« Worst. President. Ever. | Main | What We've Learned »

Good short link

In many discussions about Peak Oil, economists are smug that the market will provide an alternative when oil gets too expensive. This, of course, focuses purely on the most pedantic definition of "alternative" - which includes things like demand destruction to the point of worldwide economic collapse - which is probably not what most people have in mind when they think of "solution". Most "cornucopians", as they're called, believe the Magic Of The Market will provide new energy technologies when rising oil prices make them profitable; which ignores the reality that there is no energy source out there with anything approaching the "energy balance" of oil - i.e. the difference between the energy you get out vs. the energy you put in.

However, it's been difficult to express this - for some reason, economists believe their soft science over the hard science of physics (specifically, thermodynamics, or in this case, energy return on energy investment).

Here's a good short explanation of the difference between money economics and energy economics - IE, why the market can't beat the laws of thermodynamics - for those Peak Oil cornucopians.

Fuel cells won't save us. Electric cars won't save us. Only taxing carbon and doing it quickly has any hope - and even then, it'll be using the market to move from today's default setting of "government provides more incentives for you to waste energy than to save it" to "energy becomes very expensive, giving you automatic incentives to conserve".

This entry was posted in the following categories: Economics


If the economists are cornucopians, the peak oilers are malthusians. I'll bet on the cornucopians -- has a malthusian prediction ever come true?

There is plenty of energy. Extracting it and converting it to usable form in a safe and efficient way is the problem, and it is a technological problem. There is no iron law of thermodynamics that says that wind power, sunlight, etc. can't be harnessed efficiently and on a large scale. That technology does not exist today. It is possible that it will not exist when we need it. But it is at least equally possible that it will, based on our track record so far.

This issue doesn't really come to a head until we decide what we're going to do about it. We can hurt ourselves by being too pessimistic just as we can hurt ourselves by being too optimistic. Should we take costly steps today that will shave $100 billion (or whatever) off GDP in order to get ready for a potential scarcity in 10 or 20 or 30 years? It's a gamble. Like any gamble, it depends on the odds and the payout. I think the odds are in the cornucopians' favor.

PS, about the worst thing for technological progress are subsidies that lock in a fake solution. Ethanol, for example. It is turning into a huge boondoggle (I'm talking US and corn, here, not Brazil and sugarcane). But the worst thing about it is it discourages investment in other potential technologies, since they must not only outproduce ethanol, but they must outproduce ethanol's subsidized price.

Yes, ethanol sucks. But you're falling back into soft science trumping hard science in your first comment:

"There is no iron law of thermodynamics that says that wind power, sunlight, etc. can't be harnessed efficiently and on a large scale."

We know today how much energy can be harnessed if we develop 100% efficient ways of doing so, because we have long since had the ability to measure that energy input. There isn't enough energy there - not to support anything like the current American lifestyle (driving big SUVs one-at-a-time from exurb to exurb, I mean). Then stack on top of that the need for energy storage (hydrogen, fuel cell, battery, whatever).

A lot of cornucopians believe we'll be able to drop Energy Tech N+1 right in place of gasoline. It ain't gonna happen - even at 100% efficiency solar cells and batteries beyond that we've ever imagined, we simply won't be able to store as much energy in as little space in such a usable form as nature did with petroleum.

Oh, and the "technology will save us because it always has in the past" argument is a common logical fallacy - I forget what it's called, but it boils down to the "since the societies in which it didn't work went extinct, they aren't around to tell us otherwise".

"The 'technology will save us because it always has in the past' argument is a common logical fallacy"

The "we won't have technology to save us because we can't think of it today" is equally a fallacy.

To be fair, I am not claiming that it is inevitable that we will have the necessary technology in the future. I think there is a good chance that we will, based on our track record of innovation. A better chance than not having it. I don't think it's a fallacy to make bets based on a track record. You've got to estimate the odds somehow. Malthusians and their linear projections have been wildly off target in the past. True, they could be accurate this time, but I wouldn't wager too much money on it.

I honestly don't understand the basis for your physical predictions. Why can't cars be fully powered with rechargeable batteries some day? I don't dispute that powering cars might be more expensive relative to other energy uses, such as heating the home. But how much more expensive? 5 times more expensive than today, or 50 times? If it's 5 times as expensive, expect a bunch of small, energy-efficient cars. If it's 50 times, then you'll see more dramatic changes, including dramatic changes in housing patterns, no question. But there is no reliable basis for predicting one over the other, much less pinning a prediction down to any date. (When will we see this? In 10 years? 50 years?)

In any event, assuming your predictions of physical limitations are correct, we're still left with a "soft science" question: How should we allocate resources today based on anticipated scarcity tomorrow? That's economics, not physics.

Finally, although I wouldn't be willing to spend a lot of money, it would be worth something to hedge the bet. Certainly, government should not be _encouraging_ energy dependence -- e.g., encouraging sprawl through unnecessary land-use restrictions or untolled roads.

The risk of oil shocks is a better to reason to hedge; there's a large risk of shocks, even if they won't necessarily last that long.

I think that ultimately there will be an energy technology that is at least as good as fossil fuels.

The question is, will one of those technologies be ready soon enough to head off serious global economic stagnation due to decreasing and/or more expensive fossil fuel production? My gut says yes, because when things start getting bad, we'll really start conserving. Many industries don't use energy very efficiently right now. It's pretty hard to predict the answer to this question.

I'll add that there is tremendous potential in nuclear energy: an area that humanity is still pretty immature in harnessing.

"I honestly don't understand the basis for your physical predictions. Why can't cars be fully powered with rechargeable batteries some day?"

Energy density; and the fact that there's a well-known maximum (at 100% efficiency) to solar and wind power. Not enough power in; and not enough space in which to store it.

Despite what you hear from cornucopians, battery technology has been growing far slower than it needs to to be able to drive anything like today's vehicles anything like how they're driven today. Read my crackplog on plug-in hybrids for details on that angle (remember, I drive a Prius):


I'll admit that it's possible there's a nearly 100% efficient energy storage mechanism out there waiting to be discovered; but it is very very very unlikely given what we know about chemistry. As it stands today, people are being suckered by interests which want you to believe that fuel cells / hydrogen cars are just around the corner so you won't go buy a smaller more fuel-efficient gas car. Of course, plug-in hybrids have the same problem - some people are pushing them because they honestly believe; while others are just thrilled at the chance to sell more electricity at night even if it ends up being a dead-end.

BTW, can either one of you guys remember the name of the logical fallacy I was referring to? I can't find it.

"I think that ultimately there will be an energy technology that is at least as good as fossil fuels."

I HOPE so, but the compelling argument by the citee was the fact that fossil fuels basically represent the last N,000 years of insolation, so in order to find something better, you have to find another place where a whole bunch of solar energy was stored a long time ago for our use.

One could make a similar argument for uranium, and some have noted that nuclear fission, at least, doesn't have enough fuel lying around to be a long-term primary power solution for the whole world.

Fusion, of course, is a different story. But watch out for that damn Doc Octopus.

I scanned Wikipedia and think it's an enthymeme, an assumed minor premise.

The syllogism takes this form:

1. In the past, technological advances have always allowed us to meet our energy needs.

2. We will have energy needs in 10 years.

3. Therefore, technological advances will allow us to meet our energy needs in 10 years.

The assumed minor premise is: If technology has always provided innovations in the past when we need them, then it will always provide innovations in the future when we need them.

I'm not sure that the minor premise itself is a logical fallacy. I think it is an empirical statement about how the world works. I mean, what has happened sometimes is evidence of what will happen. Whether this particular assumed premise is true or false depends on the evidence. (I think it's false, but a weaker version is true. )


I could have sworn I saw a different form - the one paraphrased by "all these things worked out for us and everybody else we see today, so things'll continue to work out" which ignores the fact that we don't see the ones that didn't work out.

That's a heck of a thing to try to google, let me tell you.

Better to phrase it this way:

We (current Western society) are the end-result of a chain of successful transitions from one energy source to another (first from human power to animal power; then to biofuels; then to fossil fuels).

There were societies that did not manage such a transition; but they died out, so we don't think of them much today.

The fact is that we HAVE to be the result of a chain of successful transitions - this is not chance - because the ones that didn't succeed didn't survive. "We" wouldn't be here to talk about it otherwise.

The term that comes to mind is "selection bias." I think that's what a statistician would call it, anyway. E.g., drawing inferences from data slanted to favor of one conclusion.

If you assume that technological innovations are essentially independent, random events, then a good analogy would be evolution. You cannot argue that each organism will evolve to adapt to environmental changes. Some will, and they will thrive. Some will not, and they will die out. We're left with the winners.

Technological innovation is different. It is largely a function of the (i) the money and human capital we invest in it and (ii) the present state of knowledge. We can produce more if we're willing to pay for it, although there's more uncertainty in technology production than in car production. (I'll admit that my thinking on this has been affected by Warsh's "Knowledge and the Wealth of Nations." It covers the revolution in the economics of technological innovation over the past 15 years.)

I'd argue that it's not that different. For instance, if very high oil prices hit before the magical alternatives that will allow the suburban lifestyle to continue; Western Europe will become far more competitive with the US and China - they don't use much oil and much of what they DO use is more easily substituteable.

There are bounds to innovation - and I prefer to at least think of these physical bounds before assuming that innovation & the market will save us. For instance, I can offer you an infinite amount of money today to break the laws of thermodynamics - you still can't do it. Economics fails when confronted with physical science.

Likewise, you can't build an energy collector which operates at more than 100% efficiency no matter how much money you throw at it.

The other thing is that economics must also operate within the land of politics - i.e., if/when gasoline hits $5/gallon, we'll see calls for direct subsidies for gas prices from desperate suburbanites - I guarantee it.

Even if we could find unlimited supplies of energy with which to run our SUVs forever and ever to keep commuting 50 miles each way to work, with absolutely no pollution, it would still be a bad idea for a myriad of other reasons - health issues, time lost, and efficient use of space, among others.

Whether we run out of our precious and oh-so-delicious supplies of oil or not, we still need to change how we live.

"you have to find another place where a whole bunch of solar energy was stored a long time ago for our use"

And that's basically any form of nuclear fission (only it was some other star, not the sun, that stored the energy).

And nuclear fusion (even better once we get it to work safely) is basically extracting energy that was, in a sense, stored by early-universe activities on a cosmological scale.

Fission: Google on "peak uranium". It's a stopgap at best - if we attempted to replace essentially all oil/gas with fission, it'd take so many plants that we'd be using up the easily recoverable uranium fairly quickly.

Fusion: probably enough laying around to be usable for the life of our species, if we can ever get it to work. It's been 10 years off for 40 years now; and then you still need to figure out the problems of energy storage for portable users like cars.

Neither one is coming quickly enough to save us from Moderate Peak Oil (I'm no Roger Baker, as you know, but I'll bet even odds that 5 years from now we've seen oil above $120/barrel).

I wonder if running out of oil and a suitable alternative will in fact lead to worldwide economic collapse. After all, we did have an oil-free economy before 1850. It might be a terrible transition, but we can go back to living on nuts and berries.

Wow, Dan, been a while.

The problem is that the 1850s development pattern is now extinct in this country (although it survives in Western Europe). The suburban neighborhoods up near our former employer-in-common, for instance, would not have been feasible before the era of cheap oil began; nor will they be after it ends.

It's even worse than you think. In addition to peak oil, we also have peak natural gas to worry about. And the evidence seems to indicate that the availability of natural gas, once peaked, will deteroriate even faster than oil.

So we'll be getting a double whammy on transportation and power plants.

I think we can see the dim shape of how the near future might be bad. As oil prices keep rising, so will transportation costs. That will affect the price of goods, probably provoking unbridled inflation.

Now, mix in ever increasing power prices due to natural gas price escalation. Mix in stagnating economic growth in general due to rapidly dropping population growth.

If we get into a period of stagflation like that, all it would take is a regional crisis of some sort to provoke catastrophe. After all, that is one of the "benefits" of globalism - every nation's lives and economies have now been linked together. What formerly would be a contained regional event could now result in a chain reaction of unfortunate proportions.

Even a semi-objective free market capitalist such as Matthew Simmons (http://www.simmonsco-intl.com/research.aspx?Type=msspeeches) is skeptical about the appearance of alternatives to save us from peak oil.

Let's say the cost to the average driver of driving one mile is $1. (This includes cost of gas, cost of vehilcle depreciated over 100,000 miles, interest cost, maintenance costs, and reasonable charge for environmental externalities). The average hourly wage is about $17. This means the average worker must work about 4 minutes to drive one mile in a car.

How much will it cost in man-hours to drive one mile in 2037, using the then-prevailing technology? 8 minutes? 1 hour?

The cost of a specific resource is not the only variable that matters. Increases in productivity also matter. Suppose that mile costs 1 man-hour in 2037 if productivity growth is flat. Triple productivity and that mile will only cost 20 minutes.

I'm ripping off William Nordhaus and his study of the cost of illumination. (See http://econ161.berkeley.edu/Comments/FRBSF_June11.html for a more general discussion by Brad DeLong.) It used to cost hours of labor to light a room for one hour. Now it costs a few seconds. That's the cost that counts, not the relative cost of candles and incandescent bulbs.

I can't prove that we'll successfully adapt or innovate our way out of trouble. I'd bet that we will, at least in the long run, at least if we reckon costs in man-hours. (As I admitted before, oil shocks -- sudden interruptions in supply -- are a different story. But they are not the peak oil story.)

There's far more elements than just uranium that can ultimately be used for fission, but you are correct that fusion is the far more promising source, if it can be controlled.
I'll add one Bold Prediction regarding long term energy sources - one day, once the development costs make it worthwhile - I expect bioengineered bacteria to play a major role in extracting/creating fuels. Especially heavy elements (for fission) and hydrocarbons.

Post a comment