Michael C. Lynch and the 'False Threat of Disappearing Oil'
Posted by nate hagens on August 27, 2009 - 10:02am
This is a guest post by Kevin Rietmann, aka The Dude, on The Oil Drum. The post is a response to Tuesdays NYTimes Op-ed by energy analyst Michael Lynch, showing some historical track records between some 'peak oilers', Mr. Lynch and others.
Michael C. Lynch and “the false threat of disappearing oil.”
Energy analyst Michael C. Lynch's op-ed piece “‘Peak Oil’ Is a Waste of Energy, published in the August 24th edition of the New York Times, has naturally garnered much attention from those in the peak oil community itself. Lynch's dim estimate of predictions of declining supplies of energy is evident from the title of his piece, and those familiar with views espoused by him in the past weren't surprised by this new batch of commentary:
A careful examination of the facts shows that most arguments about peak oil are based on anecdotal information, vague references and ignorance of how the oil industry goes about finding fields and extracting petroleum. And this has been demonstrated over and over again: the founder of the Association for the Study of Peak Oil first claimed in 1989 that the peak had already been reached, and Mr. Schlesinger argued a decade earlier that production was unlikely to ever go much higher.
Indeed forecasts of the demise of oil in the past have often been far wide of the mark, as documented in Lynch's own 1998 piece CRYING WOLF: Warnings about oil supply. In this paper he shows how prominent peak oil advocate Colin Campbell's forecasts from 1989 and 1991 were inaccurate, which they were:
One bit of data which would have been of value to include in this paper would have been who Campbell was being compared to in the first place. Likely guesses are that “CERI” is the Canadian Energy Research Institute, “EMF 11 Avg” is the Energy Modeling Forum, and “IEW” is the International Energy Workshop. As the graph shows, Campbell's forecasts were the most inaccurate of this sampling, although whether other forecasters were even further off the mark for predicting production – or price, which wasn't included in this broader comparison – isn't remarked upon; but then this was a short piece, whose contents were intended to dismiss the then-current warnings of impending irreversible decline in oil supply coming from both Campbell and Jean Laherrère, who were recent collaborators on the March 1998 Scientific American article The End of Cheap Oil, the notoriety of which had undoubtedly reached Lynch's ear; many look to this publication as marking the onset of the modern era of discussion of peak oil, including the coining of the term “peak oil” itself a few years later.
Having dismissed Campbell's credibility with his article – or so one would imagine, anyway – Lynch then rounds things up with Colin's current batch of forecasts for the early 21st century, contrasted with a handful of others, making sure the reader will be left with the impression that Campbell's projections are in no way to be taken seriously:
Lynch (1996) argued that the Hubbert method fails because it takes recoverable (not total) resources as fixed, and assumes that to be the area under the curve of total production. When the estimate of the area under the curve (resources) is increased, the entire increase must be applied to future production. This is exactly what is happening with Campbell, as Figure 15 shows. The errors in his 1991 forecast and the adjustments he has made in his latest work are thus predicted by Lynch (1996). Campbell has not provided an alternative explanation, merely ignored them. And as Figure 18 shows, his forecast is well outside the mainstream.
Short-term prices will certainly fluctuate, and we will surely have more oil crises, since they are short-term events. Unfortunately, there is little doubt that the certain failure of the current Cassandras will be forgotten within a few years and a new round of alarms will be sounded. Hopefully, it will not receive the attention that the current (and previous) ones did, and even more hopefully, most governments and companies have already learned their lesson from the tens of billions of dollars wasted when others cried wolf during the 1970s.
As in his current NYT Op Ed Lynch implores the reader to pay no mind to those predicting a limit to liquid fuels production in any time frame worth considering; as with other sections of the article he includes an easy to digest graph, this of another camp of forecasts, including his own:
So, how have things turned out in the intervening years? No doubt Campbell and his ilk have once again shot far and wide of the mark; of course they will never learn their lesson, but people have been insisting that world oil production will soon irrevocably decline for practically as long as the industry has been around.
Well, let's see: here is a version of Lynch's graph, with bars added by me to delineate increments of 10 mb/d (horizontally) and 2008/2009 (vertically), which weren't in the original:
This is of crude + condensate (C+C); 1998 average production was 66.96 kb/d according to the EIA's International Energy Statistics. As we can see Lynch, perhaps in a burst of confidence, was foreseeing the highest level of any on this graph, which, besides Campbell, included quite sizable government institutions: the EIA, IEA, and World Bank. One might class these as optimistic by nature; thus Lynch would trump even them in a burst of glory.
The trouble is, of course, that it was Lynch that aimed wildly off course here, and it is his nemesis Campbell that appears to have hit the bullseye. Time constraints prevent me from digging up the original documents that these forecasts are derived from, but we can tell enough by simply eyeballing the chart. The plateau of supply Campbell was predicting hasn't come to pass, but if it had it would yield 2008 production of ca. 69 mb/d; actual C+C levels for 2008 were, on average, 73.79 mb/d according to the EIA; a difference of 4.79 mb/d. The World Bank's call looks to be about 79 mb/d, which would equate to 5.21 mb/d diff. No doubt we could settle the niceties of who gets the blue ribbon here with the actual numbers used; but what is incontestable is that Lynch was very far off the mark; his 2008 levels are at least 86 mb/d, more comparable to the actual level of 85.47 mb/d for all liquids – perhaps this was even his intent, to forecast increased use of oil sands and the like, compared with Campbell's pessimistic focus on C+C and little else. But if so the presentation, lacking any kind of elaboration on this point, was misleading, to say the least.
What is more, in the intervening years it appears to have been the peak oilers who have made the accurate production forecasts, not cornucopians or massive government agencies such as the EIA and IEA. This is a chart I prepared from data included in the Oil Drum's Peak Oil Update - July 2009: Production Forecasts and EIA Oil Production Numbers, prepared by Sam Foucher :
The columns show the accuracy of the forecasts at present; the line delineates the year of global peak predicted in the forecasts. As can be seen, calls for near term peaks have overall proven more accurate over the years than those foreseeing peaks decades in the future; amusingly enough, a peak based on the simple stability in historic per capita oil consumption (“Population based”) has proven more accurate than that of any of the government agencies. A remarkable aspect of these results is that, while the median date of publication of these documents is 2006, the most accurate call for 2008 was from the 1998 paper of peak oilers Duncan and Youngquist.
Some obvious caveats suggest themselves immediately; for instance, perhaps more optimistic researchers than those documented here made even more accurate calls, possibly from even earlier dates. For this we can turn to a Canadian commentator on energy issues, Freddy Hutter, who is quite optimistic as regards to the future supply of hydrocarbons. He has a “Prediction Scoreboard” on his page of Peak Oil Depletion Scenarios, which includes forecasts from a broad spectrum of researchers; perhaps one of Lynch's colleagues is way out in front? Or perhaps not?
Using projections made 9-14 years ago, Jean Laherrère earns bragging rights for the most accurate forecast for 2008 with a calculation that was within 1-mbd of the final tally. Looking ahead using Year-to-date figures for 2009 and short term projections, the USA's EIA is poised to garner the long term prediction crown for 2009 & may share 2010 with Duncan & Youngquist.
It would behoove the editors of the Times to inform their readers of these fundamental flaws in Lynch's analysis; in the forecasting department he has stumbled more than a bit badly, and those he has denigrated in the past – and present – have made long term predictions of conditions in the energy markets that leave Lynch's past work – and thus, we surmise, his present work as well – throughly in the shade: indeed, perhaps little better than that “based on anecdotal information, vague references and ignorance of how the oil industry goes about finding fields and extracting petroleum.”