"Yesterday's fuel: why demand for oil will fall": Economist front cover.

Economist Leader: “The International Energy Agency (IEA) and America’s Energy Information Administration all predict that demand will keep on rising. One of the oil giants, Britain’s BP, reckons it will grow from 89m b/d now to 104m b/d by 2030. We believe that they are wrong” “and that oil is close to a peak. This is not the “peak oil” widely discussed several years ago, when several theorists, who have since gone strangely quiet, reckoned that supply would flatten and then fall. We believe that demand, not supply, could decline.
extracts: “BP reckons that emerging economies will push demand for oil from just under 90m barrels a day (b/d) now to 104m b/d by 2030. Exxon sees the thirst reaching 113m b/d by 2040, a growth rate of around 0.8% a year.”…..With cars getting 3-4% more efficient each year, and trucks improving at about half that rate, analysis by Citi, a bank, shows demand in 2020 3.8m b/d below what it would be without such efficiency gains.
….Filling up with natural gas is sure to spread as fracking and other new production techniques pick up beyond America; using gas to drive vehicles will in many cases be easier than liquefying it for export or putting it into pipelines. This could slice another 3.5m b/d from oil demand by 2020. Copious gas is also being used instead of oil in places where oil is still burned to generate electricity, such as the Middle East, potentially putting 3m b/d more oil onto global markets. If nuclear power takes off in the region—two plants have been started in the UAE—that will displace oil demand, too. Reduced subsidies for oil use, common in oil-producing countries but increasingly unaffordable, will also dampen demand. Citi, using the most aggressive assumptions about gas substitution, calculates that oil demand could peak at less than 92m b/d in the next few years, far below what the supermajors expect.
….Half the supermajors’ long-term capital spending now goes on costly unconventional or deep-water oilfields, largely because production-sharing arrangements and licences to drill in the NOCs’ backyards are increasingly hard to find.
….Exxon has been the world’s biggest spender on exploration and production since the mid-1980s. This year, according to Barclays, a bank, PetroChina will take its place.
….Poor choices and increased competition may explain deteriorations in the supermajors’ reserve replacement ratios (RRRs), a measure of the amount of oil discovered compared with production. In 2012 total hydrocarbon replacement (including gas) at Shell was a slender 44%. BP’s was 85% and Total’s 93%; that means reserves at all three are shrinking. Exxon’s RRR, which has not fallen under 100% for decades, was a more comforting 115%, and Chevron’s was 112%. But of Exxon’s 1.8 billion barrels, high-cost shale oil from the Woodford and Bakken fields in America accounted for almost 750m. Around 50% of Exxon’s reserves are now in heavy, unconventional or deep-water oil, compared with 17% in the early 2000s.
….The supermajors are now spending $100 billion a year between them on exploration and production. But this level of effort has not impressed investors; their share prices (with the exception of Chevron’s) have been flat for years. Nor has it yielded net new oil; their output fell by 2% between 2006 and 2011. What it has delivered is greater gas production, a likely harbinger of things to come. The supermajors are finding themselves increasingly in the gas business. For most of them gas is currently more than 40% of their production—for Shell and Exxon it is more than 50%.
….It will be an unhappy thought to many, but BP’s travails in the wake of the Gulf of Mexico disaster may be guide to the supermajors’ future. Forced to sell assets to raise cash to pay fines, it has found that those sales are often followed by a rise in the company’s share price. This suggests that investors like the idea of smaller, fitter oil firms. Rather than push towards ever more esoteric frontiers, the supermajors might do better to slim down and turn away from the oil that they prize so highly but that the world may no longer want ever more of—and that others can exploit equally well. They will find this hard, though. “Oil supermajor” has a much better ring to it than “fairly large gas producer”.”
Extracts from a letter to the Economist by Dr Roger Bentley on 8 August:
As you may be aware, a range of past articles in your newspaper has indicated a rather poor understanding of the global oil situation. Here I mention just three:
a). In March 1999 you ran a cover titled ‘Drowning in Oil’, with the associated article saying in effect: ‘the world is awash with oil, and likely to remain so’. The oil price at that time was $10/bbl, with a fall to around $5/bbl seen as likely.
By contrast, scientists at that date knew that the opposite price expectation was true; that the oil price would almost certainly rise considerably in the near term.
This view was based on a range of forecasts from the late 1970s onwards – from the oil majors, the World Bank, Petroconsultants and many other organisations – that had assessed the global likely initial volume of
recoverable conventional oil (the global ‘URR’) to be in the range 1,750 to 2,500 billion barrels, and hence calculated that a peak or plateau in the global production of conventional oil would occur roughly around the year 2000.
Of course more data were available in 1999 than in the 1970s and ‘80s, but the estimate for the initial global volume of easily-recoverable conventional oil had hardly changed (and indeed remains similar today).
Thus far from expecting a price fall, your 1999 article should have expected a steep rise in oil price. As you are well aware, since 1999 the real-terms oil price has risen seven-fold.
(It may seem surprising that there has been little change in the estimated global URR for conventional oil between the late 1970s and today, given the many new large oil discoveries in places like Kazakhstan, pre-salt Brazil, offshore Mozambique, etc. The explanation lies in the fact, insufficiently recognised, that the peak of discovery of conventional oil in new fields occurred in the mid-1960s, nearly 50 years ago, with new-field discovery declining ever since. As a result it was easy enough back in the late 1970s to extrapolate the past discovery trend to estimate expected future discoveries (oil that in fact has now largely been discovered), and hence to arrive at an estimated global URR that has changed little since.
Moreover, the decline in new-field discovery resulted in a point being passed, around 1980, when new-field discovery fell below annual production.
Even allowing for technology gain, it is probably fair to say that since that date, now nearly 35 years ago, oil production has represented an inexorable draw-down of the global conventional oil inventory.)
b). The second Economist article I wish to highlight described a small meeting I convened at Imperial College in 2001 to bring together some of the world’s experts to discuss future oil supply.
This meeting was covered by one of your reporters, and his write-up in your newspaper largely mocked the views of the gathering, seeing these as wildly alarmist. He did however hedge his bets, ending the article to the effect that: ‘If these speakers turn out to be right, we had better all look out.’
c). The third article was a short piece written after it had become fairly clear that the UK’s oil production had peaked. Your writer recognised the impact of this peak on the UK economy, but consoled your readers with the fact that ‘still about half of the UK’s total oil remains to be extracted’.
I wrote your newspaper a letter explaining that this ‘half-remaining-after-peak’ is the normal pattern for basins, and hence the oil industry’s ‘proved-plus-probable’ data indicated that a ‘mid-point’ peak would also soon occur in global conventional oil production. (Note that this idea was far from new: a report to the UK government in the 1980s had warned that the UK peak would likely be followed shortly by the global
It is my understanding that your newspaper’s history of poor comprehension of the global oil situation was in part due to the strongly held view of a previous editor, who had learned the mantra that ‘all oil forecasts are wrong’ because such ‘limits to growth’ arguments fail to understand economic paradigms.
And I accept that this view, though quite erroneous, was almost universally the norm among most energy commentators for perhaps 30 years.
(However, the explanation for this error in view is simple. In the 1970s and ’80s it was widely thought that global oil would soon be in short supply; indeed might ‘run out’ in a little over 30 years; see for example
President Carter’s ‘Moral Equivalent of War’ speech. This analysis was largely based simply on the data for the then-size of global proved reserves. By contrast, as explained above, the scientific evaluation of the
situation in those years was based instead on the estimated global URR, i.e. summing past production plus proved-plus-probable reserves with reasonable expectations for yet-to-find and technology gain. Coupled with ‘mid-point’ peaking, this URR indicated that global conventional oil production could continue to grow strongly until around the year 2000, and only then reach a maximum and start to decline.)
I had hoped that given the large number of academic papers and books now covering peak oil that newer editors and staff at your newspaper (which, incidentally, I read regularly, and except for this one topic hold in high regard!) might by now have obtained a better understanding of the global oil situation.
Hence my disappointment in reading the articles in your August 3rd issue.
2.  Specific comments on the Leader: It is ingrained in me unfortunately as an academic to offer comments on articles I read. I therefore take the liberty here of commenting on your
leader article, as follows:
a). ‘Since then demand for oil has, with a couple of blips in the 1970s and 1980s, …’
– As the lower graph on page 21 of your same issue nicely illustrates, the smaller blip (fall in demand) correlated with a real-terms oil price of $40/bbl; and the bigger fall (and also the decade-long recession that followed) with the real-terms $75/bbl price. Although some aspects of the global energy situation have changed since this date, these correlations are a strong argument that a high oil price damages economies and impacts demand, as we have also seen so clearly since 2008.
b).   ‘… widely discussed … strangely quiet…’
– Not true at all. There are ever more excellent books, reports and academic papers on peak oil; see the references below.
c).  ‘We believe that demand, not supply, could decline.’
– It is certainly true that a variety of pressures might reduce oil demand, certainly in future. But as explained above the current driver is supply limitations on conventional oil production pushing up price, hence reducing OECD demand and making global demand to increase only slowly.
– Note that the process of calculating the date of the conventional oil peak requires access to oil industry proved-plus-probable data, and either ‘mid-point peaking’ or bottom-up by-field calculations. Both approaches indicate that we are at or fairly close to the peak in the global production of conventional oil.
– For discussion of the rates that non-conventional oil might – or might not – be able to take up the slack as conventional oil declines see some of the references below.
d).  ‘… shale gas billows from the ground, …’
– Not at a profit today it does not (see US company tax filings). You are right to see gas as only offsetting ‘a few million barrels of oil a day’ by 2020, where this must be set against the 2 to 3 million barrels/day loss
per year of conventional oil production once global peak is past; i.e. totalling a loss of possibly 10 to 15 million bbl/day of conventional production by 2020.
e).  ‘ Citi … Ricardo’
– I know of the Ricardo study and have heard of Citi’s, but have yet to study them in detail. I am guessing, but it is probable that like yourself, and until very recently the ‘mainstream’ oil forecasters such as the IEA,
they do not appreciate the significance of oil peaking, and so do not factor this into their modelling.
f). ‘… now cost $100 [/bbl] to extract.’
– Quite so. It is because cheap conventional oil is now past peak in so many regions (for example, in over 60 countries) that the oil majors are going to ultra-deepwater, the Arctic, and light-tight.
– And be very careful about the Middle East’s ability to pump: their numbers are uncertain, but much indicates that caution is required on their future volumes.
– Note that probably the easiest source to visually examine data for past and predicted oil and gas production by country, based on industry ‘2P’ data, both for countries past their conventional oil peak and for those where peak is still in the future, is: www.globalshift.co.uk (click on the top banner of regions to see the graphs).
3. An offer to come & talk to those of your staff (and possibly also at The Economist Intelligence Unit) who might be interested.
As the references show, for a very long time I and many other scientists have been making the case for the proximity of the global conventional oil peak at around the current date (2005 to 2015). If it would be useful to your staff, I am prepared to spend between perhaps an hour and half a day
with those of your staff that might be interested, setting out the above arguments in more detail and answering questions that might arise. (Note that some of my scientific colleagues in this field say that offering to discuss peak oil is a waste of time; people’s minds are too firmly set. But I am ever the optimist in such matters.)