« The economics of polygamy, continued | Main | The future of macroeconomics »
Cool It, by Bjorn Lomborg
That's Bjorn Lomborg's latest book on global warming. He has good arguments against the exaggerations of others, so this book is worth reading. I cannot, however, agree with one of his central claims, namely that the most serious economic research favors only mild remedies or sees the problem as only a moderate one.
I would instead claim the following:
1. Policy recommendations are extremely sensitive to the choice of discount rate, and economists do not agree on this issue. Furthermore most economists do not even know enough moral philosophy to understand the issues involved (and the philosophers don't understand enough economics), so there is no coherent consensus one way or the other.
2. The most current economists' word is from Martin Weitzman; he argues that the very high costs of the worst-case scenarios suggest an insurance-based case for significant worry, more worry than Lomborg suggests. A Salon review notes:
Harvard's Weitzman puts the current concerns of many economists clearly. Based on the findings of the U.N. climate panel, he notes that with roughly 3 percent probability, "we will [live in] a terra incognita biosphere within a hundred years whose mass species extinctions, radical alterations of natural environments, and other extreme outdoor consequences of a different planet will have been triggered by a geologically-instantaneous temperature change that is significantly larger than what separates us now from past ice ages."
3. We spend too much time wondering about what is "most believable" and not enough energy worrying about the expected value of pending losses. The major critical reviews all nail Lomborg for neglecting this point. That said, the speed with which the negative reviews of the book move to the extreme cases is itself noteworthy, and it does not exactly correspond to the image presented to the public.
4. Given that the value of risk is context-specific, economists are bad at taking the value of insurance from market data in one setting, and then transplanting that estimate to another setting.
5. The strongest argument against significant action is not from cost-benefit analysis in the narrow sense, but simply that we are not very good at producing international public goods. Especially when it comes to extended, intertemporal collective action problems directed against small probability events, with unclear periodic feedback, and dealing with the Chinese and the Indians, who feel they have the right to pollute as much as we did, and also with the not-nearly-as-cooperative-as-they-might-sound Europeans (how's that sentence for a mouthful?).
This argument sounds immoral and indeed perhaps is immoral -- "we're ruining things for others, yet if we tried to fix things we would ruin the fixing, so let's do nothing." Yet I do not think this issue should be disregarded. If I can't open up my computer, dissemble it, and then put it back together again, surely my repair plans should take that fact into account.
Here is a Jonathan Adler review of the book. Here is a very critical Tim Flannery review. Here are critical remarks from Chris Mooney. Here is the Salon review.
Remember that line from Dirty Harry?: "Do you feel lucky, kid?"
Posted by Tyler Cowen on September 10, 2007 at 06:52 AM in Science | Permalink
Comments
I do have to say that any reviews of a book on economics written by non-(real-)economists is goint to be worthless given the Caplan biases.
Posted by: Russ Nelson at Sep 10, 2007 8:17:00 AM
I think the problem with using expected value of pending losses is that as long as there's any probability of the infinite loss of the world ending above zero, then we should devote almost all of our resources to averting it
I think that the probability of people hijacking AGW for their own tastes is likely to blow up any expected value calculation and make adaptation the way to go
Posted by: Pat at Sep 10, 2007 9:03:02 AM
The line is, "Do you feel lucky, PUNK?"
Posted by: David R. Henderson at Sep 10, 2007 9:29:48 AM
Actually, this quote from Dirty Harry is one of the most misquoted ones as it is, in fact, a double quote. Harry Callahan says: "You've got to ask yourself one question: 'Do I feel lucky?' Well, do ya, punk?".
Posted by: Samuel T. Petursson at Sep 10, 2007 10:11:38 AM
I do have to say that any reviews of a book on economics written by non-(real-)economists is goint to be worthless
Modulo subject matter, many climatologists feel the same way, oddly.
Posted by: Kieran at Sep 10, 2007 10:16:06 AM
I saw an interesting seminar by Dave Rutledge last week where a guy estimated the amount of fossil fuels left underground and showed that there wasn't enough to match the carbon emission scenarios being bandied about. Potential moral of the story -- you can't be both a peak oiler and a global warmer.
Posted by: david at Sep 10, 2007 10:26:07 AM
Since reading about biochar/agichar/terra preta (I wish that people would settle on one name for it) I do not worry. This method would allow us to reduce the amount of co2 in the air at any point, so why not wait until we see some real non-ambiguous negatives before we act.
Posted by: Floccina at Sep 10, 2007 10:29:34 AM
Anyone who uses insurance rates to estimate actual probable loss figures on an event which has never occurred in human history is a fraud. The insurance companies have no history upon which they can base their expectations. Indeed, a worst-case scenario would likely bankrupt the company long before the policy criteria kicked in, so these are sucker offers. The reason rates are high are either because they are afraid of the probability of spurious lawsuits/publicity nightmares or because their estimates of the probability have been driven up by the shysters who WANT the numbers to be high to score political points.
Posted by: Nathan Zook at Sep 10, 2007 10:31:09 AM
Well, the biggest argument against action is...action against what exactly? There is no real evidence that our modeling can predict climate change on interdecadal intervals - no evidence at all - and in fact a lot of evidence to the contrary. For those with a numerical bent, I suggest scanning Steve McIntyre's Climate Audit blog, where one can get a flavor for how sloppy even the most basic temperature measurement and analysis have been. The sad answer - really, really sloppy.
It's fair I suppose that economists accept as given the risk estimates of the IPCC as a given and then fiddle with the discount rates, but what economists don't get is that the uncertainties in the uncertainties of these models make them worthless for predicting tail risk 50 years out. So, action based on these over-fit models is not warranted.
One final point - the scientific prior would be that CO2 increases will produce modest warming on the planet, but unequally in colder, drier portions of the globe. This is in fact what is being observed. The poles show a lot of warming, the US shows zero since 1930. Absent Al Gore scenarios, do we care about that? Perhaps CO2 is stabilizing a cooling trend in the civilized portions of the world - isn't that good, or can their only be losers here? The Canadians seem pretty jazzed by this whole warming thing, for example.
Posted by: JPC at Sep 10, 2007 10:34:19 AM
Great list!
Number 1 and 5 are far and away the two most important things people should be thinking about in relation to this problem.
The great thing about what economists are doing on the issue of climate change is the enormous attention given to working on prescriptive arguments/theory. However the utter weakness and non-systematic nature of the underlying normative assumptions driving economic choice theory is shown to be a total disaster in the intergenerational problem of global warming mitigation.
Political theorists certainly have a better grasp on the normative aspects, but not enough of them take seriously #5. But much more importantly those that do take #5 seriously do not have support from political science for coming up with prescriptive arguments for what we can do institutionally to improve prospects for global collective action in this area. The lack of interest and tools in political science for the kind of prescriptive theorizing economics is so comfortable with is a disaster on par with the normative disaster of economic choice theory.
OK, OK, I got too excited. Maybe the situation is not disasters left and right, but things are not looking very good.
:(
Posted by: aaron_m at Sep 10, 2007 10:43:58 AM
I think the problem with using expected value of pending losses is that as long as there's any probability of the infinite loss of the world ending above zero, then we should devote almost all of our resources to averting it.
But there are an indefinite number of possible, non-zero probability causes of the 'infinite loss of the world'. Shall we devote almost all our resources to detecting and preventing a catastrophic asteroid collision? Or preventing a nearby supernova? Or to preventing nuclear armageddon? Or to preventing global warming?
However, is there really 'roughly a 3 percent' chance that within a hundred years we will face a more severe change in climate than experienced in an ice age -- as is claimed? I'm with Freeman Dyson in being skeptical of the state of climate modeling. I don't think anybody understands the global climate well enough to make such an estimate with any confidence. And if we take history as a guide, the chances of 'greater than ice-age' climate change within a given century are much closer to vanishingly small than 'roughly 3 percent'.
Posted by: Slocum at Sep 10, 2007 10:44:28 AM
I don't understand why people are unwilling to believe the predictions of a scientific field that can't
perform controlled experiments, about events that will happen far enough to the future that we can't know
what technology will exist then.
Posted by: Person at Sep 10, 2007 11:11:22 AM
TC,
While I very much agree with number 5, you frame it in such a way as to make us think that the Chinese and Indians (and the other unnamed billions of dirt poor people plausibly moving away from poverty very soon) are a bunch of selfish, evil scum for wanting to end their poverty much like we did.
The simple fact of the matter is that the more developed countries emit an order of magnitude more GHG's per capita than these poor countries, and these poor countries have far more capita on its way to economic growth.
If the rich countries won't spend the money to reduce their per capita emissions 90+% below where they are today, then they are the morally culpable ones, not India, China and others. The notion that the rest of the world should morally have to stay ridiculously poor so that we don't have to reduce our very high standard of living is absurd.
Kyoto style emissions trading would only work if everyone on the planet on an equal per capita basis gets vouchers worth the global mean average GHG emissions. This basically means a 90+% reduction in emissions overnight in rich countries, which isn't going to happen. Alternatively it means a massive transfer of wealth from rich countries to poor countries, and that we (rich countries) trust them to keep their word that they won't emit more in the future (ha!). This won't happen either.
Similarly, a Pigou tax on GHG emissions in rich countries won't do diddly either (although we definitely should transform on a $ for $ basis all energy taxes we already have into carbon taxes), because still more GHG heavy industry will move overseas into countries like China and India.
What we need is to get green technology that is cost competitive with coal, gasoline etc. otherwise the poor countries simply won't adopt them (even then we'll have to pay them to replace legacy infrastructure). Only the rich countries can get that to happen, not China and India.
Posted by: happyjuggler0 at Sep 10, 2007 11:20:54 AM
A response to JDC's post: you're making a bunch of classic denier claims that don't stand up to scrutiny.
First, when you say that there is no evidence that current models can predict climate change over decades, what is your standard of proof? If you require that we wait a few decades and see how a model did, then by definition no current model can be proven correct. If you allow for historical back-testing, current climate models do quite well. You even seem to contradict yourself, since in your last paragraph you claim that observed changes are meeting predictions.
Second, you reference McIntyre's work. This is a classic anti-science technique: in a huge field of work, find a few bad data points, then trumpet them as evidence that the entire field is worthless. I imagine someone, on hearing that a forest is healthy, hiking in for a bit, then coming back and pronouncing "It's not healthy! I found a dead tree!" Historical work from weather stations is just one of many, many ways that the current warming trend has been documented. If you don't like weather station data, how about ice core data? or melting permafrost? or retreating glaciers? or satellite measures of upper atmosphere temperature?
Finally, the claim that the U.S. shows zero warming since the 30's is deliberately misleading. If you choose the hottest decade in the record as your starting point, you can manipulate the data to show what you want. Instead of starting in the 30's, why not pick any other decade in the record? By analogy, it's pretty easy to look at yesterday's stock market and say "If I bought here and sold there, I'd have made money." That says absolutely nothing about whether the market was up or down yesterday.
Posted by: TomR at Sep 10, 2007 11:42:10 AM
"What we need is to get green technology that is cost competitive with coal, gasoline etc. otherwise the poor countries simply won't adopt them (even then we'll have to pay them to replace legacy infrastructure). Only the rich countries can get that to happen, not China and India."
This the way economists always misunderstand #5.
The argument from technology is not compelling in the two interpretations we can make of this claim.
On the first interpretation the argument is just blind faith in some new, dramatic, and unexpected technological discovery that will radically change the global collective action problem. It is faith in a technology that will be so cheap to discover, invest in, build infrastructure for, etc..., that it will be a no brainer. Those that can will just adopt it for self-interested reasons and will even be willing to pay off the huge vested interest in the existing energy system. This is faith in something we have no reason to expect to happen.
On the second interpretation the argument is a failure to understand that #5 means that we currently lack the incentive structure for investment in adopting new technologies, improving alternative energy technology, investments in new infrastructure, etc.... The global collective action problem that we face is exactly to generate the incentives for technological change.
Given what we now have in terms of alternative technology and the improvements we can reasonably expect to make through technological innovation we could clearly affect major GHG emissions reductions. But in current market conditions these technologies and reasonable expectations in terms of return on investment for improving alternative technologies do not generate significant profit based incentives for the needed transition. So it does not make any sense to say 'well we can't address the collective action problem, lets focus on technology instead.'
Posted by: aaron_m at Sep 10, 2007 11:45:27 AM
I don't understand why people are unwilling to believe the predictions of a scientific field that can't
perform controlled experiments, about events that will happen far enough to the future that we can't know what technology will exist then.
I take it you mean economics.
Posted by: Kieran at Sep 10, 2007 11:59:04 AM
Is this your One Percent Doctrine?
Posted by: Chris at Sep 10, 2007 12:00:31 PM
"Is this your One Percent Doctrine?"
?
Posted by: aaron_m at Sep 10, 2007 12:05:00 PM
“I cannot, however, agree with one of his central claims, namely that the most serious economic research favors only mild remedies or sees the problem as only a moderate one.”
Really? Here is what William Nordhaus says in an article published in the same issue of the J. Econ. Lit. as Weitzman’s paper (and BTW, how exactly does Weitzman’s paper attain the status of representing “all economists’” views??).
“How much and how fast should we react to the threat of global warming? The Stern Review argues that the damages from climate change are large, and that nations should undertake sharp and immediate reductions in greenhouse gas emissions. An examination of the Review’s radical revision of the economics of climate change finds, however, that it depends decisively on the assumption of a near-zero time discount rate combined with a specific utility function. The Review’s unambiguous conclusions about the need for extreme immediate action will not survive the substitution of assumptions that are consistent with today’s marketplace real interestrates and savings rates.”
Nordhaus goes on to note:
“These [the Stern Report] results are dramatically different from earlier economic models that use the same basic data and analytical structure. One of the major findings in the economics of climate change has been that efficient or “optimal” economic policies to slow climate change involve modest rates of emissions reductions in the near term, followed by sharp reductions in the medium and long term. We might call this the climate-policy ramp, in which policies to slow global warming increasingly tighten or ramp up over time.”
Other economists have made similar criticisms. Richard Tol, a leading environmental economist, had this to say about Stern:
“In sum, the Stern Review is very selective in the studies it quotes on the impacts of climate change. The selection bias is not random, but emphasizes the most pessimistic studies. The discount rate used is lower than the official recommendations by HM Treasury. Results are occasionally misinterpreted. The report claims that a cost-benefit analysis was done, but none was carried out. The Stern Review can therefore be dismissed as alarmist and incompetent. This is not to say that climate change is not a problem, nor that greenhouse gas emissions should not be reduced. There are sound arguments for emission reduction. However, unsound analyses like the Stern Review only provide fodder for those skeptical of climate change and climate policy.”
You are right that there is no general agreement among economists about discounting future costs and benefits for public policy. But Nordhaus illustrates nicely the extreme nature of the Stern Review’s “no discounting” assumption:
“The effect of low discounting can be illustrated with a `wrinkle experiment.’ Suppose
that scientists discover a wrinkle in the climate system that will cause damages equal
to 0.1 percent of net consumption starting in 2200 and continuing at that rate forever
after. How large a one-time investment would be justified today to remove the
wrinkle that starts only after two centuries? Using the methodology of the Review, the
answer is that we should pay up to 56 percent of one year’s world consumption today
to remove the wrinkle. In other words, it is worth a one-time consumption hit of approximately $30,000 billion today to fix a tiny problem that begins in 2200. It is illuminating to put this point in terms of average consumption levels. Using the Review’s growth projections, the Review would justify reducing per capita consumption for one year today from $10,000 to $4,400 in order to prevent a reduction of consumption from $130,000 to $129,870 starting two centuries hence and continuing at that rate forever after.”
Would you be willing to make this trade-off? I wouldn’t, and neither (I suspect) would most voters.
Posted by: mike v at Sep 10, 2007 12:27:05 PM
Fascinating post. Some quick reactions to four of the points:
1. I am neither an economist nor a moral philosopher, but I think that the Nordhaus – Stern debate addressed this issue very clearly, and I think that Nordhaus won with the argument that it is an unrealistic / inhuman standard to expect that we either would or should apply Stern’s social discount rate of 0.01%.
2. Weitzman’s (very smart) series of articles on this subject seem to have as their crucial idea the observation we can never in practice know whether a small sample of a huge population (of unknown size) contains the most extreme cases. This is, of course, a technical version of the Black Swan or Hume problem. Said yet another way, he is arguing that we think we face climate risk, when in fact we face uncertainty. I think this is true, even without believing we have a sample – population problem, since the climate models themselves that are used to create the pdf for temperature outcomes are non-validated. I’ve written a pretty detailed article on this for NRO.
3. There are two ways to think about this: (1) our models allow us to examine the pdf of outcomes or (2) the models are unreliable. In case 1, the DICE-2007 modeling project at Yale has shown that the EV applied across a range of scenarios representing the pdf results are very close (through the next century or so) to the results for the expected temperature case. In case 2, this all boils down to the Precautionary Principle, which, as several other posters, Cass Sunstein in a book, and (far less illustriously) me in an article for National Review, have argued, is a really silly way to think about decision-making in the face of this problem.
5. There is clearly a prudential argument that it seems pretty hard to coordinate international action on this. You should also note that no individual country has apparently been able to do much about this. Is that only short-sighted selfishness, or should that e telling us something about the proposed remedy?
Posted by: Jim Manzi at Sep 10, 2007 12:31:37 PM
If you allow for historical back-testing, current climate models do quite well.
I once heard a talk by an artificial intelligence researcher (this was 20 years ago) who had written a learning program that did a wonderful job of 'predicting' historical trends in the stock market. It performed so well, that he started using it to make investments. At the time he gave the talk, it had just recently failed to predict the stock market crash of 1987.
Before the late 90's you might have built a U.S. budget deficit model that handled historical data beautifully but was completely unable to predict the dot-com driven budget surpluses. Or you might have build a population growth model that failed to predict the drastic decline in global population growth rates that would occur.
There are good reasons to be skeptical about models that are validated by their ability to predict the past. And, yes, I understand historical data is the best that scientists have to work with -- but that is the problem.
Posted by: Slocum at Sep 10, 2007 12:36:42 PM
Good summary, Tyler, as always.
I think it's to some degree valid for China and India to believe they have some right to pollute as much as we did. (It's also somewhat invalid from a standpoint of "them's the breaks.")
Policy solutions include some combination of 1) prodding them to do it anyway, 2) offering large cash transfers and 3) R&D subsidies that have beneficial spillovers, as happyjuggler suggested.
Posted by: mk at Sep 10, 2007 12:56:09 PM
i love the reverse pascal's wager take -- the potential harm is so great, even a minimal chance it occurs means we have to act now. but no one, ever, ever acts on this. why not? because there's only a chance the potential harm is that great, and humans aren't good with probability? or because it assumes what it's trying to prove? e.g.:
pascal, how should i figure out whether it's worth fighting GW?
well jimmy, let's start by assuming the downsides of GW are so great, you'd be crazy not to address it.
um, ok. so if GW is that harmful, i should address it?
exactly, well done, jimmy.
or:
well jimmy, let's start by assuming that the potential downside from GW is so high that any expected value calculation will show it's worth trying to stop it.
next, let's perform an expected value calcultion.
wait a second while i do my sums . . . . somewhat surprisingly, my expected value calculation seems to confirm that it's worth trying to stop global warming.
hmmmmm, let me double check my math . . . . yes, that's what it shows. so there you have it. etc.
Posted by: dj superflat at Sep 10, 2007 1:05:49 PM
“The effect of low discounting can be illustrated with a `wrinkle experiment.’ Suppose
that scientists discover a wrinkle in the climate system that will cause damages equal
to 0.1 percent of net consumption starting in 2200 and continuing at that rate forever
after. How large a one-time investment would be justified today to remove the
wrinkle that starts only after two centuries? Using the methodology of the Review, the
answer is that we should pay up to 56 percent of one year’s world consumption today
to remove the wrinkle. In other words, it is worth a one-time consumption hit of approximately $30,000 billion today to fix a tiny problem that begins in 2200. It is illuminating to put this point in terms of average consumption levels. Using the Review’s growth projections, the Review would justify reducing per capita consumption for one year today from $10,000 to $4,400 in order to prevent a reduction of consumption from $130,000 to $129,870 starting two centuries hence and continuing at that rate forever after.”
I dunno about this example. It's interesting but:
1) To have knowledge of such precision regarding the world of 2200 is wildly implausible.
2) The example seems to undermine the application of Expected Utility in general. Perhaps something like prospect theory (with its intense loss aversion) would yield a more sensible result. Of course, who knows what that would mean for Stern's analysis.
Posted by: mk at Sep 10, 2007 1:07:58 PM
Since there is also a non-zero chance that we face global cooling, why aren't we doing all we can to add CO2 to the atmosphere to prevent that possibility?
(No, I'm not suggesting we do that either.)
Posted by: Shakespeare's Fool at Sep 10, 2007 1:10:01 PM
