A Free-Market Energy Blog

More on Peak Oil

By -- November 2, 2009

The recent Houston Chronicle op-ed, ostensibly written to respond to my New York Times op-ed, is worthy of reading for a variety of reasons, but primarily entertainment. The reference to me as Stephen Lynch was apparently an editor’s error, but the analogy of oil fields and glasses of water was quite enlightening as to the state of the debate. The three gentlemen comment on the difference between a straw in a glass (a supergiant field) and a puddle of water on the table requires many straws.

In fact, I know of no supergiant fields that have not required many straws, since oil fields are not ‘pockets’ of oil but rather oil that is in rock, rather as water is in a sponge. Drawing all of the fluid from one spot doesn’t mean that all of the oil will flow freely and uniformly to the straw: to the contrary, a given well usually drains a very limited area, and supergiant fields typically have numerous wells, hundreds even thousands, depending on the geology and geography.
[The inappropriate use of analogy is reminiscent to the website of Colin Campbell, the founder of the Association for the Study of Peak Oil. He points out that when you have finished half the glass in a beer, you only have half left. Given that he lived in Ireland, this prompted the rejoinder that his inability to find another glass of beer should raise questions about his understanding of resources.]

There is also the rather illuminating comment that not knowing much about Russian and Middle Eastern supergiants suggests that they could decline much faster than we expect. And yet, couldn’t they also decline much slower than we expect? This selective attention is what is known as ‘bias’. In fact, while it is not possible to download comprehensive information about oil fields in those areas, I have done work in the past by relying on such information as is available, including the piece, “The Economics of Petroleum in the Former Soviet Union,” in Gulf Energy and the World: Challenges and Threats (The Emirates Center for Strategic Studies and Research, Abu Dhabi, 1997) as well as “Crop Circles in the Desert” (on this website), which show what can be done, given a little effort.

Finally, the issue of cost and the increasing difficulty of producing oil is raised. This is an interesting and pertinent matter, but one that is poorly understood. Many of the recent studies arguing that oil prices must remain near $70/barrel because that is the new ‘marginal cost’ of production suffer from two major problems: taking transfer payments such as royalties as costs, and assuming that the cost increase is due to geological effects, not cyclical in nature.

As an example, cement and steel make up a large portion of the development of any big field, especially offshore. Those costs rose sharply in the past few years, but there is hardly any danger of either being ‘depleted’. Indeed, since 1900, cement prices have fluctuated repeatedly, but have actually tended to decline over time. Steel prices (only available from 1940) have risen in the past few years before declining with the recession, but this was clearly due to the economic boom in Asia tightening markets. They had actually been on a long-term decline from 1980, dropping fairly steadily from $180/ton to $100 before the recent run-up.

And the very fact that costs have appeared to double (or more) in just three to five years should be suggestive: depletion has hardly increased during that period, nor has there been any sudden change in drilling targets. Most of the deepwater fields now being developed were found before the price/cost run-up, and the expensive projects like Sakhalin and Kashagan were begun when prices were below $50. To the extent that more expensive fields are now under development, higher prices are the explanation, not depletion. Operators are free to target marginal projects when prices are high; if prices drop, they will be ignored, bringing costs down.


  1. Andrew  

    You’ve left a hyperlink tag open somewhere.

    Very informative, though.


  2. Kevin Rietmann  

    Hi Mike,

    In your initial response critics of your NYT op-ed you linked to my piece on the Oil Drum, which was wholly concerned with your late 90s forecast of supply, which has proved somewhat wide of the mark, in contrast to Colin Campbell, whose contemporary projection has come closer to actuals. I’d hope your riposte you’d have something to say about this, but instead you focused wholly on price forecasts, indeed even stated that

    The peak oil community tends to focus on price forecasting not supply forecasting, because prices are much more unpredictable, subject to both political and psychological factors (including ‘momentum trading’ the bubbles’ favorite).

    Would welcome some input from you about those 10 year old calls. I hope you haven’t filed me in your ad hominem catalog, I tried to maintain a semblance of neutrality in my writing. I am a “neophyte,” fwiw. But this was an obvious enough bone to pick. I did welcome the opportunity to collate more forecasts from either side of the fence to see who has the most solid track record, too.


  3. mlynch  

    Hi, Kevin, sorry to neglect you earlier. The tag was added by the web editor, I had meant to refer to the page with the 200+ comments.
    The point here is that you are comparing crude+condensate production with forecasts of total liquids. In the late 1990s, no one separated them (exception: see below), but recently a few peak oilers like Simmons have begun to focus on that data because it seems to show a peak, whereas the recent boom in global gas has meant a lot more natural gas liquids. If you look at total liquids, you see my forecast was very close (as was nearly all others).
    Campbell, on the other hand, has often used ‘conventional crude’ as a subset, to avoid dealing with deepwater, tar sands, etc., but he defines it differently from nearly everyone else (excluding Arctic oil for instance) and has changed his definition of deepwater. This makes it hard to get actual data (Siberian oil, for example), but also makes it hard to compare his various forecasts.
    Also, be advised that global production forecasts are really consumption forecasts in most cases. Looking at OPEC, non-OPEC, or Middle East and non-Middle East is usually more instructive.


  4. Kevin Rietmann  

    I wondered about the link in your piece, since you described it as a forum for general comments, whereas the article I submitted was specifically about that 1998 paper of yours. Others suggested that you were forecasting all liquids as well, but mixing them in on a graph with Colin’s C+C is a bit misleading to say the least! Just fit in “(C+C)” in the legend next to “Campbell.” I seriously don’t think an editor would’ve been that stingy.

    Was he imposing all of these restrictions at the time? It has it uses, to focus on restricted segments of upstream development, but in the end oil is oil. NGLs aren’t going into auto fuel tanks though, unless you’re outfitted to run on propane. Ultimately the more data sets thrown into the pool the better, I say. One example from peak oilers I find a bit disingenous is the backdated discovery profiles – again, there’s a perspective to be gained by looking at things like that, but reserves growth is an undeniable reality; would it be such a chore to throw up an alternate chart displaying reserve growth in old fields year by year? Neglecting to do so creates the impression you’re attempting to mislead.

    My piece also briefly touched on the accuracy of various forecasters, as quantified by Sam Foucher, one of the TOD staff members. Got an email from Freddy Hutter, who’s been doing work along those lines at trendlines.ca Some of the regular posters at TOD such as “WebHubbleTelescope” would welcome the chance to lock horns with you, too; drop by if you have some free time and want to spar a bit.


  5. mlynch  

    The figures for Campbell used in the graph you cite appears to be total liquids, since it roughly matches reported production when he published. He has varied quite a bit in his coverage:
    Published/Peak yr/peak level/type
    1989 1989 65.8 Total
    1991 1992 63.0 Conventional
    1997 1999 65.0 Conventional
    1998 2003 71.0 Total
    1999 2003 n.a. Conventional
    2002 2000 63.9 Conventional
    2005 2006 66.0 Conventional
    2005 2007 83.0 Total
    2006 2005 66.0 Conventional

    Reserve growth is a tough question, because there are few data sets showing individual fields. Adelman did some work years ago on the US, but while Campbell and others are incorrect in attributing all of the growth to the use of proved, instead of proved+probable, reserves, it is a factor, so the answer isn’t comparable.
    The USGS used the same data that Campbell and Laherrere did, except they have a lot more time series, and found lots of growth. That is the source of their ‘estimate’. But to show actual growth by region over time is tough.
    I have thought about going on TOD, but what has happened on other sites is that a dozen or more people write lots of pieces, with many mistakes, misrepresentations, misunderstandings, inaccuracies, and some insightful comments. (If you look at the open thread responding to my op-ed on TOD, you might notice some.) Taking the time to respond to all of those would be a full time job.
    I had the same problem with Jean Laherrere. He sent me a lengthy comment, but it was so bad that I would have have to write pages and pages in response. With an academic, pre-web background, I try to be very careful in my comments, but on the web, many people say all kinds of things, and when they prove wrong, just shrug it off.
    Examples: Some people claim they ‘debated’ me and I failed to respond. Possibly, I got worn out, but it’s not my recollection. How much time should I spend trying to check it out? Especially when they use pseudonyms that can’t necessarily be compared from site to site?
    Or, Matt Simmons once emailed me complaining that I had falsely stated we had debated at an SPE meeting in Houston, and I sent him a link to his slides on his website. You would think he would check before making such a claim.


Leave a Reply