sensitive one, and many vested interests are involved. Both oil-producing countries and oil companies have various reasons either to overstate or understate proven reserves. While there is no shortage of estimates and forecasts, only a few can be described as knowledgeable and objective, and what those few reliable estimates make clear is this: In global terms, and recognizing the uncertainties deriving from unreliable reserve data, we can accept, at least as a strong working hypothesis, that there are about 995 billion barrels yet to be produced. Of this, about 17 percent is yet-to-find. Annual consumption stands at 24 billion barrels a year and rising, and contrasts with the current discovery rate of just over 5 billion barrels a year on a falling trend. On this basis, the midpoint of depletion and corresponding production peak will arrive when a further 93 billion barrels have been produced, which will be in less than five years.

The reaching of peak production is not in itself a disaster; obviously, the sky will not suddenly fall the morning after it occurs. Large old fields of conventional oil in the Middle East and elsewhere will continue to produce, and their capacities will not be affected as lines cross on a graph in the metaworld of geopetroleum analysis. Spot rates for oil will not suddenly and sharply rise, and the multi-year contracts that determine price for most crude oil sold in the world will continue to be determined by standard factors: artificially limited production, especially by swing producers in the Middle East; business considerations; and diplomatic sensibilities. But the world will have passed a point of no return that will ultimately have an impact on supply, and therefore on price. Market signals alone are not the only measure of human rationality, and, in the case of the oil market, parochial short-term interests may well take precedence over the longer term interests of oil producers and consumers alike.

The Thirty Percent Threshold

But that is not the whole story. Equally as important as the size of the earth's total oil endowment is its geographical distribution, and here we must return to the Middle East, where the consequences of reaching peak production may announce themselves even sooner than the year 2003.

The oil shocks of the 1970s carried a particular message about the distribution of oil, demonstrating that a few richly endowed countries could exert a disproportionate control on supply. By 1973, after many years of steady increase, the share of world supply coming from just five Middle East countries--Iran, Iraq, Kuwait, the United Arab Emirates, and Saudi Arabia--had risen to as much as 36 percent before falling off as new fields in the North Sea, Alaska, and elsewhere began to deliver growing supplies. These latter discoveries were made before the shocks and were not a response to them, as is often suggested, although it is true that high oil prices did prompt their more rapid development. By 1986, the Middle Eastern share of world production had fallen to 16 percent, and prices had fallen to a low of $6 a barrel. The price collapse was partly engineered by the United States, in cooperation with Saudi Arabia, as part of its policy to undermine the Soviet Union by economic means.

Since then, however, an important change has occurred: share from the five Middle Eastern countries has again begun to rise, reaching 27 percent of world supply by 1996. This time, however, it will continue to rise because, as shown above, there are few if any new major oil provinces waiting to be discovered. Certainly, there is no prospect of major early-flush production from new giant fields, save perhaps the Caspian, an area obviously not free of serious political and contractual risks. Accordingly, we must ask: What will happen when Middle Eastern producers reach the critical threshold that will enable them to virtually set the price of oil? It is a more serious prospect than it was in the 1970s because it is not likely to be temporary this time. Nor has the question been made any less serious by the benign changes in Israeli-Arab relations since 1973. The Palestinian issue was never much more than a convenient pretext for oil-producing Arab Gulf states; remember, after all, that OPEC's charge on the pocketbooks of the Western world in 1974 was led by the Shah of Iran, whose country had quite cordial relations with Israel and an alliance with the United States.

About half of the world's conventional oil endowment that remains to be extracted lies in the five major Middle East oil countries listed above. On current trends, assuming a 2 percent annual increase in world demand, their share of world production will pass 30Â percent by the year 2000, and that is the threshold that made possible the price shocks of the 1970s. No one knows what this may mean in terms of oil price, but it does not stretch the imagination to conclude that the scene for another radical increase will have been set by the turn of the century, and most certainly by a decade after that. Political motives for an increase may not be necessary--shortage alone might suffice. But in that part of the world, political factors should never be discounted, and the political and social fragility of Saudi Arabia, far and away the most significant producer, is reason for real concern.

A massive increase in price, were it to occur, would no doubt curb demand, especially in the developing world. This would most likely give rise to very volatile prices for about ten years, at the end of which the Middle East producers would be supplying as much as 50 percent of the world's oil, and would be close to their own midpoint production peak. Absent sharp declines in consumption, literal shortages will be inevitable a decade or two after that, as production of a finite resource starts its inevitable decline.

Since these numbers are not and cannot be exact, there is some latitude in the timing and scale of the consequences--but the essential picture is clear enough, at least to those who are willing to look. Strangely, though, not many people are willing to look, being content instead with the bland pronouncements of the oil companies and the mistaken views of economists, who have difficulty accepting the very concept of depletion. Accordingly, energy policy to meet this historic discontinuity affecting the life-blood of the world economy is far from the top of most countries' agendas.

As I have argued, this complacency owes much to the widely held view that there is always a technological solution to scarcity and that more oil will be recovered from existing fields if price signals touched off by impending scarcities make it economically worthwhile. The industry has indeed recorded remarkable technological advance, its knowledge of oil geology has made huge progress, and no one has ever accused it of not understanding how to make a profit. But that is, paradoxically perhaps, really the point: The only way to reconcile these considerable achievements and the high level of worldwide exploration with the sharply falling discovery rate is that there is much less left to discover. On present well-founded trends, to find significantly more than is estimated here would take a very long time and require an astronomical number of wells. And even adding a hypothetical 500 billion barrels to the yet-to-find category would delay peak by only ten years at current and projected rates of consumption.

Remember, too, that most technology aims to increase production rate, and with it profit, without necessarily adding much in terms of reserves. To understand the reports that are made about the very real technological achievements we have lately seen, it is important to keep a sense of scale. For example, the North Sea, the largest new province found since the Second World War and one which relies heavily on high-technology for extraction, has a total endowment of some 60 billion barrels--enough to supply the world at current consumption rates for less than three years. That its production is about to peak is common to almost all studies, and new technology will not make much difference as to when precisely that peak occurs.

Non-conventional oil will undoubtedly make an increasingly important contribution after the peak production of conventional oil has passed, but due to the long lead time and the slow rate of extraction, it cannot impact peak production or the growing control of share by the Middle East's "big five." There is indeed great scope for new technological solutions to overcome the impending oil crisis, but they will have to be aimed at finding new ways to use less oil-based energy rather than in trying to find what is not there to be found.

Man arrived on the planet some two million years ago and lived sustainably on it until about 1850, when he numbered about one billion. Since then he has multiplied six-fold and the growth in oil extraction has increased in direct proportion to it. Since oil has provided abundant cheap energy to sustain the human population increase--particularly over the past half century--it can hardly be doubted that the imminent peak of oil production will mark the beginning of a historic discontinuity that will change the world we know and the energy economics that drive it. We are not running out of oil, just out of the cheap abundant supply on which we have come to depend.

There are many long-term solutions to the problem engendered by the depletion of cheap oil in terms of renewable energy sources and, above all, of using less of the endowment we have. Our main concern, however, must be about the transition from the era of cheap oil to whatever replaces it, a transition that will bear with it many grave economic and political consequences. These cannot be avoided entirely, but they can be mitigated with a modicum of planning. A proper understanding of the resource base is critical to any sensible political response. But a proper understanding seems to be precisely what governments are lacking, and these days are laboring actively to avoid having. Fundamentally the problem is not the greed of a few politically recalcitrant Middle Eastern countries that were especially favored during the Jurassic period--something that in the last resort could be solved by sending in the Marines. Rather, it is the immutable physics of the reservoir.