I have written before on peak oil, so much of what I am about to say has been said before, but it still seems a good idea to put it all together in one place. Also, I am including some new content, for example, the fact that the entire Alaska pipeline was built after we supposedly reached peak oil in the US, so there will be some new content. In addition, b putting it all in one place I think it will be much more clear for those who are curious about the topic.

As I mentioned before, there is a rather interesting article in the June 2008 National Geographic magazine about peak oil. And, while it is obviously hardly a definitive description of all aspects of the topic, it does quite clearly present the reasoning behind those arguments, and, much more importantly, it shows that my original argument is basically correct, that those forecasting the end of oil production are doing so by confusing known reserves with total possible reserves. Not only that, but it shows that many predictions are founded on an even more spurious basis, and one I did not anticipate.

Let us start by summarizing the basic "peak oil" argument. The argument is that there is some fixed amount of oil in the earth*, and as we extract that oil we will eventually reach a point where we cannot reasonably produce any more oil per unit of time. This point has been dubbed peak oil. After that point, oil production will either plateau or level off.

Now, let us first deal with the concept of peak oil. I will then argue that, even if peak oil is meaningful, the methods used to predict it are incorrect.

First, the whole definition of peak oil reminds me of nothing so much as Malthus' predictions of starvation is population growth outstrips agricultural expansion. And just as with Malthus, the peak oil theory ignores the impact of technology. Even now, wells which were considered exhausted are producing oil through a variety of new technologies. That means that what was yesterday's peak oil could be far from tomorrow's peak. Not just reusing spent wells, but oil sands, coal, oil shale, and a variety of other sources all could yield oil economically in the future. So any prediction of a peak has to make assumptions about future technologies, and all such predictions have been far from reliable in the past**.

A second, and bigger, problem is whether peak oil is even meaningful. Julian Simon pointed this out when he argued that we will never truly run out of any natural resource. The fact is, should we reach peak oil, when demand rises, prices will rise as well. This increased price will make alternatives more economically desirable. Gradually, the rising prices caused by a production plateau will both spur searches for substitutes and for technologies which allow us to extract more oil. (Not to mention exploration for as yet unknown oil sources.) Which means that the price rises caused by a production plateau will both reduce our demand for oil as we use substitutes and may allow us to produce more through technological innovation. In short, peak oil may either prove not to be the peak after all, or else will force us to wean ourselves off oil. Which means that peak oil may not prove the crisis so many are predicting. For a time the price increases may be uncomfortable, as they are proving now, but in the end it may prove no more devastating than the whale oil shortage which moved us to petroleum, or the wood shortage which moved England to coal. Shortages occur all the time, and most often the change to an alternative proves beneficial int he long run.

But let us ignore both of those arguments and assume that peak oil is something we would wish to predict, the question remains, are current predictions of when peak oil will be reached valid? For purposes of this argument, I will leave aside questions of technology, even though those are quite significant, even when looking at no more than a five to ten year time horizon. Technology moves quickly, making predictions based on present methods invalid in a short time. But, for the moment, let us ignore technology and look at the other issues involved.

In an earlier essay I argued that peak oil was based upon currently known reserves, which is a mistake. I actually went a bit farther and allowed that some predictions started form known reserves and extrapolated possible future discoveries. It turns out that I was far too generous. According to the National Geographic article, one of the more dismal predictions of peak oil for Saudi Arabia is based precisely on the overly simplistic methods I described:

Since the mid-1990s he had been studying data from the 250 or so major fields that produce most of the world's oil. He looked at how much crude remained in each one and how rapidly it was being depleted, then added all the new fields that oil companies hoped to bring on line in coming decades.

In other words, he basically looked at oil reserves we currently know and said "We are running out of oil." This is precisely the simplistic reasoning I have argued against in earlier essays, the same reasoning that has lead to forty years of scare mongers telling us oil will run out in 25 or 30 years.

For the benefit of those who have not read my earlier essay, I will reproduce here the argument against that position:

I wrote about this before, but I want to correct one misapprehension, the idea that we KNOW how much oil there is in the earth. Any number you see is an extrapolation from "known reserves". Known reserves, almost without fail, tend to be about 20 to 30 years of oil at current consumption. Why? Because oil companies have no interest in prospecting once they have that much in reserve. It is costly to prospect, and if you have oil to run for 20 or 30 years, why look for more. Well, as oil is drilled, they will continue to prospect for replacement reserves, but that would still need to keep known reserves at somewhere between 20 and 30 years.

That is why since the1960's we keep hearing "Oil will run out in 30 years!" Because scientifically illiterate journalists confuse known reserves with total reserves.

Or, put simply, we always have known reserves of about 25 to 30 years, but that means nothing. The quantity of reserves is almost always 25 or 30 years AT PRESENT CONSUMPTION, which should tell the doom mongers something, that the amount of known reserves is not a measure of the absolute amount of oil available, is not even related to the total amount of oil in the earth, but is, instead a function of present consumption. So, unless you are drawing conclusions about present oil consumption, known reserves is useless.

However, the error of confusing known reserves and total oil in the earth pales in comparison to another mistake, one so bizarre, I would never have imagined anyone making it had I not read it in the article.

I first saw this article in my doctor's office, and almost fell over when I read one quote. My first thought was "I have to write about this". At the time, I forgot that my wife had bought me a subscription to National Geographic for my birthday, so I despaired of ever being able to quote it, but since then I have found my copy, so I can share with you the quote which shocked me:

After geologist M. King Hubbert correctly predicted that U.S. oil production would peak by the early 1970s, analysts adapted his mathematical formula to calculate the peak of world oil.

As soon as I read this, I realized that there is even less basis to peak oil predictions than I had first thought.

Now, I am sure some are wondering what is so damning about that quote, so let me explain. Basically, it argues that all peak oil predictions are based on the work of Hubbert, and Hubbert's claim to fame is predicting that US oil production would peak in the early 1970's. But there is one big problem with that. US oil production did peak a little after 1973, but the reasons for that peak were entirely political, not technical. Which means that Hubbert's prediction was right entirely by accident, and that the remaining theories which base themselves on Hubbert are thus based upon an incorrect model which only happened to fortuitously predict the moment when political forces caused US oil production to drop.

Once again, I covered all of this in an earlier post:

The woes of the oil industry began in the Nixon administration. With the signing of the Smithsonian Accords, effectively taking the US off the gold standard, inflation suddenly became a huge concern for the government. Worried that the rapidly falling dollar would cause economic collapse, Nixon imposed wide ranging price controls. Worse still, from the perspective of the oil industry, Nixon imposed strange price controls on domestic oil production, distinguishing between "old" and "new" wells, favoring the former.Finally, even when he lifted price controls on the rest of the economy, oil remained controlled. As a result, the oil industry virtually stopped exploring. As "new oil" was priced lower there was little incentive to search for it. So, for the entire decade of the 1970's, until Reagan lifted price controls at long last, the oil industry did very little exploration, and known reserves dwindled.

From the 1980's until the present, a second factor has been keeping the known reserves low. That has been environmentalism. As soon as  price controls went away, opening up the possibility of oil companies again exploiting oil fields in the US, environmental groups began to flex their political muscles and shut down exploratory drilling everywhere. As a result, few new fields were added to the already small known reserves.

Which means that while Hubbert's prediction happened to get the date right, his reasoning is still completely wrong. There was no technical reason for oil production to drop in the US in the 1970's. Price controls caused "peak oil", not any technological issues.

For those who doubt this, I would point out a simple fact. Once the price controls were lifted, and prior to the imposition of modern, crushing environmental controls, we opened a number of new wells, including the Alaska pipeline. Had we reached our peak, it is unlikely we would have opened so many new wells. Nor is it coincidental that this sudden burst of drilling took place almost immediately after the price controls were lifted. The evidence suggests very strongly that price controls had much more to do with our lack of new drilling than any technical issues.

All of which brings me back to where I started, my assertion that even if we allow that peak oil is a meaningful term, our present methods of predicting when it will occur are far from plausible. Not that I think it is even a meaningful term. Even if we see a plateau in production, that does not mean it is "peak oil", there may not even be a single "peak oil". We may see a plateau, but future technology may make that temporary "peak oil" only a temporary peak, with future oil production higher still. So what seems peak oil may prove to be nothing of the kind.

And even if there is a "real" peak oil, some point at which we can find no way to extract any additional oil, it will likely be, at worst, a temporary inconvenience. As I said before, we have experienced numerous shortages, of whale oil, of wood at various times in history, of a variety of essential resources. And, throughout history, when we have started to run out of an essential resource, we have found a replacement, and progress has continued. I see no reason to think oil is a special case, so I can only assume, should we eventually start to run out of oil, we will find a replacement. We may suffer some discomfort, prices may rise even more than they have now, but we will find a solution. I know that past events are no guarantee of future events, but the pattern has happened so many times in the past, our response to a shortage has been so consistent, that I cannot help but believe it will be the same with oil.

Of course others will argue that we should start looking for that substitute now, so as to avoid that future discomfort. But that is simply incorrect. First, as I said, we may never really see "peak oil", only a series of transitory plateaus overcome by technological innovation and substitution. Or we may not even see that. But even if we do eventually reach a permanent plateau, it is far more efficient to allow the market to come up with the mix of substitution, economization, and new technologies that pleases the greatest number of consumers, rather than confiscating money now and decreeing a government solution. We may be able to find a substitute energy source by doing so, but at best it will simply produce the same result as the market and more likely it will leave consumers less pleased than had they been allowed to come up with their own solutions.

All of which is a very wordy way of saying "don't worry". Peak oil may never truly arrive, and even if it does, the market will solve it in a more satisfactory way than any of the various "Manhattan projects" that the fans of central planning want to undertake immediately.

-----------------------------------------------------------------------------------------------------

* This point is not really in dispute. The only counter arguments are those raised by the abiotic theories of oil formation, which at present seem unlikely. Other than that, I would argue that, if oil is formed form ancient organic material, there is still oil being formed, so the quantity is not absolutely limited. However, formation is so much slower than extraction that it can safely be ignored, so, effectively, there is a fixed amount of oil available.

** One of my hobbies is collecting old "futurological" books. I love finding books from the early 1960's predicting what the 1970's or 1980's would hold. It is quite a laugh to see how far off they are. I even own some specialized texts, such as computer texts from the early 1980's which try to predict the future of computer technology in the next 10 years. What is interesting is, even within a very limited field such as personal computers, and with a very limited time horizon of 10 years, how far off the experts can be. For example, most computer books I have found from the early days of computing see nothing replacing the magnetic storage devices of that time. And, while it is true we continue to use magnetic hard drives, these predictions fail to imagine anything like nonvolative RAM (NVRAM) or CD/DVD technology. So, if the experts can miss so much when looking only 10 years into the future, I am a bit leery of how well the oil experts can forecast the technologies which will effect when peak oil is reached.

------------------------------------------------------------------------------------------------------

POSTSCRIPT

My previous essays on peak oil and reserves can be found here:

A Thought on Oil Reserves
A Brief Comment on Oil
Why I Doubt Peak Oil Predictions

Other essays relating to oil, and energy in general, can be found here:

In Defense of Speculators
Authoritarian Oil Talk
Stop Big Porcelain Now!
A Shortcoming of Conspiracy Theories
Oil Speculation?
A Brief Comment on Oil
G-d Save Us From Simple Solutions
Absurdities on Oil
Those Darn Speculators
Economic Illiteracy
Obama's (Lack of an) Energy Policy
One Specific Idiocy
A Thought on Solar Energy
Oil Logic
Solving High Food Prices
Several Convenient Untruths
It's Not An Energy Policy
A Question for Those Supporting Green Energy
Life Without Villains

This list may not be comprehensive, but it covers most of the major essays on my blog.

And finally, one recent essay, part of which deals with the proposals to resolve our current energy woes:

What We Deserve

Most of the essay is off topic, but I include it as it is interesting to see the absurdly authoritarian solutions being proposed by nominal conservatives.

POSTSCRIPT II

I know many will point to current oil prices as evidence of "peak oil"'s validity, but I would instead point to oil scare books of the 1970's. It appears that whenever we see a sudden rise in prices, whether from political causes, such as in the 70's, or from increased demand, as today, we begin to hear the fear mongers talking of oil running out. My only question is this: If it is so clear oil is running out, why was no one screaming about this impending doom when gasoline was ready to fall below $1 per gallon? Why do we only get these scare stories when prices rise and audiences become more receptive?

Is it just me, or does it seem that the scare stories seem to crop up only when oil prices are high enough to make them an easy sell?

CORRECTION 08/20/2008

It turns out my memory was incorrect and the Alaskan pipeline was opened prior to the removal of price controls. However, as most of our other drilling still took place in the early 80's after price controls were lifted, my initial argument stands. Then again, the fact that Alaskan oil was being sold while price controls were still in place discriminating between "old" and "new" oil, may explain why most oil went to Japan, as Japan had no such price discrimination. Having been driven into Japanese contracts by such restrictions, and with Japan as close or closer than US refineries, it is no wonder oil would continue to flow there.

In any case, whether opened in the late 70's or early 80's, all of our Alaskan oil still started flowing after "peak oil" was reached, making a bit of a mockery of those predictions.