July 11, 2005
How to talk to an economist about peak oil
I for one would like to see better communication between economists, geologists, and petroleum engineers about the timing and consequences of the eventual decline in global annual production rates of crude petroleum. In part the failure to communicate better with each other stems from differences in the language, assumptions, and paradigms with which those of us from different specialties approach this issue. As one small step toward bridging that gap, I'd like to lay out for noneconomists a few of the key aspects of how economists might think about peak oil.
Suppose you told me that, as a result of a careful examination of oil reservoirs, you were certain that annual oil production was just about to plummet, and would be 30% below its current level in two years. I realize that's a more extreme example than anybody is advocating, but perhaps you'll bear with me in examining its implications for a few minutes before turning to a more subtle scenario.
Let's see if we can first agree on how society ought to respond to these facts that you would be giving us. I would say that the first thing we should do is curtail current oil consumption drastically. Oil is going to be an incredibly valuable commodity in two more years, and we've got to stop wasting it now. By leaving more of the oil in the ground now, we could stretch out the time available to us for developing alternative sources from oil sands and coal and to make radical changes in our transportation systems. And we would need to start immediately making huge investments in those alternatives.
Now let's consider what would happen if the government doesn't make any policy response. Oil is going to become extremely valuable under this scenario in a very short period of time. Let's say for discussion we're talking about $200 a barrel two years hence. Then I would like to make the observation that, if the facts were indeed as we just conjectured, oil surely could not continue to sell for $60 a barrel today. Anybody who pumps a barrel out of a reservoir today to sell at $60 could make three times as much money if they just left it in the ground another two years before pumping it out. The same is true for anybody with above-ground storage facilities-- they're throwing away money, and lots of it, for every barrel they sell at $60 that they could have instead stored for two years and sold for $200. If oil producers did respond to these very strong incentives by holding back oil from today's market, the effect would be to drive today's price up. This profit-seeking wouldn't drive the price all the way up to $200, because you have significant interest, storage, and idle capacity expenses from trying to wait around a couple of years before getting your profit. An economist would expect the end result of this profit-seeking to be that the price today is lower than what it will be in two years by an amount that reflects these interest and other expenses, but certainly far less than the difference between $60 and $200 a barrel.
Suppose, again for sake of discussion, that the outcome of this profit-seeking behavior by oil sellers was to drive the price of oil to $180 a barrel today, (that is, supposing that $180 plus two years worth of forgone interest equals $200). What effects would that have? For one thing, it would be a very powerful and effective incentive to force today's users of oil to reduce their consumption immediately. It would likewise be a very powerful incentive for investing heavily in oil sands and alternative technologies. And, of course, it would leave us more oil in the future to keep the economy going as we make the needed transitions. In other words, the consequence of oil producers trying to sell their oil for the highest price would be to help move society immediately and powerfully in the direction that we earlier determined it ought to move in anticipation of what is going to happen in the future.
I know that many physical scientists feel that economists have a misguided, mystical faith that "markets will always solve everything." Though I understand how outsiders might get this impression, I would guess that more than half of the published research in economics has to do with how the market can misallocate resources rather than how it always does a perfect job. But one thing in which most economists do place a great deal of faith is the powerful forces that are unleashed, for good or ill, by people's efforts to make themselves richer. The argument I'm making here is not an abstract, mystical claim about the market, but rather a very specific claim about the particular matter of interest. The claim is that profit-seeking works strongly in this instance to make the oil price rise now rather than wait until production actually declines, and that this force further works to produce the kind of changes that society needs to make now in order to prepare for the coming production decline.
So, if you thought you were right about the physical scenario, and yet saw oil selling today for $60, how could you explain the situation to an economist, who says that, if you're right, oil should be selling for $180? One thing you might argue is that, in some of the oil producing countries, the rulers have a precarious hold on power, so they just want to pump all the oil they can right now rather than wait a few years, even if the extra profits from waiting might be enormous. That's a good argument to make to economists, one we can understand and listen to. But, we would respond, what about private oil companies operating in market economies? Why would they throw away enormous profits?
Here I have heard some conspiracy- or incompetence-based arguments that frankly are difficult for an economist to follow. But rather than get into those, let me just observe that, even if every single oil producing government and every single oil company in the world had no desire or is too stupid to reap the huge profits that, under the scenario we're discussing, could be theirs for the taking, that still wouldn't be enough to account for the current price structure. The reason, as I've explained here and here, is that you don't need to control a single barrel of oil in order to profit quite handsomely, if what we've conjectured so far were true. If today's price is $60 and in two years oil will sell for $200, anybody with any money to invest could profit enormously by purchasing oil futures or options. And we don't see $200 oil, not in spot prices, not in options, not in futures.
So how could it be that there are billions and billions of easy dollars to be made, and nobody can be bothered to collect them? Unless you have a clear answer for that question, an economist at that point is going to ask whether you're sure that you've got all the facts straight, that oil really is going to sell for $200 a barrel in just two years.
Now, you may protest that you never claimed that oil production was going to fall off as much as 30% and the price was headed for $200 in just two years, that this whole extreme scenario was just a straw man I brought up to steer the discussion my way. Fine, let's say your story is that we'll be at the point we just analyzed in n years rather than 2 years. The point I wish to make is that you shouldn't expect that the oil price will stay stuck at $60 a barrel for n - 2 years, with everybody cluelessly wasting oil and ignoring the need for alternative investments along the way, and then all of a sudden jump up to $180 in year n - 2, for exactly the reasons we just discussed. Somebody would be missing a huge profit opportunity in the price jump from $60 in n - 3 years to $180 in n - 2 years. What economists would therefore expect to see under the n-year scenario would be for the oil price to rise steadily over all these n years, gradually producing greater and greater incentives for the needed conservation and the needed development of alternatives between now and year n.
This particular understanding of the natural consequence of profit-seeking behavior is I think the heart of the issue that needs to be addressed in order for economists and noneconomists to understand each other on this issue.
There's much more that can and should be said to improve communication between economists and others studying this question. I have learned that my readers are a highly informed group representing quite a range of different viewpoints and disciplines. So, at this point, perhaps you'll allow me just to turn the discussion over to you.
Posted by James Hamilton at July 11, 2005 04:31 PMdigg this | reddit
Listed below are links to weblogs that reference How to talk to an economist about peak oil:
» A Real Economist on Peak Oil from The Bit Bucket
James Hamilton, who my roommate says "wrote the book on time series econometrics," on peak oil: I know that many physical scientists feel that economists have a misguided, mystical faith that "markets will always solve everything." Though I understand ... [Read More]
Tracked on July 12, 2005 07:32 AM
» Economists and Engineers: Bridging the Communications Gap on Peak Oil from Green Car Congress
James Hamilton, Professor of Economics at the Univerity of California, San Diego, publishes Econbrowser, and has just posted a piece entitled: How to talk to an economist about peak oil.I for one would like to see better communication between economists, [Read More]
Tracked on July 12, 2005 02:02 PM
» James D. Hamilton on Peak Oil from Deinonychus antirrhopus
James Hamilton has a very good comment about peak oil, I agree strongly with most of what Robert said.... The one aspect of Robert's analysis to which I might add a qualification is the suggestion that oil prices need not rise much until after we're pa... [Read More]
Tracked on August 6, 2005 03:02 PM
» Talk from a rock guy from The Oil Drum
Normally I prefer to post on the events that are carrying us inexorably forward, but cannot resist the urge to respond to the recent article by Econbrowser that Prof G cited. Given that I am prone to the odd comment on economists, and, with your indulg... [Read More]
Tracked on August 17, 2005 08:21 PM
» How to talk to an economist about peak oil from The Oil Drum
A quality piece tonight over at Econbrowser; it captures one of the things we try to do here at the oil drum: integration of the many aspects of the peak oil problem. Here's a snippet: [Read More]
Tracked on August 17, 2005 08:23 PM
» Interesting Reading on oil prices from The Devil's Excrement
To all of those who worry about the price of oil and want to learn more about it, there are a bunch of interesting articles, starting with last week's Magazine article in the New York Times "The Breaking Point" which suggests there will be a huge crisis i [Read More]
Tracked on August 28, 2005 04:39 PM
Where were your theoretical profit seekers three years ago? Oil prices have nearly tripled over the last few years, why are you so sure what just happened is unlikely to happen again?
Doesn't the fact that oil prices have nearly tripled over the last three years suggest that information in the markets for oil is not nearly perfect enough to make your arguments relevant?
Posted by: felixrayman at July 11, 2005 07:38 PM
Can an unforeseen development cause oil prices to change dramatically? Absolutely, as Felix points out, that happens all the time. There's no way a profit-seeker, theoretical or otherwise, can profit from something that's completely unpredictable. But you most assuredly can profit from something that is foreseen and is predictable, and peak oil is, I thought, conceived by its advocates to be a well foreseen near-term event.
Are you suggesting that there's so much uncertainty about when peak oil arrives that, in fact, there is no opportunity to profit from it?
Or perhaps some might suggest that the increase in oil prices of the last few years is itself the first phase of peak oil arriving, not because production has actually dropped (as you know, it's in fact way up), but through an anticipation effect such as I describe in the main post. This to me is a very intriguing hypothesis, but I've been unable to see how to fit it with the observed facts. Maybe others have ideas about how that explanation could work.
Posted by: JDH at July 11, 2005 09:17 PM
All I can say is, "amen, and amen."
That seems to me to be the main problem with all large problems that need to be solved, the solution to the problem requires the integration of many disciplines, economics and the physical sciences being just two of them. (Politics, psychology, etc., etc.).
Cheers to you on an excellent piece. We'll be linking you up here soon.
Posted by: The Oil Drum (profgoose) at July 11, 2005 09:33 PM
A couple of points:
As somebody who is both an engineer (two degrees, 10 years practice) and an economist, I can tell you that it is generally true that engineers recieve almost no training in either (a) how markets behave or (b) human action - which is basically the explanation for (a). Engineers are the ultimate technocrats, and as such are the most acute sufferers of Hayek's "fatal conceit". This isn't some sort of personal failing, just a consequence of the way they're trained. Their world is excessively deterministic.
Second point: what you're describing above, of course, is the Hotelling theory in a nutshell. Having studied energy economics at the doctoral level, I know all too well. As a consequence, I also understand it's failings, the most obvious being that it requires the extent of the reserve to be known. That is, it kicks in when we are near the "endstate". Since it has not been observed, we can be pretty sure we have yet to enter the endstate. Which is one of the many reasons why I feel that "Peak Oil" is a sophistic folly.
Felix: there is a difference between volatility and trends. The price swing of the past 18 months is much more likely to be volatility than trend. And as JDH pointed out, volatility is basically impossible to predict.
Posted by: Barry P. at July 11, 2005 09:41 PM
I'm not clear whether you're setting up this argument as a straw man to be knocked down, or whether you're advancing it as an active opinion.
In any case, it seems that the argument makes several assumptions. For example, it assumes that the future scarcity of oil is known with enough accuracy that a significant fraction of the market players would bet on that predicted scarcity.
This leads to the parenthetical observation that you're not actually dealing with future reality, but rather with present beliefs. If a sufficient number of players now confidently believe oil will become scarce (and thus expensive) in the near term future, it is reasonable to expect that the current price would rise.
However, examine that premise more closely. "Scarce (and thus expensive)" -- suppose price controls are imposed? What happens to your bet in that case?
Suppose the U.S. government releases oil from the strategic petroleum reserve? Could that affect your outcome?
Suppose that, at the first sign of real shortages, ownership of oil reserves is quickly nationalized in virtually all producing countries? Your bet may depend on oil from leases. Such a lease can suddenly cease to function. Your company may then be transformed from an asset holder to a distributor. All your "withhold supply" bets are suddenly cancelled, because you no longer control the supply.
Then again, perhaps you're thinking of the protagonists as oil producing countries. What happens when a superpower invades your country and forces you to sell oil on their terms? Did your saved-up underground assets help you in that case?
Markets thrive on uncertainty, but they function best when there is quantifiable uncertainty -- that is, when probabilities can be estimated, based on past experience. When a market enters completely new territory, it's difficult to make intelligent bets.
Right now, the players are still mostly betting that things will continue as they are at least for several years. Put yourself in their place. Wouldn't you have a hard time facing your shareholders if you didn't lay down the majority of your chips down on the square labeled "status quo"?
That's called fear overwhelming greed, and it happens all the time.
Posted by: Ralph at July 11, 2005 10:12 PM
This is a good explanation. If it were generally understood among the Peak Oil community it would be a big help. The fact that so many of them distrust economics really hurts their credibility with me. It's like an inventor of some new technology who has nothing but bad things to say about the laws of physics. Not exactly confidence producing.
One problem faced by markets in the current situation is a lack of information. Most of the oil companies seem to accept that non-OPEC oil production is likely to peak in a few years, so the question is whether OPEC will be able to increase their production to fill the gap. The problem is that OPEC is notoriously secretive about the details of their oil production situation. On the one hand we have the sheiks saying that they can increase as needed, but on the other we have people like Matt Simmons in his new book saying that the Saudis are overextended and hitting a peak of their own. It comes down to one guy's word vs another's, and markets don't have any magic formula to penetrate to the truth.
The result is that the markets become nervous and unstable, easily swayed by rumors, everyone looking at everyone else to try to get a sense which way things will break. Prices often become metastable, sticking to one trading range for no particular reason and then suddenly switching to a new price range, seemingly on a whim.
The point remains though that even though the situation is uncertain and the markets don't know what to do, that reflects the reality of the situation. Nobody has a crystal ball. Peak Oilers are deluding themselves when they claim that the situation is so clear and obvious. It's not like they are privy to insider information that the markets don't have. Just as you said, if things were really so clear that a peak was imminent, the markets would be reacting very differently than they are.
Posted by: Hal at July 11, 2005 10:12 PM
The difficulty I have with a lot of economists' writings on this subject is illustrated well in your post. The underlying model that people are a bunch of rational actors who individually project the future well and then act on their projections is fairly inconsistent with my experience of humanity (and I think with much psychological study, though I have only an amateur acquaintance with the latter). Markets are just a community of humans. Humans are very emotional in their decision-making. They cannot really credit that the future will be sharply different than the past until they have experienced the emotional consequences of a new regime. This is especially true of humans in groups, who's individual judgement gets subjugated to that of the herd, until the situation become so obviously untenable that everyone starts to change (as in the sudden changes of prices across a bubble-crash sequence). We create fire brigades only after major fires, earthquake codes only after major earthquakes, departments of homeland security only after 9/11, and we will figure out how to price post-peak oil only after the peak.
There are exceptions of course - they are the entrepreneurs who found successful new businesses and the Warren Buffett's who consistently beat the market. But the market is mostly made up of the 95% who are all following each other in the conventional wisdom of the day until that wisdom becomes untenable.
This is not to say people aren't primarily motivated by self interest in their market decisions. Just that their judgement is too emotion driven to make the traditional economic assumptions a very good model of reality. We humans are not very good about imagining how a particular situation will make us feel until we are really in it, and we decide to buy or sell in the market based on our feelings (specifically fear and greed).
My training was in Physics, and major transitions in markets strike me as rather like a very famous model called the Ising model. It began as a model of the spins of neighbouring atoms in magnetic materials, but actually models a wide range of physical situations. Any atom can point either up or down. Atoms energetically prefer to point in the same direction as their neighbour. At zero temperature, all spins point the same way. At great temperature, thermal excitations cause all spins to be random. In between, clusters of spins form domains of mostly up, or mostly down. If you start with such a system at modest temperature in a mostly up state, and add a slowly increasing bias favoring the down direction, you'd see the system start to take bigger and bigger oscillations towards the down state, and then flip over quickly until all the spins pointed down.
Replace "up" and "down" with "bull" and "bear", replace the atoms in a crystalline grid with humans in a social network with the appropriate graph topology, and I think you have a qualitatively reasonable model of the way a market makes a relatively sudden major paradigm shift in response to slow steady changes in the underlying fundementals, rather than smoothly adjusting to future change in the way you suggest in your post.
(I don't know the economics literatures, but I figure somebody somewhere must have noticed this correspondence before and already written it up).
Posted by: Stuart Staniford at July 11, 2005 10:21 PM
The point about the Ising model is really interesting. Economists, have you heard of that?
Stuart Staniford, you should be writing a book. Your model (well, Ising's) might explain all kinds of historical and economic events.
Posted by: Ralph at July 11, 2005 10:37 PM
This will be a moot point soon----
We will not be considering superstition based market economics (with invisible hands, and unlimited expansion in a finite system), but the equalizer:
The Second Law of Thermodynamics------
This is what the bank account will look like after we have
blown the lottery that we inherited from all this stored solar energy-
Posted by: Scott at July 11, 2005 10:51 PM
To my knowledge,
the only semi-quantitative attempt to answer the
question of whether market forces are adequate to
handle the approaching problems of fossil fuel
dependency is presented in the "Limits to
Growth" series, which culminates in the recent
"Limits to Growth: the 30 Year Update", by Meadows, Randers, and Meadows. And the answer
is clearly NO, because market forces do not look
far enough ahead, while large amounts of investment and time are required for mitigation.
Nonetheless, thanks for the essay. It is far
past time for serious attention to this problem
by mainstream economists.
Posted by: Walter Cook at July 11, 2005 11:07 PM
"Are you suggesting that there's so much uncertainty about when peak oil arrives that, in
fact, there is no opportunity to profit from it?"
You argue that, "If oil producers did respond to these very strong incentives by holding back oil from today's market, the effect would be to drive today's price up". In the past what would have happened in such a situation is that the Saudis would have used their spare capacity to drive the price back down. They would act as a price-setter.
And if you look at what the Saudis said 3 years ago, they claimed they would use their spare capacity to keep oil prices around $20 a barrel. They haven't done so. One reason could be that they don't want to play that role anymore. Another could be that they have been lying about their reserves and they can't play that role anymore.
If the latter is true, prices may be adjusting to reflect information that obviously can not be reflected in a price-setter's market.
I'm suspicious of the idea expressed above that a tripling of oil prices over a short period is just "volatility". Aren't properly functioning markets supposed to smooth out volatility?
Posted by: felixrayman at July 11, 2005 11:47 PM
A quick search of http://scholar.google.com with "Ising" and "Market" reveals a small pile of papers, mostly by physicists, on the topic of Ising inspired markets of market discontinuities. Doesn't look like it's penetrated very far into the economics literature - more than half the papers are in physics journals.
Posted by: Stuart Staniford at July 11, 2005 11:55 PM
I have two points, one pro and one con. I do believe that we are coming up against the end of oil, but do think market signals will make the end more gradual as you indicate. So I am a peak oil/soft landing believer.
1. Ralph presents an argument that is very prevalent in the peak oil community, which I think is theoretically sound, but not accurate. I have made comments similar to yours on the excellent The Oil Drum website and the most frequent response is the severity of consequences of shortage are so extreme that oil will no longer be a freely traded commodity, but instead a political commodity in which ownership is by force and previous contacts are not honored.
If I can take the liberty of paraphrasing, the idea states that the holders of oil might well believe that prices will rise and that the future economic value of their oil will be higher in the future. However, they are not convinced that they current market structure will exist in the future and fear they would not be able to exercise their claim. Therefore the present value of the future price of oil is not present price plus storage price minus convenience yield, as theory would hold but instead has another variable for risk. So the present value of a future claim on oil would equal the discounted future value + storage cost minus convenience yield minus risk of loss of asset.
My belief is that markets are pretty robust and in any event, there is no indication that I have seen that a significant number of investors hold this belief. In fact, a significant number would have to mean almost all since if a well capitalized 1% who disagreed they alone would drive the price up.
2. On the contrary Barry P. posits that the current price rise can be credited to volatility, not any sustained trend. And based on historical evidence his point is very strong. Oil has a 100 year history as a volatile commodity and one has to assume that the very nature of the market has changed. Prices have been high in the past and have reverted to and overshot the mean and could do so again. Four years ago, oil was at close to $10 a barrel.
However, in any scenario in which the end of oil is driving the price up, it will be indistinguishable from the effects of volatility. If a car comes close to hitting you three times, and then approaches a fourth, the evidence is identical to the first three. But it is also identical to what you would see before you did get hit.
Other commodity markets have been exhausted or the demand has been offset by discovery of a substitute. Whale oil has been presented as an example of the first, with oil becoming the substitute. But oil is a special commodity, like gold. High gold prices do create incentives to mine, but at the end of the day gold is finite and can only be found where it is. But unlike gold, oil is consumed and needs to be replaced. I do think we will someday find substitutes but a high energy density liquid that comes out of the ground in near usable form is going to be extremely difficult to substitute.
I am not an economist, geologist or an engineer but I have worked in and studied the energy sector. From a pure theoretical standpoint, oil would appear to be no different than any other commodity and we all know that Malthusians were wrong about those. However, in the case of oil it seems obvious, to me at least, that many individual fields (or countries) have peaked and declined, the U.S. being the most prominent example; there have been no major discoveries in thirty years; and that the next steps in technological innovation are not ready to bring on large amounts of new supply. (I like this presentation as a primer on non-OPEC supply www.csis.org/energy/040908_presentation.pdf ) Take the refining subsector. US capacity is currently near 98%, the market is expected to grow, margins are wide, but no new investment has been announced and there is a five year minimum time required to get a green field project on line.
This post has already been too long, so I will conclude by echoing the sentiments of the original post. The global oil market is important and complex. It contains economic, political and technical arguments. No one discipline explains everything and none of us really know what will happen.
Posted by: Jack at July 12, 2005 12:22 AM
How an economist might respond...
"The outstanding fact is the extreme precariousness of the basis of knowledge on which our estimates of prospective yield have to be made. Our knowledge of the factors which will govern the yield of an investment some years hence is usually very slight and often negligible. If we speak frankly, we have to admit that our basis of knowledge for estimating the yield ten years hence of a railway, a copper mine, a textile factory, the goodwill of a patent medicine, an Atlantic liner, a building in the City of London amounts to little and sometimes to nothing; or even five years hence. In fact, those who seriously attempt to make any such estimate are often so much in the minority that their behaviour does not govern the market."
Posted by: Walker at July 12, 2005 12:31 AM
We'd be remiss if we did not point out that the largest collection of market participants are generally good at recognizing established trends, but mostly awful at recognizing when the trend is ending and more importantly when the trend has reversed, at least until such time as the reversal is then painfully obvious.
And a goodly number of participants become absolutely certain of their convictions right at the very top (or bottom) - the turning point. Once most everyone is convinced that sets the stage for the reversal...
Posted by: Michael Watkins at July 12, 2005 01:03 AM
I think the reasons why oil futures markets are under-pricing oil futures may be:
i) information- they are relying on production and reserve data from the IEA, US DoE and Shell etc. Analysts recycle data rather than going and finding their own. Some isnt available to be verified (eg Saudi reserve data).
ii) status quo/normality- global peaking of oil production will be a once in earth history event. Most people find it difficult to believe that it will occur, even when Hubbert's theory is explained and examples of US, Norway, UK, China, Indonesia, Australia... are given.
The 'peak oilers' have used the power of the web to gather, verify and debate alternate information on oil production and demand data from credible sources such as Matt Simmons in the US, Dr Ali Samsam Bhaktiari in Iran and Colin Campbell in the UK. If they were traders they would be pricing oil futures at your $200 per barrel.
Posted by: Davidw at July 12, 2005 01:30 AM
Bear with me and let me include two price charts here as well in order to argue the point that what we've seen lately happening with crude price is not "volatility" but a trend. If one is willing to throw in more descriptive words in the mix. I'd accept the description "volatile trend". One can dismiss any minor "volatility" that serves the purpose of proving or disproving support for prices at given levels simply by moving up in time frame.
For example, on the weekly chart - crude, continuous contract:
Here we see price has been in a classic up trend at least since 2003, although for practical purposes the uptrend really can be counted to be in place since mid 2002. When in doubt move up a time frame:
CL continuous contract, monthly data
There have been a series of higher swing highs and higher swing lows since mid 2002, the classic definition of an up trend.
Typically when price accelerates and rises in near parabolic fashion, as the price of crude has over the past 24 months, prices do break and fall hard, often taking years to recover. Such was the case in Gulf War I times, or 2000's Nasdaq peak, and literally hundreds of internet bubble stocks.
I really hate even thinking the words "this time it might be different" let alone uttering them, but its possible. Unlike Gulf War I, the current situation is of longer term duration. What premium is built into oil due to the ongoing "war on terror" / conflict in Iraq and elsewhere in the region - is longer term not short term thinking at work, quite unlike Gulf War I - the outcome of that conflict was as predictable as the initial onslaught in Iraq in this decade.
However now the situation is different. We have a longer term disruption at the same time as global markets as a whole are increasing demand at the same time as China and India (and other developing nations) are stepping up demand growth something fierce.
Posted by: Michael Watkins at July 12, 2005 01:38 AM
Availability of information as Davidw points out is a key component in keeping prices where they are. No one will go too far out on a limb when there remains substantial doubt as to whether we are just over the horizon going to see peak oil or whether the process of "peaking oil" is going to be with us for many years before the "peak".
So today markets have priced in a premium, in my opinion, based on what is known - demand is up, substantially over the past two years; supplies are not growing at the same rate and there is fear that little additional raw crude capacity is available.
All the other tidbits of data which push price day to day - refinery capacity, storms, etc - are pieces of the puzzle and important to be sure but the basic backdrop is one where demand is bumping up against what is believed to be the current ability to supply and thus...
... thus more traders will fear being on the short side (there have to be sellers there somewhere believing prices are not going higher) until something dramatic happens, such as sudden new supply (less and less likely unless OPEC really does have an ace up their sleeves) or until prices rise so high that demand is substantially altered.
Posted by: Michael Watkins at July 12, 2005 01:45 AM
Hello all, saw your discussion.
At oilcast.com we are very interested in questions of supply, peak oil people seem to like our news etc. We've covered the subject way before most other journalists.
Did anyone note that JODI failed to bring their data-set analysis to the table again? Years after formation? Any day now a new piece by me will go up at the BBC about data. It is a foundation stone to the arguments over supply issues. Basically have politics interefered with accurate data? No one knows. Not me, not anyone else.
Also a new piece for BP will be out soon. You might like to find it. Energy transition, data transparency, alternatives liquids for transport, exploration, investment etc etc are all subjects for discussion in the industry. Just like Peak Oilers discuss them too. What is the big difference?
But also 'free markets'. Are they the best way to allocate resources? If they are why is there such a huge imbalance in consumptive terms? Do markets transfer wealth away from the poorest to the richest by force of consumption? If all barriers were stripped away would the poorest countries suddenly have lots of cheap energy? Enough to create stable economies?
Are questions about markets and their usefulness really at the heart of all these discussions? Are all the peakers and non-peakers and general ruckus really just a subset of market arguments?
Posted by: Adam Porter at July 12, 2005 01:48 AM
James, great post, one small detail
in one of his books Varian uses a slitly different model for current oil price, instead of using an hypothetical figures (aka $200/ba in n years) he uses the cost of substitute (ie what would be the cost of a substitute to a barel of oil in n years, discounted by the interest rate) hence assuming that oil will only be used as long as it costs less than other sources of carbon+cost of conversion.
Posted by: Anonymous at July 12, 2005 02:02 AM
Pretty much all Hotelling modeling involves the use of a backstop technology to create a theoretical upper limit on price.
For example, for transportation fuels in the US you might use coal liquifaction as a backstop technology.
Another point: coal liquifaction is commercially viable in South Africa, and has a production cost of about $30-$35/barrel. When the markets are convinced that the price of oil stays above that for more than a few years, we can expect to see planty of coal-to-liquids plants built. And there's enough coal in the US to last for a long, long time.
Of course, the peak-oilers always pooh-pooh the concept of an alternative picking up the slack. To them the only possible solution is a massive die-off of the human race.
Posted by: Barry P. at July 12, 2005 05:42 AM
Things you left out:
1. Opportunities for conservation may not be as big as you thought. Suburbia (at least in the USA) can't be served by transit (or even in many cases carpooling).
2. Rebuilding the suburbs so that they CAN be served by transit (or the bike or the foot) is a decades-long trillions-of-dollars project.
The market can't beat physics. Most suburbanites are clinging to the hope that the hydrogen economy will save them - but energy density is not a matter of economics, it's a matter of physics. Likewise, distance from the suburban house to the suburban grocery is not a matter of economics, it's a matter of geography.
Posted by: M1EK at July 12, 2005 05:57 AM
I thought this commenter:
"Where were your theoretical profit seekers three years ago? Oil prices have nearly tripled over the last few years, why are you so sure what just happened is unlikely to happen again?
Doesn't the fact that oil prices have nearly tripled over the last three years suggest that information in the markets for oil is not nearly perfect enough to make your arguments relevant?"
nailed it, and nobody's bothered to respond. IF the poster's theory is correct, then logically the tripling of price over the last three years is part of the "gradual" rise in price of oil, which means the rate of rise is fairly high - which WOULD in fact get us north of $100/barrel pretty soon.
OR, the original poster is wrong, since we saw a tripling in price of oil which wasn't part of the expected 'gradual rise'.
Which is it?
Posted by: M1EK at July 12, 2005 06:02 AM
Did you see comments #2 and #4, M1EK? There's also some follow-up discussion of these.
Posted by: JDH at July 12, 2005 06:06 AM
Engineer here. Better find another backstopping technology. You will not be able to build enough coal liquefication plants (which in SA are really coal gasification and Fischer-Tropsh) to replace oil in anything less than twenty years.
The main divergence between economists and engineers would be that economists ignore time lags for actually building things, and the consequent large resistance of corporations to make the massive investments required. After all, an investment that makes tons of money ten years from now does not get anyone bonuses now.
Posted by: Tim H. at July 12, 2005 06:31 AM
The dismissal of economics by some "peak oilers" probably has a lot to do with comments made by Ken Deffeyes and Colin Campbell, two prominent petroleum geologists in the PO field. Campbell and Deffeyes both take issue with a notion they attribute to economists that higher prices will magically lead to increased supply. I'm not sure how many, if any, energy economists actually believe this, but "peak oilers" have repeated this so many times around the web that it has become gospel.
Posted by: JLA at July 12, 2005 06:46 AM
Relying on coal gassification and/or tar sands (use of natural gas to cook oil out of sand) as comparable alternatives may not be such a great idea since we already allocate nearly all current production of coal and natural gas for daily use. I would love to see the gameplan for someone proposing to rapidly and massively increase coal and natural gas production in the next five years. The challenges of siting LNG terminals only hints at what's involved.
Posted by: olson at July 12, 2005 06:54 AM
The answer to the question raised by James has to do with denial. A single person showing symptoms of denial can be treated by a competent psychologist. That option is not avaiable when the whole society decides to deny that there is a problem. The sysmptoms of denial are everywhere. The topic of depletion of energy resources is a taboo in the public domain (media and politics). When president Bush decided to go to war with Iraq, there was a big silence in the senate - everybody understood and agreed silently with a few notable exceptions (Robert Bird was one such exception). Even in personal conversations, the topic of oil depletion and peak oil is not an easy one as many people interested in that issue will confirm. As a civilization we do really seem to believe that the earth is for all practical purposes infinite and that the second law of thermodynamics is only a theoretical concept which does not have consequences to our lifes.
Indeed, the idea of peak oil is absurd to conventional thinking. Standard economics teaches us that the human mind is superior to any constraint posed by nature. We can always solve the problem of scarcity by inventing or using something else. Since this type of thinking is so deeply ingrained in our collective mind and since we, as individuals, do not really understand how the world works, we can not really respond in a rational way to the threat of death to our civilization posed by peak oil. It is therefore not surprising that collectively, we behave very much like patients who suffer from a form of terminal cancer but do not have the emotional strength to face the truth. Indeed, there are many doctors who claim it would be unnecesserily cruel to tell these patients the whole truth.
In conclusion, to answer the question raised by James, I think the future market is no exception to the rest of society. A permanent decline of oil production is only a distant possibility of little consequnce for the next few years.
It is also worth pointing out the the greatest opportunity to profit from the rise in oil prices is actually not in the futures market but in the stock market. "The cheapest oil can be found in the stock market". Indeed, many oil stocks are still valued at a level reflecting a long term oil price of $40 - far from the present $60 and the $200 quoted by James.
Posted by: Robert Sczech at July 12, 2005 07:14 AM
I'm an engineer with a minor in economics, and I agree with Barry's comments above about engineers having little training but having a predilection to autocracy (though I think that physicists and biologists may be even worse). Tell all the "Nut Room" parables you want, I don't know how you persuade a certain kind of person that knows a little science and just enough economics to be dangerous. I suspect that anyone who has ever bought or sold anything believes they know as much about "practical" economics as you do (and everything you know that they don't isn't "practical"). This suspicion has me convinced that economics ought to become part of core high school curriculum so that people don't start their adult lives with misunderstandings about how the world works.
Regarding oil, I tend to reject the herd psychology and oil-producer conspiracy counterarguments because these prices are largely set by highly motivated, informed, and competing professionals with their own methods of information gathering (including professional tanker counters at the major ports - there was a WSJ article a few years back detailing the private information networks). In other settings, I might be more inclined to accept herd and conspiracy theories, though.
I think the hardest problem is that peak oil (if it ever occurs) is not going to give us a hard signal. New finds will come in more slowly, new fields will be smaller but that may not be obvious until they've been explored, and gradually rising prices create more incentive to invest in infrastructure that increases the pace of discovery and extraction and masks the Peak. But those same high prices serve as incentives to both conserve and to replace oil, reducing the pressure on the oil industry to find more. That last phenomenon serves as both mask and solution - which dominates? If it is more mask than solution, we may be surprised by the actual peak and falloff and there may be a period of hardship. But as you've documented, the recent price rises are caused by demand pressures, not supply falloff.
It is urban myth that markets only work in the short term: the parts of the markets that people notice (like commodity markets) might work that way, but the parts that are actually in play tend to have a very long term outlook because it's a human scale: approximately one generation. The best study of the ability of markets to respond is not the Limits to Growth, it's The Ultimate Resource (and so begins the "nuh-uh, uh-huh" exchange).
Look at the evidence available to us today: oil is $60/bbl, but it's still not the highest real price in my lifetime and natural gas and electricity are also much cheaper than they have been. Despite that, Toyota and Honda can't build hybrid cars fast enough, VW is building some fantastically efficient turbo diesels, the University of New Hampshire has a program to explore the use of algae in waste streams to produce biodiesel, researchers at UWM recently demonstrated an efficient method for converting farm waste to biodiesel, RMI is trying to prod superlighting into reality, BP and Shell are investing billions into photovoltaic development, and LED lamps are about to replace CFLs which have all but eclipsed incandescents. If we accept the myth of short-sighted markets, we must believe that it is pure luck that these and other technologies are here and/or on the way. God may or may not play dice (how would I know?); profit-seeking men and ambitous visionaries certainly don't.
Posted by: Eric H at July 12, 2005 07:18 AM
I'm an engineer, who rode the option curve at a dot-com company, and put it perspective by reading "The Winner's Curse" and "The Blank Slate."
I think we humans can be proud of our economic ability, but we aren't perfect. We carry a lot of pre-human psychology into any economic exchange. We eat while food is in front of us, and drive while the station has gas. We have the built-in expectation that tomorrow will bring another meal, somehow.
Looking around now, at the Peak Oil question, I see a lot of governments and corporations preparing in various ways. There is no denying that action, but at the same time I don't think we can expect perfection. I expect some amount of "creative destruction" when price spikes hit.
I think the most valid criticism from the Peak Oil folks is that "most people think oil will last forever" (on a human emotional level) and that we aren't seriously looking for alternatives. We invest in high-profile ethanol/hydrogen programs seem more designed to reassure us, than to achieve practical results ... they allow us to believe that tomorrow will bring another meal.
If we were rational beasts, fleet mpg would not be declining, and we would have at least one high EROEI fuel replacement waiting in the wings.
Posted by: odograph at July 12, 2005 07:39 AM
BTW, I think coal liquifaction is our best bet for a high EROEI fuel replacement ... it's kind of interesting that it only has a fringe following, isn't it?
Posted by: odograph at July 12, 2005 08:10 AM
Well, the direct liquefaction technologies (such as solvent-refined) were never developed commercially during the 80's and require a lot of natural gas; the indirect technologies that have been around since the 30's require large amounts of water and investment (take the investment required for a refinery for your desired production and triple it).
Posted by: Tim at July 12, 2005 08:58 AM
I'd assume that wide-scale coal liquifaction would also require some adjustment in environmental values?
Posted by: odograph at July 12, 2005 09:18 AM
I am one of those Government scientists working behind the scenes looking into the validity of the peak oil argument, probable market and geo-political issues, and demand reduction and substitution options.
I will keep it short.
There is a very serious looming evergy problem within the next 5 years, however, it will play out differently in different countries/markets, due to a fracturing in supply. The US will get the lions share through means fair or foul, however they will also suffer the greatest adjustment pains.
There is no ready replacement for oil, and the hydrogen economy of the future will cover only about 20% of current transport energy demand... most hydrogn vehicles will not use fuel cells. All other viable energy sources will make up about another 10%, so as you can see our cities and farming systems will need to be far more efficient.
As far as economics go, countries that remain stable will move into a new form of Keynes/Command economics. Your Harvard MBA wont be worth the paper its printed on. Perhaps the best way to understand the present and the future is to have a closer look at world history, especially the past 200 years or so.
Posted by: scientist at July 12, 2005 09:28 AM
a) Ising model and variants: search under "local interactions", you'll find some papers in econ, mostly in finance but some in other fields. Some of the dynamics noted above appear - but, surprise, fundamentals eventually kick in. (Ising models don't overturn things.)
b) Uncertainty does not invalidate the logic, contra most of the comments above. We don't need to know things exactly in order to make bets.
c) Keynes/command is dead for a good reason. Your Harvard MBA will be more valuable than ever - that is, if you learned the economics.
d) PLEASE learn some econ, people. Why do we seek out a PhD in physics to learn about physics, but everyone is automatically an expert in econ.
Posted by: randy at July 12, 2005 10:02 AM
I may be missing something here, but your post reads to me as saying:
"If your theory is correct it should be in the price. It isn't (in my judgement as an economist) in the price, therefore to me your theory appears to be flawed."
Just to be clear, is this the gist of what you're arguing?
Posted by: FT at July 12, 2005 10:14 AM
FT, I would prefer to say "therefore I'm puzzled" rather than "therefore to me your theory appears to be flawed." What I'm trying to do is draw attention to what seems to me to be the key issue that needs to be addressed in order to reconcile the differing viewpoints.
I guess I'm also pointing out that, in order to complete your theory, you might want to have an answer to the question, why aren't more people trying to make the billions of dollars in profits that your theory implies they could? I don't view that as a question that can be dismissed lightly or casually. As Jack points out in the comments above, even if you think that 99% of the people in the world are stupid herd animals, a well-capitalized rational 1% would be enough to drive the price up. And if your theory is right, that 1% can become spectacularly well capitalized by placing their bets with money rather than with words.
Posted by: JDH at July 12, 2005 10:43 AM
"stupid heard animals" is a straw man. We are the very best species on the planet when it comes to pricing oil futures. ;-) We just get to see the distance between that "best" and the actual outcome.
The thread might also be a microcosm of the problem - to the extent that engineers are concerned, and ignored.
Posted by: odograph at July 12, 2005 11:14 AM
Just on a personal note, I am seriously considering investing something in oil futures, but haven't made a decision. The central risk I see is this: I believe supply will probably get tight, will increase somewhat more, and will then start to decline at a rate which is currently very hard to determine, since it depends on the effect of newly introduced technologies in holes a mile deep. It is unclear to me (and I think to most of the professionals involved) how high oil prices have to go to cause serious demand reduction.
(As an aside, it's perhaps a telling statement that the elastacity of energy is a poorly known quantity in economics: the relationship of the fundamental quantity that makes civilization go to the fundamental object of economics, prices, is not a number that professional economists and analysts can agree on to within a factor of two).
I believe the market will probably overshoot and produce more oil demand destruction (conservation etc) than is really needed in the short term, causing a significant drop in prices, before they go up again. The timing of this is hard to predict. So thinking that the balance of the evidence does point to a peaking in oil supply before too long doesn't lead in a straightforward way to a prediction about oil prices on a particular date.
On the coal liquifaction issue: besides the long ramp-up noted in other posts, it's not going to fly in the Kyoto countries. I think climate change is starting to get too noticeable for the world to stay in denial very much longer. The first ever hurricane in the South Atlantic is an interesting bellwether this March. The North Pole melting seems like a pretty strong hint that society is getting onto thin ice, too. If you look into the science of the last ten years on abrupt climate change carefully, it will make you think very hard about burning all the coal in the ground. (One sentence summary: it appears that historical climate change is not gradual but instead flips very rapidly between quite disimilar states, sometimes in only a few years).
On the argument that the oil traders and analysts are smart competitive professionals and therefore must usually be right. In the industry I make my living in (high tech), I've had the privilege of talking to many analysts and seeing the results of many of their pronouncements, as well as raising modest amounts of money from investors. In a public forum, let me just say no more than that the business folks are all too human. I can rest my case just by pointing to the record of high-tech stock prices in the last decade. I imagine their brethren in the oil industry are similarly human. If one looks at past predictions of mainstream analysts (eg IEA, USGS) they have not been very good.
On the other hand, one has a record of folks like Campbell repeatedly calling the peak too early. His past predictions of decline rates in particular countries also seem to be quite unreliable. Even if he's right on the broad general issue, getting the details right is clearly difficult.
I am of course always happy to learn more economics - that's the reason to post in a place like this. I believe JDH was inviting comments from a broad range of perspectives. But I can see plainly that classical economics is based on unrealistic assumptions about humanity - there's a big trend in behavioral economics which seems to be much more grounded in psychological studies of how people really behave, rather than the rational agent abstraction. (Needless to say, I'm not an expert, having only read a couple of books on it).
I started to read Julian Simon's book (UR II). I view it as a caricature of what good economists must think: his treatment of any subject other than economics is so ill informed that I just got disgusted and couldn't finish the book (I lost any sense that I could rely on him as an authority on anything). The chapters on infinity and entropy are particularly bad. I'm hoping that first rate economists view him as a second rate popularizer and idealogue.
As to the well capitalized 1%, I note recent news stories that Warren Buffett is buying utilities and investing in nuclear power.
Posted by: Stuart Staniford at July 12, 2005 12:49 PM
JDH, okay, thanks for your reply.
This post of yours seems to be trying to reach over the divide between economists and scientists in the peak oil debate, and as such I'm hesitant to add criticism. However, I'm having a great deal of trouble with many aspects of what you've written, and I say this as a reader who holds your previous posts in the highest regard.
I am not a peak oil zealot, in fact I'm not even convinced yet by the evidence put forward. However, I have many objections here, and the principal one is this:
"Let's say for discussion we're talking about $200 a barrel two years hence."
Although you give disclaimers for this example, and even draw attention to a 'straw man' argument, the fact is that $200 is constantly referred to throughout your commentary. It becomes the benchmark. Suddenly $60 oil, which was considered a staggering price level just two years ago, now appears to be 'cheap'.
Once the $200 benchmark has been set, the rest of your analysis is harder to refute, except for the argument for a "steady rise" in the oil price over a number of years if the peak is further out. Show me a commodity market which has moved in such a fashion.
So, my question back to you is where do you get this example $200 figure from? Maybe a single peak oil proponent has put forward such a figure at some point, but is this collectively what the peak oil advocates are arguing?
If the figure you had used in this post had been $100 instead of $200, then the whole thrust of the argument would be different wouldn't it?
Ah, but you'd probably say: "But if what the peak oilers claim is true, then $200 is the sort of price that oil would have to be."
And I'd have to answer: "Why would it?"
You haven't explained what the peak oilers are arguing. As far as we know you believe as Barry P does that "the peak-oilers always pooh-pooh the concept of an alternative picking up the slack. To them the only possible solution is a massive die-off of the human race."
I would humbly suggest that one of the principal reasons for the current gulf between peak oil scientists and economists is a misunderstanding and misinterpretation of their claims. Much of the mainstream view of the peak oil argument has been formulated from sensationalist doom-mongering which bears little relation to the scientific analysis which has been put forward.
It's no wonder that frustration in the peak oil community currently persists. This is likely to remain while economists express 'puzzlement' as to why prices have not yet hit three digits if the theory has any credibility.
Posted by: FT at July 12, 2005 01:16 PM
Laymen are used to living in a world full of disagreements. There is nothing odd to them about having one group of people say X is true and another group say X is false. Then people fight about it, argue back and forth, but few opinions get changed.
Economists have a very different view. They see the market as an institution where agreement is reached. Virtually everyone in a given market agrees about the price. At least, unless or until someone is willing to spend every last dollar he owns investing in a market he thinks is mispriced, eventually he reaches a point where he's not willing to invest an additional dollar, given the risks involved of being wrong. Each individual trader is in equilibrium with the current market price. He's not going to buy, he's not going to sell. He's going to wait and see what happens. Then once the price changes, he may do more buying and selling, but again he will quickly reach an equilibrium at the new price, where he is content to leave it where it is. This is the sense in which traders agree about the price.
Economists see markets as engines which generate these kinds of agreements. Given this understanding, it is hard for them to understand the lay view that disagreements persist and are normal.
Most Peak Oilers, on the other hand, are totally comfortable with disagreement. They see themselves as Jeremiahs, prophets in their own time, who are ignored and scorned. For the layman, there is nothing strange, paradoxical or worrisome about the concept that the market disagrees with them. Lots of people disagree with them all the time. Disagreements, for them, are a fundamental fact of life.
But for economists, disagreements and markets are like oil and water. They don't mix; they shouldn't coexist. Markets exist to eliminate disagreements.
In fact, there is a theorem, the No Trade theorem, which says that in an ideal futures market, no trades will even take place. Rather, the price will move to a level where everyone is happy with it, so no one is willing to make a trade since they don't disagree about the fair price.
Now, obviously this is an over-idealized model and of course economists understand more or less why traders don't go this far. But the fact remains that markets are powerful institutions for eliminating disagreements. The Peak Oilers' seeming comfort with disagreeing about future oil prices, in the context of a market designed precisely to eliminate such disagreements, is baffling and disconcerting to economists. It makes it look like Peak Oilers are behaving irrationally.
Posted by: Hal at July 12, 2005 01:36 PM
I mentioned reading "The Winner's Curse" earlier ... did everyone catch the relationship between that book, ecnomic behavior, and the dollar valuation of oil reserves?
It seems somewhat amazing to me to defend the "understanding" of traditional economic theory, in the very field that helped launch the study of "anomalies."
Posted by: odograph at July 12, 2005 01:57 PM
FT, I only intended to use the $200 price in order to be able to give a concrete numerical example. The main point I wanted to make was that, rather than a sudden lurch off the edge when oil production plummets, what economists expect to see is a gradual increase in price and gradually increasing pressures for conservation and development of alternatives. If it's easier for you to follow my discussion if "$X" is substituted everywhere for "$200", by all means please read it that way.
As for where such a number might come from, I was taking as given for the hypothetical scenario that we required a 30% reduction in global oil demand within the next two years. Again, that was just a number to fix some ideas, and deliberately an overdramatic one. Feel free to substitute Y% for 30% and n for 2.
So how much would oil prices have to rise to reduce global demand by 30% within 2 years? I don't want to pretend that I or anybody knows the answer for sure, but one way to come up with a ballpark figure might be to go back to 1976-80, when the price of oil rose from $40 to $95 (in 1985 dollars) per barrel, a change in the natural logarithm of 0.86, after which world consumption declined from 63 to 55 million barrels, a logarithmic change of 0.14. If you assume a constant elasticity, to get a 30% reduction, you might need twice as big a price increase. The value I actually used is less than that (ln(200/60) = 1.20) in part because I think it's quite possible we'll see some demand reductions already in the pipeline from the price increases that have already occurred, precisely the point you were making that $60 is already a high price.
If you have a longer time in which to make the consumption reductions, you would use a smaller price increase for such calculations, because the long-run demand elasticity is greater than the short-run elasticity.
But, as I say, I really didn't intend to be defending these particular numbers. I just wanted to use them to try to explain more clearly the basic principles. Sorry if using concrete numbers got in the way of accomplishing that objective for some readers.
Posted by: JDH at July 12, 2005 02:24 PM
Since people with graduate degrees in Science and Economics have responded, I thought it might be time for someone with a graduate degree in Business to respond (who has worked at an oil refinery).
The original post is very interesting and makes some good points. But there is one problem that I don't think has been addressed yet.
The article assumes that business will act on the idea that they “make three times as much money if they just left it in the ground another two years before pumping it out.” That might make good “economic” sense but it doesn't make good “business” sense. Most businesses, including oil companies are concerned with short term profits. They need quarterly results to satisfy management and shareholders (oil refinery managers look at daily profits).
It usually doesn't make sense to give up those quarterly results even if they might get better results a few years down the road. I new start-up company might want be willing to go without profits for many quarters but oil companies are not accustomed to doing this.
By keeping the oil “left in the ground” the oil companies would also have to go through all the restructuring costs of laying part of their work force off. There are also huge costs associated with shutting refineries down and then starting them up again.
Posted by: r_ at July 12, 2005 02:33 PM
In response to Hal Finney's excellent post: my response would be that "scientists" and "traders" (to speak in very gross generalizations) are not very overlapping populations. The kind of people who think it would be fun to spend their life trying to figure out new truths, publish them, and hang out with other scientists, are a pretty different breed than the kind of people who think it would be fun spending their lives trying to make a bunch of money by trading energy futures.
It would be a rare working scientist that has the first clue how to buy or sell an oil future. So it's not surprising that oil markets don't express their opinion (not to claim that all scientists share any particular opinion about peak oil - most have never heard of it). It's fairly unlikely that he would think it fun to spend a lot of time trading oil futures, and he wouldn't have become a scientist if making money was a principal driving force in his life. So the barriers to him actually expressing his opinion in the marketplace are much higher than an economist might assume.
One only has to look at very recent history with Internet stocks circa 2000 to see that professional investors are capable of going through radical shifts in their opinion of particular kinds of assets even though the fundamentals of those assets hadn't changed very much. So a scientist might reasonably ask why we aren't in the first stage of such a readjustment in oil prices now?
It seems to me a major market repricing of some asset class is going to proceed somewhat like a paradigm shift (in the sense of Thomas Kuhn's "The Structure of Scientific Revolutions"). It's going to start with a few individuals pointing out inconsistencies in the existing paradigm, slowly gain more adherents, then more and more and more, until it becomes the dominant view. In the meantime, the currents of fear and uncertainty drive a lot of volatility. One can argue that Campbell et al are the earlier harbingers of the changed view. Or of course that the are a bunch of lunatics who will retreat to the fringe again as soon as oil prices drop. The presence of significant numbers of oil industry insiders in the peak oil discussions might be suggestive, however...
Posted by: Stuart Staniford at July 12, 2005 03:08 PM
Peak oil analysts have vastly differing views as to when peak will occur (see the SAIC Hirsch report).
One thing they do seem to agree on is that an approaching peak and subsequent plateau of production will lead to significant price volatility. They do not see a remorseless and predictable upward trend in the oil price. For example, if high prices lead to global recession, then prices could easily fall back to $20-$30, peak or no peak.
Even if the peak oil story has merit, trading oil futures on that premise is fraught with risk. It certainly isn't a one way bet.
Posted by: FT at July 12, 2005 03:39 PM
Re comments by "r", I agree - leaving oil in the ground theory at this point doesn't wash.
Public companies have responsibilities to deliver, in a regular and transparent fashion, value for shareholders and this means today's profits and *cash flow* (for reinvestment) are king, not future profits.
Anyone that has worked for a large upstream oil and gas co knows that there is a very high percentage of white collar work vs the folks out on the rigs. Regardless, it takes a lot of humans to keep the big virtual machine running (from legal to a+d to exploration to production to IT to accounting to...). White collar and blue collar talent alike in a period of high demand like this require occupation, reward and challenge in order to keep them engaged.
There's no choice but to keep producing.
Posted by: Michael Watkins at July 12, 2005 03:54 PM
OK. I see a hyper-intellectual discussion here. 99% of people who would be affected by Peak Oil are not going to bother reading this, let alone "study Economics," as one pundit here says.
I tried reading the article. Halfway through, my eyes begin to cross. "You're dense," I say to myself. "You're not smart enough to understand these deep subjects."
This, even though I have a degree in English.
Economists and the like speak in easy, glib abstractions. They try my patience. There is nothing appealing about their "science."
I can understand such subjects as evolutionary psychology, peak oil, ecology, but I cannot fathom these economics arguments.
Back to my point about the populous who won't bother trying to understand it:
It's much more practical for people to assume that the end of cheap oil means a decline in living standards, and the best remedy for that is to Take Care of Business. Why leave it up to governments, corporations, or--god forbid--"markets"?
You can live a comfortable life with less energy if you try. You can learn the myriad ways to grow and preserve your own food.
You can learn to raise your own meat, milk your own cow, prepare your own meals.
This is a language people can understand. As for dissertations on forums like this--forgive me, but they don't mean a goddamn thing.
Posted by: Anonymous at July 12, 2005 06:17 PM
Actually, I'm not "anonymous."
Posted by: mikeB at July 12, 2005 06:18 PM
It's been hard to stake out a moderate position in this thread, perhaps because the discussion is abstracted at least once, to the level of trust in economic theory.
FWIW, I don't think we really have to worry about growing our own food (unless we want to). I think there is a lot of padding built into the current energy economy, and lots of inefficient uses of oil that (with higher prices) will fall away.
It may be hard for folks who work in the most cash-strapped and oil-dependent industries (airlines?), but there is a lot of distance between reduced air travel and hitching up the horse and buggy.
I don't know how to frame it in modern economic theory (behavioral economics?) but I think there is "some" underreaction going on, and that there will be "some" hardship which will occur when adjustment (and perhaps even overreaction) hits the fan.
Posted by: odograph at July 12, 2005 07:32 PM
I brought up URII because it and Limits to Growth define the two sides of the controversy as Tom Tietenberg explains in his textbook, _Environmental Economics and Policy_. Yes, they are both popularizers, one of the optimistic and one of the pessimistic model. One of them has proven to be a good predictor, the other not so good. (I had a problem with Simon's velocity charts, as I recall)
In contrast to what FT says above, it seems that the general counterargument to James' original point is, "If your theory is correct it should be in the price. You are claiming - as an economist - that it isn't in the price. But I know - as a scientist - that Peak Oil is imminent if not already passed, so either your theory or your opinion about the pricing is incorrect, markets are flawed, or all of the above." A lot of hand-waving is going on, blaming "imperfect markets" or "irrational actors" by citing the existence of imperfections and psychology. However, except for the Winner's Curse (which ends up showing that market actors learn, if you read all the way to the end), no actual examples of such imperfections or irrationalities specifically in the oil market have been posted, so this looks to me like, "perfections exist in some other markets, therefore they must exist in the oil market." If that's the case (and I have probably overgeneralized), I have read that Vernon Smith's experimental studies show that markets work surprisingly well despite imperfections (such as actors without perfect information), and there are other reasons to believe that some of these imperfect actors aren't so imperfect, so I don't consider the conventional economics to be dead, yet.
A problem I perceive with scientists' approach is the same one Jevons made in his coal report. They assume that demand will continue at present rates, regardless of supply. Knowing that supply is finite, they can easily predict a time when it will run out (Hubbert's use of a curve is only a slight modification to the constant extraction model - it was right for American production, and has so far proven wrong for world production). They claim, accurately, to be using physics to prove their point. The problem with that is that supply, demand, and price are all linked, not independent. Price is a function of supply and demand, and demand is a function of price and time Price conveighs information to both suppliers and consumers, and they adjust their behavior dynamically. If humans aren't always rational actors, they darn sure are not ballistics equations or automatons.
MikeB - thank you for illustrating my point about what people think about an economist's "impractical" knowledge.
Posted by: Eric H at July 12, 2005 07:53 PM
In terms of information hitting the market, I think that all the oil traders must now have at least heard of "Peak Oil" as one of those crazy theories. The fact that it has already happened here in the US makes the overall argument harder to dismiss, but the timing is very hard to predict.
Having said both those things, I think there is already a $5-$10 premium on oil because of the peak oil theory. In my opinion, the markets response to Saudi announcements of increased oil production targets seems support such a premium. Add in $5 for a war premium, and a weak US dollar, and you get $60 quite easily.
Posted by: fringy at July 12, 2005 08:09 PM
I would suggest that my colleague Heading Out (a "rock guy") has published a worthy response to Professor Hamilton's piece. I hope you'll come over to The Oil Drum and have a look.
Posted by: The Oil Drum (profgoose) at July 12, 2005 08:13 PM
Instead of oil selling for $60, Prof Hamilton is suggesting oil should sell today for $200 - the price it will command in 2 years. I think this argument ignores the following fact: In order for oil to command the price of $200 in two years, it is necessary that oil be depleted severely within the next two years. That could not possibly happen if oil would sell already today for $200. That is the logical error which underlies the puzzle presented by Prof Hamilton. Markets can not be fooled.
Posted by: Robert Sczech at July 12, 2005 09:10 PM
"A problem I perceive with scientists' approach is [that] ... they assume that demand will continue at present rates, regardless of supply."
I don't think scientists assume this (see for example "Out of Gas" by the Caltech physicist David Goodstein). Whatever economists may think, most scientists are not fools. They recognize that demand will - must - drop as prices rise.
But they believe that this drop in consumption will not be painless. They argue convincingly that very facet of our economy and lifestyle depends on cheap oil; that oil will not be cheap for much longer; that there is no readily available substitute for oil; that viable energy alternatives will require massive investments in research and infrastructure and will take many, many deacdes to complete; that we are not making these investments now; and that therefore there will be considerable pain for a considerable time when the oil crunch arrives.
So if I agree that the above scenario is likely, why (asks JDH) don't I go out and buy oil futures? I'd have to set up and fund an account (non SIPC insured) with a futures broker. I'd have to pay commissions and spread in buying and rolling the futures. I wouldn't get paid a fair interest rate on my margin. I would run the large risk that I could loose a great deal of capital from short term price fluctuations. There'd be all kinds of hassles at tax time dealing with mixed straddle treatment on the 1040. It might be five years before the peak hits. So for various reasons which
seem to me at least to be rational I don't run out
and play the oil futures market - even though I'm pretty sure I know the long term direction of real oil prices.
I do own some oil company stocks though.
Posted by: Jeffrey Miller at July 12, 2005 09:37 PM
With a finite, non-renewable commodity of known size, the profits (what economists call rents) will increase at the rate of inflation (or the risk-free interest rate).
It doesn't matter that the oil be "severely depleted" in two years: given a fixed demand curve, a constant cost of extraction and some backstop technology, what JDH describes will happen. I'd do the math for you, but the space here is limited. It is explained at this link:
Of course, there are some facts extant that make this model not what is happening now.
1) The quantity of the finite commodity is not known.
2) Demand is not fixed over time: as a price rises, some people are priced out of the market. They will seek an alternative. The increased demand for an alternative will drive investment and technology in that alternative, making it cheaper. This will make some of the people who can still afford oil to shift to the alternative.
3) Supply cost is not constant. It can be reasonably assumed that future oil will be more copstly to extract than past oil.
Concerning future versus current profits: it is accepted wisdom by many that current profits rule. But empirical tests of this idea have shown that firms tend to operate on about a five-year profit maximization horizon, with profits in the future being discounted by the appropriate discount rate. Furthermore, stock markets reward firms that invest earnings in ways that increase future profits, as opposed to making those profits "real" by paying dividends. Also consider that no stock price is determined solely by this quarter's profits.
This post could go on forever, but no end of reason, logic or empirical exposition can blunt the apocalypticist's sense of impending doom. If one really wants to believe the end is nigh, there isn't much that can be said to change their mind. Even the passing of the mystical "end date", which gets magically revised everytime the rapture fails to descend upon us on schedule, fails to dull the prophet's faith. If that doesn't shake his resolve, mere words surely will not.
Posted by: Barry P. at July 12, 2005 09:54 PM
"So, if you thought you were right about the physical scenario, and yet saw oil selling today for $60, how could you explain the situation to an economist, who says that, if you're right, oil should be selling for $180?"
The argument you put forward between economists and other experts is odd. No one knows what the physical senario is. This statement seems to assume that peak oil is happening right now. Someone who speaks of peak oil is someone who believes that oil, particularly sweet light crude is a finite resource and that at some point it will peak and follow Hubbert's curve, production will slowly decline. When that happens or what will happen when that does is anybodys best guess. Some experts believe that it will happen soon others believe we have 40 or so years. Couldn't it be that the ucertainity of it all is what keeps the speculators unsure of how to proceed regarding prices?
Posted by: Lamb at July 12, 2005 10:10 PM
"And, of course, it would leave us more oil in the future to keep the economy going as we make the needed transitions."
If this is what economists think of peak oil, then everyone is in trouble. You're missing the whole point - there are no "needed transitions" (alternative energy sources) currently available that will allow the world to continue as we know it today. It doesn't matter when the price rises occur or how they occur - the fact remains that oil is a limited resource and depletion will be amazingly difficult to deal with.
Posted by: nick at July 12, 2005 11:12 PM
In order for the price of oil to rise, it is not necessary that oil production peaks or declines. It is only necessary that the rate at which demand increases is greater than the rate at which oil production grows. I think we are at this stage now. Peak oil could still be a couple of years away. Right now the market is watching production spare capacity. As this capacity decreases, oil prices rise. Still, the market is not convinced that peak oil is near. I think that is an important distinction.
Posted by: Robert Sczech at July 12, 2005 11:25 PM
Barry, I am afraid you misunderstood my post. If prices rise within two years from $60 to $200, then this rise must be either due to increased demand and/or lacking supply (I do not believe speculation can affect prices that much). If present prices would jump immediately to $200 (as Prof Hamilton recommends), then present demand and therefore future demand must decrease which in turn improves future supply (the cost of extraction is not constant - it steadily increases with every barrel produced) which in turn contradicts the original assumption of a future price of $200. That was my point. To say it in another way: Future prices are a function of past demand and supply. (Hypothetical) changes in the past affect prices in the future.
Posted by: Robert Sczech at July 12, 2005 11:57 PM
The main point to understand, IMHO, is that the whole peak oil issue is full of uncertainties ( date of the peak, duration of the plateau, depletion rates, maturity and scalability of alternatives etc ... ). You cannot expect markets to have a rational behavior consistent with absolute knowledge ( oil will be at $200 in two years ) while real information available is scarce, short term and in some important cases guarded as national secret. You still can bet, but these bets will reflect these uncertainties.
Posted by: Z at July 12, 2005 11:59 PM
In response to Eric H, and at the risk of further turning off the English majors, I don't think it's accurate to claim Simon has made better predictions than the Limits to Growth sequence of books. The latter have fairly consistently had a base case scenario in which things peak in the 2010-2020 timeframe and then start to crash. For those who believe in a near term Hubbert peak and a sharp depletion rate, that still looks remarkably prescient starting from the early '70s. However, Meadows et al are very quick to point out they only mean the model to be a qualitative general explanation of the core system dynamics. They wouldn't claim the specific dates or the exact evolution of the system should be taken too seriously. Obviously, it models a zero dimensional world - if they get within +-50% of reality on any given thing, they did well.
The basic logical errors in Simon's book are as follows. Most of his argument comes from looking at past data over the last 50-200 years. Since things keep getting cheaper and better in that timeframe, he argues it must always be so in the future. This is logically a non-sequitur - an exponential curve, a logistic up to around the inflection point, and a peak-and-crash halfway up the peak are going to be statistically indistinguishable looking back down the curve, given enough real world noise added to the basic dynamics.
Also, almost all his data series come entirely from the fossil fuel era, so I would argue that most of the effect he's seeing is precisely the accelerated growth you get from discovering such a fantastic source of highly ordered energy that you can get out of the ground with huge EROEI, and very short payback times. (There's a much longer data series on Japan, but he omits to look at the historical quality of life over the last 2000 years in the Yucatan peninsular, or Italy, which would make one think a lot harder about the idea that human civilizations only get better and better over time).
His arguments that economic growth can continue exponentially into the future forever are where he gets completely farcical. The total energy in the planet in a thermodynamically usable (ie low entropy) form is a large but mathematically finite number. The annual solar flux is a finite number. So civilization can only double it's energy use some fixed number of times before it can't do it any more (absent a miracle). You can argue that day is sooner or later, but arguing the day will never come places you outside the bounds of rational discourse. Continuing exponential growth tends to cross even enormous chasms in quite modest numbers of doubling (he slips around actually doing the calculation and throws up a heavy smoke screen in that chapter).
(Most people who hold Simonson's views, when you really push them to the wall, turn out to be relying ultimately on interstellar space travel. That strikes me as approximately on a par with hoping that co-ordinated prayer will avail us something good).
Where I do agree with him somewhat, and somewhat disagree with LTG, is that many many other problems are survivable *if* you have enough energy to throw at them. You can grow massive amounts of food on tiny amounts of land if you can illuminate the hydroponic plants with large amounts of artificial light. You can mine incredibly poor quality metal ores and then fix up the land afterwards if you have the energy to move all that material. You can augment your denuded soil with crushed rock if you have the energy to quarry, crush, and transport it. You can sequester almost all the CO2 from your coal fired power plants if you can stand to lose the energy required to do it.
But if you *don't* have the energy, all those bets are off.
The thing that worries me (and I don't claim to fully understand the situation) is this: *all* our well developed post-oil options share the following characteristic to varying degrees: if you make a large investment of energy, you can eventually get more energy back than you put in, but the payback time is rather long. That's true of nuclear plants (with their massive capital investment up front, and the need to mine the uranium), it's true of solar (with it's high embodied energy panels or mirrors collecting very diffuse sunlight), it's true of oil sands (where you need a refinery and a bunch of steam generating equipment to get the stuff out of the ground and into a usable form), it's true of wind (where you put up a big metal thing that produces an unpredictable trickle of energy). The energetic annual ROI will be rather poor on all these. And yet we would need to make truly massive investments in these poor energetic ROI options to replace the energy we have with oil. Even conservation takes investment (Priuses take quite a bit of energy to make and the payback time on that embodied energy is long).
Those investments might be rather hard to do when the available energy is declining and already spoken for just keeping us all in the style we've become accustomed to (assuming for the purpose of discussion that the oil depletion rate becomes real and is somewhat rapid). It will be like raising capital for a startup with a so-so economic ROI story when the stock market is tanking. It won't be easily done (I was raising capital in 2001-2002, so I know whereof I speak).
The present cash economics of these things is probably completely misleading in most cases because almost all the embodied energy will have been supplied by cheap fossil fuels. To know if an option has any long term viability as an alternate basis for civilization, we have to look just at the energetics.
A good solution for rapidly declining fossil fuel would have the following characteristics: you get more energy out than you put in, the energy payback is fast (1-3 years, not 10-30 years), and it doesn't release greenhouse gases. All the candidates with that potential are still at the only-a-dream stage, as far as I can tell. The list might include tabletop nuclear fusion, gossamer thin solar panels with hardly any embodied energy in them, and bio-engineered plants that capture a far higher proportion of solar energy than the ones evolution has provided, and provide it in a form that can be trivially turned into hydrocarbon fuel.
I wouldn't rule out any of these possibilities. I hope the guys working on them are staying in their labs at least as late as I'm staying up to write this. UK North Sea oil production is now declining at comfortably over 10% a year. If that turns out to be a harbinger for what oil decline looks like after you've developed fields with multilateral maximum reservoir contact wells and peripheral water injection from day one, we're going to need to pull one of those technological rabbits out of the hat one of these years. Hopefully its some freak factor with the nature of the reservoirs in the North Sea instead (oil industry insiders assert that the same technology used to develop the North Sea is now widespread in the industry).
Finally, for our self-sufficient English major: the last time humanity was subsisting by human/animal powered agriculture alone, the population of the planet was around a billion, with regular famines to keep the numbers in check. We're now at 6.5 billion on a very rapid path to 10 billion from existing demographics alone. What's the plan for the other 9 billion? And how are you going to stop them from raiding your organic garden if they get hungry?
Posted by: Stuart Staniford at July 13, 2005 12:21 AM
Erik H, all you have to do is read the news, to hear of people (and businesses) caught by surprise at these new gas and oil prices. Obviously, imperfections exist in their planning.
Why should an asphalt contractor feel a pinch? All he had to do was buy options to hedge his oil demand ... or something.
Creative destruction is fun to read or write about, but much harder to live.
(If T. Boone Pickens is right, and we see $3 gas later this year, I think they'll be surprised again.)
Posted by: odograph at July 13, 2005 04:28 AM
Rising Oil Prices won't Reduce Oil Consumption
...it only affects WHO will do the consuming, not how much. The money paid for barrels of oil is used for ... buying more barrels, just different people are doing the buying. Our increasing use of energy has increased our ability to use it, to transport wealth around, to build all sorts of energy-comsuming devices, to move energy supplies around, etc. All of which contributes to "demand".
How much is consumed is determined, in the absence of any sensible global rationing system, by how fast we can pump it out, or by the rate of supply of other flexible sources of energy. However, if a large amount of money was somehow "marooned" in time and not spent, that might reduce the comsumption of energy, but that seems highly unlikely. If this money were "saved" in a bank, that would be equivalent to "investing" it, so would also add to demand. Since the utility value of oil seems still to be vastly greater than its extraction cost, will be still try to keep producing it faster and faster despite this increasing the energy required to get it out of the ground. The end-point is that the efficiency of oil/gas extraction decreases to the point where it is almost requiring 1 barrel of oil to get the next one out.
It is also worth noting the effects of hard-to-produce oil. Suppose it takes 1 barrel of oil of energy to get 2 barrels of oil extracted, refined and distributed (an extreme situation). What is means is that oil-fields suddenly only have half the oil you thought they had, and can only produce half as fast.
Still excited by Tar Sands?
Posted by: Stephen Lawrence at July 13, 2005 06:02 AM
Stuart: the energy payback time for a large windmill is, depending on locatiaon, around 4 years which is close to your desirable range of 1-3 years. For solar panels, that time is presently around 10 years. Further progress is likely given that serious research in this area has been only done since the 1970's. You need a century of development in order for any technology to mature. Bicycles have been build for over 200 years, yet progress is still being made. You are right in your remark that present day cash prices are misleading (since they are based on cheap fossil fuels). However, I do not believe that the present per capita energy consumption in the western world can be sustained indefinitely. Fortunately, in order to maintain life standards, it is not necessary to maintain energy consumption. Energy consumption can decline significantly if efficiency can be improved dramatically. Most energy is wasted anyway. Why do we still heat houses if houses can be build which do not require any heating? The answer most often given is: because it is cheaper to heat the house than to insulate.
Posted by: Robert Sczech at July 13, 2005 06:36 AM
I think that high oil prices will not curb demand for the foreseeable future.
The reason is what I would call differential marginal return.
If prices go up, those oil uses with the least actual or felt marginal return on the investment will be reduced first. This lowers demand pressure on prices from these uses, just to be compensated by uses with a higher perceived marginal return taking advantage of the demand "gap".
I guess this effect will happen on all levels - locally, across industries, nationally and, most importantly, internationally, due to the large differences in general industrialisation around the globe. China, for example, seems to be just fully entering the steepest part of the energy use ROI curve, still being far from its diminishing returns part, whilst "the west" actually gets relatively little real benefit from using even more energy.
So my estimation is that prices will go up to the point where
- all "easy" demand destruction like reducing direct energy waste has been taking place,
- every oil consumer feels about the same pain from *not* buying the next barrel of oil *and*
- the price for the next barrel is higher than the expected/needed ROI from that barrel.
I expect this new equilibrium to manifest itself very quickly after the actual production plateau is reached and at very high prices indeed, as there is currently a huge number of users that can make very profitable use at double or even triple prices.
A thought that was coming to my mind just now: According to Liebig's Law, the least available necessary resource limits the growth of a system. So far, oil has hardly been the limiting factor for economic growth.
Slowly increasing oil prices below a certain threshold are "just" reducing profit margins and sparking inflation in the form of increasing prices for goods and services.
But after peak oil, actual scarcity will become critical. Prices will not just be painful because they directly increase bills and lower margins, but also because lack of (affordable) oil will diminish the leverage effect oil has as an enabler for the other elements of added value in products and services.
Imagine a pizza home delivery service: at first, they might try to live with a reduced margin in order to stay competitive. Then they would have to nudge up the prices for their service in order to survive. Then there is a point where the whole service becomes obsolete because the gas for the delivery vehicles adds such a premium to the price of the food that it can no longer compete with the "inconvenience" of preparing your own food.
The whole service would go bust, the skill of the pizza cook and the friendly delivery personnel could no longer create added value for the operator of the service.
(Sorry for the crude and somewhat incomprehensible phrasing - I struggle with the concepts as well as the language.)
Posted by: Davidyson at July 13, 2005 09:42 AM
I was a little sloppy in my comments. I agree wind is reasonably favorable as long as one can add the electricity into the grid, which can be done in small quantities. To rely on it at large scales (since it's erratic), it needs to be turned into something transportable and storable (eg hydrogen) and transported: that worsens the energetics a lot.
I agree with you that a mixture of conservation and development of better renewables is the only half-decent answer over the course of the century. It just seems to me that that process can only be moderately painless if the post-peak oil depletion rate is very low and/or the plateau around peak is very long (because the energetic ROI of these things is so marginal right now). It's not at all clear that those assumptions about depletion are good (nor is it clear to me they are wrong). If the transition becomes too abrupt, there's a risk of breakdown in order (one can argue that's already started with the Iraq war). Also, there's a sharp limit to how much we can reduce our need for transport energy without cutting the global food supply a lot (since most of the people don't live close to where food is produced any more). Making this much better again requires large scale investment in more energy-efficient transport technologies (eg railways). Even R&D is an investment that societies can only make when they are in reasonable shape (eg soviet scientists spent the nineties digging in their vegetable gardens).
Finally, the low hanging fruit in energy conservation are mostly in the middle-class consumer sector (home energy use and autos). But's that not where the market will pick to conserve first, since energy is not yet a very noticeable part of most household budgets. So conservation will not kick in at scale till rather late in the game (absent a lot more political leadership than we are seeing at present).
Posted by: Stuart Staniford at July 13, 2005 10:06 AM
I'm not suprised to see "the peak oil folks" (sorry for the generalization) to swoop in on this entry.
We are correct that markets do have aspects of them that are incorrect. If the data is wrong, then people are making in correct decisions. The only thing worse than no data is bad data. The trick is, we need to work with probabilities. The response to questions in the accurancy of projections for future oil reserves / production should be, well, how likely is to be 10% low? 10% high? The response from the peak oil folks isn't that. It's simply "the numbers are crap; you know nothing; the sky is falling". That isn't correct either. Futhermore, the markets know of these issues and they are responding. The prices we tend to talk about are for the 30 day deliveries, not the long term. That is one reason why the argument "...but prices have tripled in the last three years" doesn't work. Prices for immediate delivery of oil may have tripled, but that speaks nothing about the rest of the market.
The other flaw with that argument is that it doesn't explain why a 3 test for that specific type of price change works. Why wouldn't the large drops in oil prices before that show that peak oil is not approaching? Why does this one specific 3 year period apply? Because it's the most recent? If that's the case, then why a 3 year period instead of 1 year or 5 years? I don't know. The only thing I can guess is that it's convienient. It goes back to a recent low in prices and compares that very low of the market to a peak; it's dramatic. If you go back 5 years, prices have only doubled. Or going back to @1986, they're the same. Saying prices have doubled in 3 years fits peak oil. Saying that prices for oil now are where they were 20 years ago doesn't.
It reminds me of something I run into all the time in software development. People write test cases that fit what they're looking to find, that is, that the software works. And they point to the test cases as proof. But when asked why they case that test plan, they can't really tell you.
I'm not aruging that we should all stick our heads in the sand and pretend the world is peachy. On the other hand, let's not claim the lack of peachiness is proof that the sky is falling. I suspect that in a way this is what Dr. Hamilton is trying to do; to tell us to find ways we can discuss the issue and work on that will help to move us forward.
Posted by: Allen at July 13, 2005 10:29 AM
Allen, I think it is easy to identify the two extremes to the Peak Oil argument. At one end are the "die off" folks, who see human popluations as unsustainable in their livetimes. At the other end are the "there won't even be a speedbump" folks, who think smoothly rising prices will yield more extraction, and low cost alternatives.
I feel like I'm in the middle (there will be some disruption) and that I can criticize in both directions (toward excessive optimists and pessimists).
I'm sure part of what is happening in this thread is that relative moderates (dare I say "just left and right of center?") are being viewed as more extreme than they really are, by folks on the other side.
FWiW, I'd add an important word in your comment:
"The response from the [extreme] peak oil folks isn't that. It's simply 'the numbers are crap; you know nothing; the sky is falling'."
Posted by: odograph at July 13, 2005 10:59 AM
A lot of papers using Ising models, often labeled as something else, such as statistical mechanics or statistical physics models, have come out of the Santa Fe Institute. Some better authors include Lawrence Blume, William Brock, and Steven Durlauf. There are indeed a lot of papers in Physica A, Physica D, and Physical Review Letters E, by so-called "econophysicists" as well.
One implication of these models with their dynamics instabilities is that there can be overshoots of prices in various directions. This may be very important for the discussion at hand. Thus, $200/barrel may be too high, but we might see it as an outcome of some kind of herd behavior Ising overshoot. But such an overshoot would probably bring an overshoot the other way, quite likely too quickly for an increase in production from more serious alternatives to kick in seriously. Thus, a more measured increase to say $120 might bring on line the very abundant tar sands (500 billion barrels worth in Lake Maracaibo in unpleasant Venezuela alone and a lot in Canada I hear), of which there has been little discussion in this thread.
One observer had it very pointedly. A major problem here both theoretically and practically is deep uncertainty about actual supplies. Again, the key players here are the Gulf producers and most especially the Saudis. Part of the reason for the recent upward spike is clearly the rumors that they may be reaching the end of their apparently bottomless pit, and if their al Ghawar, the world's largest pool, runs dry, that alone might bring "the peak," even if it is not as traumatic an event as many fear, and even if the Saudis really do have a lot more in their higher cost Empty Quarter.
Posted by: Anonymous at July 13, 2005 11:05 AM
Ah, exactly. Markets respond quite well to perfect information. However, many of the variables in this morass are unspecified (e.g., oil reserves, intentions, geopolitical wranglings, awareness of the coming supply/demand imbalance in Q4 to whenever, the potential for declining growth, etc., etc....).
These all add uncertainty to the markets' abilities to respond rationally, does they not?
Also, I'd like to be clear, I am with odo on this...I consider our perspective over at The Oil Drum somewhere between the trip to Olduvai Gorge and a car running over a speed bump. We're not Kunstler-esque...I think we believe something can be done.
However, the longer we wait though to destroy demand and invest money in R&D&I, the more petroleum is going to cost. It take petroleum to efficiently innovate because of its very high EROEI. The longer we wait, the slower the innovation will be, it seems. It is a catch-22.
Posted by: The Oil Drum (profgoose) at July 13, 2005 11:30 AM
Nobody knows what price is needed to cause demand destruction. Once demand destruction kicks in, demand will go down and the price will go down, aside from peak oil. Nobody knows whether that is $60, $80, $100, or $200. The price will probably hesitantly rise until that point is hit.
Peak oilers would say that the supply side is relatively well understood, the demand side isn't.
Posted by: Greg at July 13, 2005 12:09 PM
I am just identifying myself as the author of the comments that started noting papers at the Santa Fe Institute and went on to talk about price overshoots, uncertainties in Saudi, and tar sands.
Posted by: Barkley Rosser at July 13, 2005 01:27 PM
One could argue that $60 oil (and future prices fairly flat at that level) is the market's blended estimate of the probability that the Saudi's are telling the truth: "the market is well supplied", implying oil should fall back to $30 or below, or Matt Simmons is telling the truth, implying oil should go to $100+ to shave a few million bpd off demand in the near term (or whatever it takes - this was the Goldman Sachs estimate).
I have a question for the economists. It would be nice to see a plot of the autocorrelation r^2 between oil futures and the price at the time the contract came due, as a function of the lag. In other words, how good have oil futures markets empirically been at predicting prices? Has this been done? Is the correlation of the 3 year future (or the 5 year or whatever) stronger with price at the time of contract fulfillment, or at the time of sale of the contract?
If the plot isn't available but someone can point me at the raw data, I'll do it.
Posted by: Stuart Staniford at July 13, 2005 01:30 PM
I am not defending space travel or Simon's views on infinity; those are red herrings. You are correct that it is a fallacy to claim, "Things have gotten better so far, so we can expect them to do so indefinitely," but it is equally fallacious to claim, "Things have gotten better so far, therefore everything will come to a sudden crash any time now." Those are both extreme forms of the Optimistic and the Pessimistic views (and the pessimists had to keep revising their dates so often that they now claim that such predictions were for illustration only, not intended to be real predictions), but I am not defending either one. At best, we can claim, "Things have gotten better so far, however we need to analyze A, B, and C in order to determine if things will get better, level off, reverse, or fall off dramatically in the future. In any event, our forecasting ability gets poorer the further forward we look." Our problem, then is to determine what A, B, and C are, and see if we can come to some agreement about what the trends are going to be into the foreseeable future.
I believe that population, wealth, energy, and technology are the factors that are important for determining human well-being, where I incorporate legal, political, and cultural institutions along with scientific knowledge into "technology". Population is dependent on wealth: subsistence societies (e.g. most of Africa) are small and stable, slightly wealthier agrarian societies are marked by rapid growth with periodic declines (Asia), while even wealthier societies see their populations stabilize and perhaps even decline (US, Europe). Wealth depends on energy and technology: with those inputs, societies can either make or make and trade for their needs. Energy also depends on technology. So population will stabilize if we have enough wealth, which means having enough energy, which means having enough of the right kind of technology. A steady population with increasing technology will experience increasing per capita wealth.
Pre-industrial societies relied on renewables - biomass and solar - but those weren't easy to produce in mass quantities with technology available at that time. When fossil fuels came into use, they were so cheap and plentiful that we quickly abandoned everything else. They're still so cheap that we don't come anywhere close to using them as efficiently as possible (the comment above about it being cheaper to heat than to insulate in on target), and it doesn't make sense to spend too much wealth developing a replacement (more below). Fossil fuels are the defining energy of our times, but renewables (and possibly non-renewables more abundant than fossil fuels) will define our future, like it or not. We have just barely begun to develop renewable technology because fossil fuels are still so cheap.
Daily insolation provides much more energy than we will ever need, so we only need to capture a portion of it. One form, wind, is limited to a small percentage of the earth's surface, but has rapid energetic return where it is applicable (because they are simple devices). Photovoltaic is rapidly gaining, but still requires more energy input than you get out (it is economically viable if you calculate over a 10-30 year period, but is also location-dependent). Heliostatic generation probably has greater potential (mirrors are simple to make), but is also location dependent. Conversion of biomass to fuel is probably the best potential source, is less location-dependent, and has the added advantage of utilizing existing infrastructure (energy returns are at least 3.2 and take less than a year). However, an even greater potential source of energy is conservation, since eliminating waste is no different in effect than finding new oil.
It isn't necessary for "the answer" to completely replace fossil fuels for at least a generation. "Peak oil" doesn't mean it will all be gone suddenly, it only means the supply will decline over a finite period of time, so we still have oil, coal, and natural gas for some time (generations, in the case of coal). We only need to augment them with renewables until technology allows us to replace them (and the faster we augment, the longer fossil fuels will last). Meanwhile, we still have a lot of wiggle room in the form of increased efficiency in application. More efficient transportation, lights, heating and cooling, industrial processes - we waste a lot of energy because fossil fuels are still the cheapest source of energy around. Since fuel expenses have tipped up from 4% to 5% of household income ((NY Times, 23 April), people are starting to pay attention, hybrid vehicle sales are doubling every year, and hybrids are expected to constitute half of the vehicles sold in 2012.
There is one other factor I haven't mentioned: time. It takes time to react to unexpected events (a well-coordinated terrorist strike in a certain Middle Eastern Kingdom could not only reduce oil flows now, but could interrupt them for a long time and, if the pressure is bled off, render those fields unusable). It takes time to develop new technology. If we underestimate the amount of wealth we need to spend on finding alternative, renewable energy sources, we won't have time on our side. However, the more time we do have, the greater our potential response. It may not be logical to extrapolate future natural resource extraction from past trends, but it is reasonable to assume that human knowledge will only increase.
Odograph: Your comment to me was addressed in comments #2 and 4.
Posted by: Eric H at July 13, 2005 01:30 PM
It's easy to declare anything that happens in real time as "volatility" ... you can always come back in ten years and write a paper about the "trend."
Posted by: odograph at July 13, 2005 01:51 PM
Stuart, here are a couple of papers that address your question about the correlation between the futures price and the subsequent spot price:
The basic answer from this research is that the correlation is pretty weak. I read that as saying that in fact it's very difficult for the market or anybody to predict oil prices. Many of the comments in the discussion above also hit on the theme that it's hard to predict oil prices. But I think the question of how hard it is to predict oil prices is only tangentially related to the point I was trying to make. I intend to have a follow-up post on this issue in the next day or two, so I hope you'll check back to look for that.
Posted by: JDH at July 13, 2005 01:57 PM
Eric H: I think I agree with most of what you say except for the part about biofuel - biofuel appears to be a non-starter energetically. See:
As I've said earlier, I believe the depletion rate is going to be the key to whether the process is miserable or fun, and that remains very uncertain.
JDH: Thanks for the links, fascinating. It does indeed look like the oil futures prices contain almost no information about the future price of oil (only 5% of the variance). I look forward to seeing how you recast your arguments in light of that.
Posted by: Stuart Staniford at July 13, 2005 04:29 PM
JDH writes "So how could it be that there are billions and billions of easy dollars to be made, and nobody can be bothered to collect them?"
This happens all the time. Consider, for example, the Google IPO. This was open to all bidders, and all financial info was publicly available. The final IPO share price was $85. Today, less than a year later, the Google share price is close to $300. Billions and billions was left on the table by those who could, but did not, place higher IPO bids.
So why should I believe that the oil futures market is going to reliably predict peak oil?
Posted by: Avo at July 13, 2005 05:52 PM
"So how could it be that there are billions and billions of easy dollars to be made, and nobody can be bothered to collect them? Unless you have a clear answer for that question, an economist at that point is going to ask whether you're sure that you've got all the facts straight, that oil really is going to sell for $200 a barrel in just two years."
Two economists, Stefano DellaVigna of UC Berkeley and Joshua Pollet of Harvard wrote a paper a few years back called "Attention, Demographics and the Stock Market":
They asked the question, does the stock market correctly discount the effects of very predicatable demographic events? The idea is that
certain age cohorts tend to consume certain things - young men drink beer, old people buy annuities, and b) the size of these age cohorts through time are quite predictable long in advance.
What they show is that the market takes near term information into account - say the next 3 to 5
years - but beyond that horizon, the market does not correctly discount what is almost certain to occur. They show this by constructing portfolios of stocks that simple demographics indicate should do well between five years and ten years out, and showing that these portfolios trounce the market.
So let me ask this question:
"How could it be that there are billions and billions of easy dollars to be made in the stock market based on simple demographics, and nobody can be bothered to collect them?
Now in the case of oil, when you consider the
high risks and high transaction costs of playing in the futures and options markets - and you add to this consideration the natural uncertainty about when exactly the peak will occur, and you also reflect on the conclusions of the paper quoted above, do you really think it so unnatural that oil sells currently at 60 when there is a high probability that 3 or 5 or 10 years it will sell at a much higher amount?
Posted by: Jeffrey Miller at July 13, 2005 06:36 PM
Yeah, Pimentel appears to be right about corn for ethanol. Corn-based ethanol and soy-based biodiesel are being driven by politics and ADM, not sustainability and (unsubsidized) profitability. However, two more recent developments hold more promise: algae and cellulose from farm waste.
Not all biofuel sources, processes, or even outputs are the same. Biodiesel is way more efficient on the road than ethanol.
Posted by: Eric H at July 13, 2005 07:22 PM
Would Mr. (Government) Scientist be willing to post any additional details to his/their research?
Posted by: Tim at July 13, 2005 07:35 PM
One good entry point to a "physics" approach to market behavior is the work of Doyne Farmer and his collaborators:
I think the economists and peak-oil "engineers" are largely talking past each other. Yes oil production will peak, no it won't lead to overnight armaggedon.
Nothing in Hamilton's economic argument refutes the possibility of another 'gas-lines-around-the-block' situation like in the 1970s. Such an episode may be required to fully alert the short-attention-span SUV driver set. But a lot of the peak oil hype sounds like Y2K all over again.
Posted by: STS at July 13, 2005 08:28 PM
I think "gas-lines-around-the-block" is about the scale we are looking at, and I think it is enough of a "problem" that we should avoid what trouble we can. That's enough to cause some "economic dislocation."
Some people might be operating at the level of "well industries die, and industries are born, that's natural." I'd actually agree ... if the overall economy remained strong, and if unemployment (and more seriously, homelessness) did not rise.
I'm in pretty good shape. I could afford a price doubling of gas and groceries ... but I don't want to see ANY more homeless on the way to the market.
I worry about the poor dumb SOBs with a $30K credit card debt, a big mortgage, and a lease on a shiny new SUV. Read the news, they're already acting like deer in the headlights.
Posted by: odograph at July 13, 2005 09:09 PM
Anyone who thinks peak oil is a "gas lines around the block" problem should read "Geodestinies", a 1997 book by geologist Walter Youngquist that thoroughly examines the geological basis of industrial society.
I'm a professor of physics. I first heard about peak oil two months ago, and I've been reading up on it ever since. Bottom line: I'm terrified. In my opinion, those who aren't simply haven't done enough research yet. Start with Youngquist.
Posted by: Avo at July 14, 2005 12:07 AM
First of all, let me say that this weblog and its comments is a gem. Very interesting things are being said, congratulations!
I am interested in the way economics theory will change in the light of a decreasing availability of high EROI sources of energy (if that's finally the case, of course).
I believe economics and markets, "per se", are, or at least should be neutral, just an efficient way of processing information. Well, if peak oil hits and we don't discover a new source of energy with similar qualities, the way economics and markets will be thought would have to change dramatically, correct?
I am for "saving the face" of our current "turbo capitalism". Not that I love it, but I think that if we over correct, the results could be disastrous (going to a totally planned economy, for instance).
The problem, as it has been said here before, is one of ideology. There will be a lot of people that will say: "look, capitalism brought us here, to this unpleasant world of reduced energy availability, while destroying the planet, it's time to get ride of it and its promoters".
And in the other end of the spectrum, others will say: "it is not our fault, the oil is there, it's just they are hoarding it and they won't let our companies in. The problem is not of capitalism or free markets, but the lack of it in those places were energy resources are".
Ok, maybe I went too far. What do you think? What we should change in economics to reflect better reality? Is that concept about "nature being part of economics" the first that must go if we want to have a sustainable economy? (We'll see if it's a stationary one or if it is bound to keep growing ad infinitum)
Posted by: Daniel at July 14, 2005 01:41 AM
Avo, have you seen this presentation by Dr Steven Koonin, who was appointed BP’s Chief Scientist last year?
I think I buy his physics-eye-view that when all else fails, there's coal.
He does point out the harder problem of global warming ... but (in my dark view) if the "lines around the block" last long, or keep coming back, people won't care quite so much about the environmental aspect.
Posted by: odograph at July 14, 2005 06:00 AM
A few comment response:
To Nick, who said:
"there are no "needed transitions" (alternative energy sources) currently available that will allow the world to continue as we know it today. It doesn't matter when the price rises occur or how they occur - the fact remains that oil is a limited resource and depletion will be amazingly difficult to deal with."
That's a very definitive view of the future. Malthus would be proud.
To Robert, who said:
"In order for the price of oil to rise, it is not necessary that oil production peaks or declines. It is only necessary that the rate at which demand increases is greater than the rate at which oil production grows."
Quantity demanded = quantity consumed = quantity supplied. The driver of price is the cost of the marginal barrel of oil. If low-cost producers increase their output, then the price of the marginal barrel stays the same, and price is unchanged. In the recent past, Saudi Arabia did this. Then they hit a production plateau. They say this is because of the unexpected demand growth in China and India used up all their excess pumping, transporting and storing infrastructure. They say they're increasing their infrastructure so they can supply more oil and bring the price down. However, increasing the infrastructure capacity will take about 3 years. We will see if they are telling the truth or not.
Thirdly, to the people who talk about the concept of "energy return on energy invested". This is an utterly meaningless concept. It would make about as much sense to talk about a "mass return on mass invested", yet we care little about whether the weight of the products is greater than the weight of the inputs. All that matters is the financial return. EROEI is sophistry.
If oil, as an input to oil production, gets more expensive, then this will drive efficincy improvements in oil production.
the peak oilers see a rise in price and swear that the end is nigh. They don't realize that a rise in price predicated upon an outward movement of the demand curve, coupled with a (likely temporary) production capacity squeeze has absolutely nothing to do with problems of "oil is running out".
Deffeyes has pegged this Thanksgiving as the date of peak oil production. All I can say is, we'll see. Personally, I'm sure consumption will increase for at least the next ten years. Probably 20.
Posted by: Barry P. at July 14, 2005 06:50 AM
I'm glad we can throw out EROEI, and get those perpetual motion machines running!
Posted by: odograph at July 14, 2005 07:35 AM
I didn't mean to imply that the sum total of the peak oil impact on the world's economy would be an episode of gas-lines-around-the-block. There obviously needs to be an enormous transformation in our energy economy. But the timescale is long enough for capital markets to do much of the work of this transformation.
Many peak oil arguments I've heard on the "engineering" side of the case have the form: here are the many, many things we rely on petroleum for. Now imagine ALL of them GONE!
In the real world, the plug is not pulled so abruptly. There are many sources of elasticity in the energy system, ranging from uncertainty about exact supply and new methods of extraction on the engineering side, to substitution effects, futures markets, and new economic opportunities on the 'economic' side.
We've been using petroleum for what? 150 years, give or take? I could easily believe that 150 years from now we will be down to a trickle in a few esoteric applications. Trying to work out how we get there is obviously overwhelming. But we have a huge, distributed, neural computing system called capital markets (or "the invisible hand", if you prefer to attach it to the collective effects of human greed) which is already taking the first steps towards adapting.
The mainstreaming of hybrid cars going on today is quite encouraging -- just the tip of the iceberg to be sure, but many once said it would "never" happen. (Maybe if Detroit had been in charge it wouldn't have.)
The social function of another 'gas-lines-around-the-block' episode at some point will be to do the work of shifting consumer preferences, which so far have underreacted to changes in oil prices.
Posted by: STS at July 14, 2005 08:47 AM
Barry P. said:
"the peak oilers see a rise in price and swear that the end is nigh. They don't realize that a rise in price predicated upon an outward movement of the demand curve, coupled with a (likely temporary) production capacity squeeze has absolutely nothing to do with problems of "oil is running out".
I think it is a misconception that all "peak oilers" believe that the price of oil right now is due to peak oil. I don't know what your definiton of a "peak oiler" is but from my understanding it is someone who believe that oil is a finite resource and that someday it will peak and we will see a decline in oil production. This does not seem to me an extreme viewpoint. Todays prices may very well reflect other constraints, but it does not mean that peak oil will never happen, even the most conservitive researchers believe that it will happen in the next 20-30 years. I think that the point is, that if we are to transition, without major diruption, to other energy sources, we need to start preparing 20 years in advance. That probably means now, thus, this is the time to begin the discussion on the peaking of light sweet crude oil.
Posted by: lamb at July 14, 2005 08:56 AM
To Barry P. regarding
"Quantity demanded = quantity consumed".
This is not true. Not every barrel demanded is a barrel consumed. A significant part of world demand is given by the desire to create national strategic reserves (hoarding). For instance, the US stores 100 days of national demand. Other countries start doing this now (China, India, Japan etc).
"quantity consumed = quantity supplied. The driver of price is the cost of the marginal barrel of oil."
Not true either. Not every barrel supplied is a barrel produced. For instance, part of the supply from Saudi Arabia comes from huge storage tanks (production buffers) which were filled a long time ago.
One problem with economics is that the models used by its practitioners are too simple in order to reflect the true complexity of the real world.
Posted by: Robert Sczech at July 14, 2005 09:33 AM
To Daniel regarding
"What we should change in economics to reflect better reality?"
As an answer to this question I make the following proposal.
The remaining fossil fuels of this world should be priced at a level at least as high (if not higher) than the equivalent cost of renewable energy. The same pricing rule should be applied to all nonrenewable resources.
The rationale for this recommendation is obvious. Once we run out of all non-renewable resources, we will be forced to pay the price of the equivalent renewable resources anyway. So why not pay the same price for the remaining non-renewable resources today? If we follow that strategy, we will never run into an energy crisis and the issue of peak oil will only be of interest to oil geologists. The rest of us could start doing more meeningful things in life than wasting our time on the internet by reading/composing nonsense on the economics of peak oil.
Posted by: Robert Sczech at July 14, 2005 10:09 AM
Odograph, I took a look at Koonin's slides. I've met him, and I think he's a straight-up guy. But I also think he seriously underestimates the challenges of switching to coal as an oil substitute. There are huge up-front investment costs, and huge mining operators are required.
Production of North Sea oil declinded 17% from March 2004 to March 2005. Houston investment banker Matthew Simmons of Simmons & Co International has done a detailed study of existing technical papers on Saudi oil production; he thinks it is likely that they are about to enter into a production decline, possibly at a double-digit rate. See his book "Twilight in the Desert" for details.
Posted by: Avo at July 14, 2005 10:16 AM
As far as the failures of futures prices to correlate well with what prices turn out to be: the problem is not with markets; the problem is just that it is hard to predict the future! As an institution, I would still claim that markets are as good as we are going to get in terms of making reasonable predictions. Exploiting people's greed, eliminating self-serving motivations like deception, manipulation and social conformity, all serve to eliminate some of the most potent sources of error.
Imagine that you got a private offer that you will win a million dollars if you predict the price of oil in December 2005 within $5. Which method would you prefer: the oil futures market? Or how about setting up a poll on the peakoil.com forums asking people to vote on what they thought the price would be at the end of the year? Do you think that a poll of random Peak Oilers is more likely to be correct than the predictions of futures traders who are putting their own fortunes on the line?
The point is not that markets are likely to be right; it is that Peak Oilers are not likely to be right when they claim to see a future so clearly, given that the markets don't see it. Now, the POers may turn out to be right about the facts, but they are wrong about the *clarity*. The markets aren't saying that oil can't go up next year, they are saying that it is far from clear now that oil will go up like that. They are saying that the chance for a big increase is balanced by the chance that it may stay the same or even fall.
This is the big disagreement between the most vehement Peak Oilers and the market. Read the posts on the Oil Drum and PeakOil.com and the other boards, and you see again and again these statements of strong belief about problems arising in the short term, this year or next year. And for many of these people, there is little room for doubt. They scorn and disparage those government and financial reports which predict that the situation will remain stable for at least the next five years or so. They call them crazy or claim that they are lies meant to lull the populace.
It is these extreme views which the markets show are fundamentally false. The point is not that the markets are right or that Peak Oilers are wrong about oil possibly going up soon; the point is that the POers are wrong to say that it is *obvious* that this is going to happen. It's not obvious, and to see that we need only to look at the markets. If it were obvious, then as Prof. Hamilton says, the markets would be acting very differently than they are.
The markets aren't an institution set up to lull people or fool people. Participants in the market are there for one reason: to make money. It's the most honest institution mankind has ever invented, because it is the only one where you make the most money possible by being completely honest about your beliefs.
Posted by: Hal at July 14, 2005 11:20 AM
A quick note: notice how most (if not all) of the economists here seem to lean on the side of disbelief regarding "Peak Oil." JDH, I think the position you're trying to take here can be approximated with that which mainstream filmmakers and writers try to attempt: status quo.
Great job on succeeding, by the way. One thing I'd like to mention, though - it doesn't mean it's entirely accurate or fair as far as the assessment of our current situation goes.
Methinks the reason why so many on this blog tend to lean towards conservative predictions in regards to our future is because they're afraid of being labelled as "fringe" (which admittedly Peak Oil is. I mean, come on, the subject matter is the end of the world, here).
It is, admittedly similar to the conundrum one finds when committing oneself to "faith": do you decide to err on the side which offers you a sense of comfort but small possibility, or do you take that leap of faith and decide to err on the side of probability, that is, with all available information known (EROEI, available statistical data, overall trends, perhaps even your own amateur analysis of human behavior, et. al) can you come to make a definite value judgement on the issue?
I think you can, and I choose to err on the side of Peak Oil.
Posted by: albert at July 14, 2005 11:21 AM
Hal asks, "Do you think that a poll of random Peak Oilers is more likely to be correct than the predictions of futures traders who are putting their own fortunes on the line?"
Yes I do. Because the great majority of futures traders are just that: traders who move in and out on a very short term basis. They bail quickly when the market moves against them, so they are not really putting their fortunes on the line. If they were required to buy and hold contracts to maturity, I think you would see very different prcing.
Posted by: Avo at July 14, 2005 12:10 PM
Hal writes, "Peak Oilers are not likely to be right when they claim to see a future so clearly, given that the markets don't see it."
I can't agree. Maybe this is the root cause of the different thinking of economists and scientists. In science, it is very often visionaries who go against the market (the consensus of scientific ideas of the time) who are right, and the market that is wrong. You know the names of these people: Newton, Darwin, Einstein, Hawking. Eventually, the market came around, and validated their ideas. But it took time.
Can we afford the time lag before the market comes around on peak oil. I don't think we can.
Posted by: Avo at July 14, 2005 12:15 PM
Odograph wrote, "We've been using petroleum for what? 150 years, give or take? I could easily believe that 150 years from now we will be down to a trickle in a few esoteric applications."
OK, fine. Now remember that crop yields worldwide have tripled due to the use of petrochemical fertilizers and pesticides, and that world population has grown to 6 billion during this period. Estimates of the carrying capacity of the earth without petrochemical enhancement vary greatly, but let's say it's 2 billion.
How are we going to go from 6 billion people to 2 billion people in 150 years?
I think the market will use its oldest and most reliable mechanisms: war, famine, and disease.
Posted by: Avo at July 14, 2005 12:21 PM
Whether or not the self-styled "peak oilers" are right or not, there is not much point in rushing around doing anything about it one way or the other. Indeed, if and or when the peak comes, presumably the price will go up, either in the sudden spikey and overshooty way it has often in the past, or more gradually. If so, at that point economic agents will find it worthwhile to substitute away from petrochemicals and to more vigorously pursue R&D and investments in alternatives.
Near as I can tell, there is nothing that we do that absolutely depends on petrochemicals, although if we were to wake up tomorrow without any, life would certainly get more difficult.
Note, these arguments hold even if markets are imperfect, which I think they are.
Posted by: Barkley Rosser at July 14, 2005 01:27 PM
Actually, I wrote the "150 years" passage you attribute to odograph.
Our agricultural use of petrochemicals actually strikes me as one of the more easily (if you'll allow that word for a decades long process) modified applications. A more environmentally sophisticated level of biotech should allow us to develop much more targeted and less toxic ways to maintain high yields while removing a lot of crap we are poisoning everyone with nowadays. Not trivial, but I think it is far from clear that we 'have to' revert to 2 billion population.
That said, I'm pretty confident that there will be wars, famines, etc. in the next century and a half. Not at all convinced that they will engulf the entire developed world in a sort of 'mad max' scenario, however.
In a geopolitical sense, we've sort of rewound to pre-1914 globalization and are trying to run the tape back with a little less nationalism and a whole lot less totalitarianism. The jury is very much out on that. But I think more universal human virtues/vices will determine the outcome than the single 'peak oil' issue.
Posted by: STS at July 14, 2005 01:39 PM
Sorry STS for misattributing the quote!
Posted by: Avo at July 14, 2005 02:15 PM
"It's not obvious[that oil is going up soon], and to see that we need only to look at the markets. If it were obvious, then as Prof. Hamilton says, the markets would be acting very differently than they are."
The paper I referenced above shows rather clearly that markets do not react to events that are more than three to five years in the future even when these events are highly predictable.
The paper also shows that you can profit from this market failure.
So while I agree that, most of the time, the market correctly discounts the immediate future, it does not discount events that are beyond five years - even if these events are almost certain.
The conclusions is that if it's going to take say twenty years of R&D and infrastructure development
to transition from the oil economy, we can't wait
for the market to tell us when to start. Because the market can't "see" twenty years ahead - at most it can see five years ahead.
Posted by: Jeffrey Miller at July 14, 2005 02:20 PM
Barkley Rosser writes "Whether or not the self-styled 'peak oilers' are right or not, there is not much point in rushing around doing anything about it one way or the other."
Arrgh! This kind of thinking drives me nuts!
There is a lot "we" could do. Rebuild the railroad system in the US, for example, since it's vastly more energy efficient to transport goods by rail than by truck.
The market can't do this by itself, because a huge amount of governmental regulatory issues are involved: no matter how much money I have, and no matter how profitable I think it will be, I can't just decide to build myself a railroad. So, we could encourage our elected representatives to pass legislation that would help make revived rail transport possible.
That's just one idea.
Or, we can sit around and wait for those oh-so-pleasant market mechanisms to kick in.
Posted by: Avo at July 14, 2005 02:23 PM
War, famine, and disease not only predate markets, they predate humans. The later two have in fact been greatly reduced in countries with market economies. They are certainly not market mechanisms.
Posted by: Roy at July 14, 2005 02:27 PM
Hal Wrote: "Do you think that a poll of random Peak Oilers is more likely to be correct than the predictions of futures traders who are putting their own fortunes on the line?"
I honestly think that they would end up being about the same. There are a wide range of beliefs among "peak oilers" - some believe that peak oil is occuring now or has occurred, some believe that it will occur in 5, 10, 15, 20... years, and some aren't going to guess because there are too many unknown variables (i.e. how much oil does Saudi Araiba have left?). There is also a wide array of beliefs about what will happen to the world economy when peak does occur. On the other side of the coin, there seems to be a wide variety of beliefs amongst economists as well. Simmons and Leeb take a more pessimistic view. The problem seems to be rooted in the fact that there is so much unknown, and the markets cannot react to that. If we were allowed into OPEC oil fields and were able to get factual numbers then the markets could react in a way that reflects reality, but at this point it is all guess work. We can only hope that SA is not in decline - but we don't know. Most of the research that the economists use is based upon numbers that OPEC countries give us - we can only hope that they are honest and accurate, but we do not know for sure.
I don't really understand the argument here. Peak Oilers vs. Economists. If a "peak oiler" is someone who believes that oil will peak someday and an "economist" is someone who does not believe that oil will peak someday, what do economists believe? Is there a group of economists out there that believes oil is not a finite resource?
Posted by: Anonymous at July 14, 2005 02:41 PM
"Is there a group of economists out there that believes oil is not a finite resource?"
Yes, there is. Morris Adelman, Professor Emeritus of Economics at MIT, say "For the next 25 to 50 years, the oil available to the market is for all intents and purposes infinite.”
Posted by: Avo at July 14, 2005 02:49 PM
War, famine, and disease certainly are market mechanisms. Consider Russel Roberts famous "nut room" analogy. (If you don't know it, you can find it here:
But let's add some twists: first, you can't leave the nut room. (It represents the earth.) Second, those nuts are your only source of food. What happens when it takes too much effort for you to sift through the shells to get enough nuts to sustain yourself? You die of famine, that's what. That's how the market solves this particular supply/demand imbalance.
If you don't want to include famine as a market mechanism, then your only other choice is to say that the market has failed and been superceded by a more powerful mechanism. Surely an economist wouldn't want to claim that!
And either way, you're still dead.
Posted by: Anonymous at July 14, 2005 02:56 PM
Oops I forgot to sign the nut room post.
Posted by: Avo at July 14, 2005 02:57 PM
Well, Avo, I haven't seen anybody here advocate Adelman's view, and there are an awful lot of posts accumulated. I think we all agree that oil is finite and production must eventually turn down. Maybe some are using the expression "peak oiler" to refer to somebody who has very strong feelings that this eventuality is coming very soon and will have profound effects on our way of life.
Posted by: JDH at July 14, 2005 03:03 PM
I find Hal's comment rather ridiculous : "It's not obvious because it is not obvious for the markets."
I remember rather well thinking it was obvious, during the dotnet bubble, that the whole situation was ridiculous and doomed to be corrected sooner or later. As is it now wih the housing bubble. The markets was wrong then, then it could be wrong anytime.
The economists should start to develop a little bit of humility, and recon that despite hundreds of years of trying, their 'science' is unable to provide any meaningful prediction and as such is not a science.
The markets are not an intelligent being, despite the religion of certain. It's just a mechanism to reach an equilibrium and it sometimes fails miserably. It does not think. It does not predict anything. It just reacts to events.
Posted by: Z at July 14, 2005 03:07 PM
It's tautological that the supply of any non-renewable resource will peak and then start declining. We all seem to agree that nobody (including the futures speculators whose track record doesn't seem to be all that good) really knows when this will happen. So lets take two boundary cases, one deliberately chosen to occur sooner than anybody thinks is likely, and one deliberately chosen to occur later. Then let's attempt to model the impact of both scenarios on economic and technological growth.
There are huge technological unknowns here, but like any unknowns they can be bounded. The relevant quantities I can think of are...
1. The EROEI on the next best backstop energy source/s currently available.
2. The time it will take to retool the infrastructure to utilize the next best available energy sources.
3. The EROEI of the best energy source currently in the research pipeline but empirically known to be feasible.
4. The time it will take to bring the best such experimental energy source to the market.
I think there are other variables to be estimated, but this is a start. Nor am I going to buy the argument that there are too many unknowns. Given the importance of the question (and even the skeptics here believe the question is important as proven by their spirited participation) any reasonable theoretical model is better than nothing. Once we can agree on what the relevant variables are, we can have a meaningful discussion on their relative influence and their likely upper and lower bounds.
Inaccurate and naive as such a model might be in the beginning, it is still superior to intuition or blind extrapolation of futures prices that economists on this very page are admitting are volatile and have a poor track record as predictors. To think otherwise would be to part ways with science, and if we take that path we might as well all just go ahead and believe whomever we feel like-- economists, geologists, politicians, or astrologers.
Btw, to the individual who said we don't care about the mass return per mass invested on manufactured goods... are you joking?! That's called efficiency. People spend their careers thinking how to squeeze as much use out of raw materials as possible. Do you think you'd stay in business for long if you discarded half the metal you bought for making machine parts while your competitor discarded only 10% of theirs? Economics is an abstraction of thermodynamics, not the other way around!
One final comment-- ever hear the one about the economist who walks right past a $100 bill lying on the floor? Someone says "Hey, look! A $100 on the floor!" The economist doesn't even look-- he turns to the speaker and says "Nice try. We both know that if there was a $100 on the floor someone would have picked it up by now"
This is not intended as an insult to economists (maybe just a friendly little jab). It's intended to illustrate the point that we're all making the unspoken assumption that we can't possibly be the first people to have seriously considered the possibility of peak oil. Enough people must have already heard of it by now that the market should already have responded to it. The extent to which this is true depends on how many people out there actually have an opinion about peak oil, how quickly their ideas get communicated, how quickly people make up their minds on the issue once they hear about it, how readily they change their minds in response to countervailing arguments, and how quickly their opinions translate themselves into economic decisions. I don't know if these "pre-economic" factors behind consumer/investor behavior are even in the purview of this field or of some other one, perhaps information theory.
Posted by: Al at July 14, 2005 03:14 PM
Well, heck, I support more government support for railroads in the US. However, I would support this even if I thought oil prices were going to go down for a lot of reasons, including the negative externalities involved in the internal combustion engine (global warming and all that). Again, I think markets are imperfect, but not so imperfect that if oil is really becoming scarce that this will fail to manifest itself in rising prices.
OTOH, I am under no illusions that a big push on railroads in the US will save the world regarding the future of the global oil market. People in the US are too addicted to their cars. They will only respond to high gas prices. Then maybe we'll really see a serious switch to hybrids and maybe the development of even other techs using even less oil.
Posted by: Barkley Rosser at July 14, 2005 04:12 PM
One form of "rushing around" about peak oil that I think is very appropriate:
If you have a good idea for a superior energy technology -- better solar cells, whatnot -- please do go out and raise some capital and build it!
The futures markets may not "care" about the future more than 5 years out, but inventors sometimes have to. A big assumption in the economic modelling is that people will innovate. Those innovators very frequently are just the type to be a bit put off by the anodyne "everything is the for the best" optimism of neoclassical economic theory.
Maybe that's you ;)
Posted by: STS at July 14, 2005 04:24 PM
JDH, I certainly didn't mean to imply that Adelman's position is widely held, but it does exist. As far as I can tell, he's basically making the nut room argument.
I'm curious as to your opinions on the research cited by Jeffrey Miller showing that (at least some) markets discount info that is more than five years out.
Posted by: Avo at July 14, 2005 04:38 PM
Avo, the effectiveness of the incentives I was talking about die out at the rate 1/(1+i)^n where n is the number of years ahead the peak in oil production is and i is the interest rate. If i is big and n is reasonably sized (as several of the commenters here and elsewhere are basically suggesting), this may not be a very powerful incentive today. This issue of how i, n, and the necessary lead times for alternative investments all interact is the sort of interdisciplinary discussion that I was suggesting might be very helpful.
Posted by: JDH at July 14, 2005 05:49 PM
So, JDH, it seems to me that are you saying that the answer to your original question "if you thought you were right about the physical scenario, and yet saw oil selling today for $60, how could you explain the situation to an economist, who says that, if you're right, oil should be selling for $180?" is "the effectiveness of the incentives die out at the rate 1/(1+i)^n where n is the number of years ahead the peak in oil production is and i is the interest rate. If i is big and n is reasonably sized, this may not be a very powerful incentive today."
Do I have that right?
Posted by: Avo at July 14, 2005 06:13 PM
Good op-ed piece on oil economics in today's Washinngton Post by Paul Roberts:
Posted by: Avo at July 14, 2005 06:35 PM
No, Avo, you don't. I'm not sure how to communicate more clearly to you.
Posted by: JDH at July 14, 2005 06:36 PM
Hmm ... well in that case I'll stop taking up space on your blog. Thanks for the opportunity to make an attempt at communication between economists and scientists.
Posted by: Avo at July 14, 2005 06:55 PM
More specifically, Avo, the reason the 1/(1+i)^n (where n is the number of years ahead the peak in oil production is and i is the interest rate) formula doesn't help is because Prof. Hamilton's scenario was, "Let's say for discussion we're talking about $200 a barrel two years hence." So if n = 2 then no reasonable value for i could cause this formula to discount $200 to the $60 we see today. The interest rate would have to be nearly 100%, which is not reasonable.
The formula only works to explain the markets discounting a much farther out Peak Oil scenario, five or ten years away. Then the formula would be large enough to significantly weaken the price signals. But if the peak is in a year or two as many Peak Oilers believe (see the poll at http://www.peakoil.com/fortopic9054.html where 3/4 of respondants expect it by 2007 and almost half by the end of 2005!) then the formula is not relevant.
Posted by: Hal at July 14, 2005 07:34 PM
"the effectiveness of the incentives die out at the rate 1/(1+i)^n where n is the number of years ahead the peak in oil production is and i is the interest rate. If i is big and n is reasonably sized, this may not be a very powerful incentive today."
You are describing what how the market would discount the future if the market discounted the future correctly.
The whole point of the Pollet paper, however,
is that the market does NOT discount events that will occur more than three to five years in the future correctly.
Correctly forecasting 5 to 10 year demographic trends - which is not hard to do - will (according to Pollet) yield excess returns NEXT year - not in five years.
As the last line of their abstract summarizes: "The results [of the study] are consistent with short-sightedness with respect to long-run information."
So my answer to the question you originally posed is that economic research shows that the market doesn't deal correctly with what will happen more than a few years out and that you CAN profit from this.
If this is the case, then relying on market signals as a guide to long ranged policy analysis appears shortsighted.
Posted by: Jeffrey Miller at July 14, 2005 07:40 PM
I thought I'd be outa town today, but all we did was load the truck ...
STS encourages us to invent. I'm all for that, but I always like to add a caution, based on my 20+ years in high tech engineering. The short story is "look for revolutions in neglected fields, look for slow/hard itterative improvement in popular and competetive fields."
New engineers, breaking ground on something like solar cells will make huge improvements. Once you get a few dozen teams working, in a few different countries, the easy gains will be quicly found. From there on it is usually (until something comes in again from left field) a slow hard grind.
Check out the solar cell improvement curves ($/watt). They fell fast and then leveled in the 70's and 80's. Only now, when left field ideas like "nano" and "thin film" are bandied about *might* we get another significant drop.
I *don't* mean this as a direct support/proof for peak oil ... but I personally am concerned with a lot of our "alternative energies" because the *have* gotten so much attention in the last 30 years, without breaking through to success.
It gets harder and harder to find an idea that hasn't been tried.
Posted by: odograph at July 14, 2005 08:04 PM
BTW, it is interesting to note that the Ford Model A got between 20 and 30 mpg. Some sites peg it right at 25 mpg. Amazingly enough, 75 or so years later, our consumer fleet mileage (20-22 mpg?) is at or below that. Our best-of-breed four-seaters (apples-to-apples) barely manage to double the Model A's fuel economy. Some of this is slow, hard, work.
(Maybe if you kept the modern cars down to Model A speeds they'd do a shade better in mpg, but only a shade.)
Posted by: odograph at July 14, 2005 08:12 PM
Let me take this opportunity to thank all of you for participating in and contributing much to this discussion. In order to keep this file a manageable size, and to make sure that subsequent discussion proceeds linearly (so that everybody can see which comment follows which and where the discussion has developed most recently), I'm closing off comments on this post, in hopes that you will resume discussion on the next one that continues this topic. Please feel free to pick up again there any topic from this thread right where it was left off here. Again, thanks much for your contributions.
Posted by: JDH at July 14, 2005 10:22 PM