The Corner

Krugman on Climate I: The Cost-Benefit Analysis

Paul Krugman had a long article in the Sunday New York Times Magazine on climate change. Refreshingly, it attempts to compare the costs and benefits of proposed climate policies rationally. But it also, in my view, loads the dice in certain key passages. In the interests of not making this post endless, I’ll discuss only two points today: (1) It asserts an expected level of global warming that is about twice as large as that provided in the current IPCC Assessment; and (2) it appears to compare a resulting inflated damage estimate in the year 2100 to a cost estimate that we would start to bear by 2050. In sum, a corrected simple cost-benefit comparison of a policy to reduce greenhouse gas emissions indicates that we should expect the “solution” to cost more than the damages it averts.

In the article, Krugman says this about how much we should expect temperatures to rise:

At this point, the projections of climate change, assuming we continue business as usual, cluster around an estimate that average temperatures will be about 9 degrees Fahrenheit higher in 2100 than they were in 2000. [Bold added]

Later, he says this:

Nordhaus has argued that a global temperature rise of 4.5 degrees Fahrenheit — which used to be the consensus projection for 2100 — would reduce gross world product by a bit less than 2 percent. But what would happen if, as a growing number of models suggest, the actual temperature rise is twice as great? Nobody really knows how to make that extrapolation. For what it’s worth, Nordhaus’s model puts losses from a rise of 9 degrees at about 5 percent of gross world product. Many critics have argued, however, that the cost might be much higher. [Bold added]

According to the currently governing 2007 IPCC Fourth Assessment Report (AR4), expected warming through about 2100 under the typical reference case scenario A1B is about 5F (Table SPM.3), not the 9F asserted in Krugman’s piece. No marker scenario for future population and economic growth evaluated by the IPCC in AR4 has an expected temperature increase of 9F by 2100. In the standard Nordhaus models referenced in article, damage estimates rise super-linearly with temperature; therefore, not quite doubling the estimate of temperature increase from 5F to 9F would more than double the damage estimate from 2 percent of GDP to the 5 percent of GDP that Krugman cites.

The essence of the argument for greater warming is that several new studies project greater warming than those used in the AR4, and that therefore a rational observer should update this forecast and expect a larger temperature increase. Of course, there is a long history of individual studies that project greater future warming than the consensus estimate. These go all the way back to the first major attempt to create such a consensus estimate: the NRC Charney Commission of 1979. [The scientist cited in the Charney Commission working papers who led the climate modeling team that projected much higher warming than the consensus estimate more than thirty years ago was…James Hansen.] In spite of these frequent challenges, however, the estimated value for climate sensitivity — the central scaling parameter that projects how much warming will be created by a given emissions scenario — has remained essentially unchanged for decades at about 3C, from the Charney report through each of the four successive IPCC assessment reports, and remains the estimate in the current AR4.

Synthesizing the various individual studies that have been released over the past several years to determine if projected warming estimates need to be roughly doubled requires integrating knowledge from numerous physical science specialties, software engineering, economics, and various civil and energy engineering disciplines. I’m not competent to do it; but then again, neither is Paul Krugman. No one person or small group is competent to do this — if it were, why would we bother with the immense time and expense of the IPCC assessment process? We’d just go ask that guy to tell us the answer.

Krugman says this about the economic costs to the U.S. of imposing a carbon-mitigation regime like that proposed in the Waxman-Markey bill:

Over all, the Budget Office concludes, strong climate-change policy would leave the American economy between 1.1 percent and 3.4 percent smaller in 2050 than it would be otherwise.

He then says this about the costs to the world of applying a similar regime to the world:

One recent review of the available estimates put the costs of a very strong climate policy — substantially more aggressive than contemplated in current legislative proposals — at between 1 and 3 percent of gross world product.

Consider the global cost assertion first. Krugman does not cite the specific review study, nor does he define a “very strong” climate policy, nor does specify when we should expect costs to reach 1 to 3 percent of world output. This last point is important, because most serious analyses of this subject that I’ve seen project that a mitigation policy will create costs that continue to rise as a percentage of economic output over the course of the upcoming century.

There have been many reviews of cost estimates produced over the past few years. I’ll purposely choose one produced in 2009 by Resources for the Future (RFF), since it is a moderately left-of-center, well-respected environmental organization. Its estimates are broadly consistent with those produced by many similar bodies. This report provides estimates for the costs of maintaining a global concentration of CO2 at 450 ppm and 550 ppm. I’ll take the more stringent of these two (450 ppm) as representing a “very strong” climate policy. Finally, I’ll use the RFF estimate for the “least cost” estimate (i.e., the most efficient possible policy that assumes, for example, global coordination around emissions reductions without any more realistic geopolitical complexity). It’s possible that Krugman is using an uncited report with very different projections than RFF, but by the choice of study sponsor and political assumptions, I’ll, if anything, bias toward low cost estimates.

According to the RFF review, global cost of the 450 ppm policy should be in the range of 3 percent of gross world output in 2050. But by 2100, this is projected to increase to a cost of about 6 percent of gross world output (with a very wide range of estimates). This is very different than costs of “between 1 and 3 percent of gross world output,” for two reasons: (1) If we use damages as of 2100 in a cost-benefit comparison, we need to be consistent about dates, and (2) we should expect to bear large mitigation costs for decades prior to realizing any expected benefit in the form of avoided damages, which has an enormous impact on the expected value calculation in the cost-benefit analysis.

Krugman says this about a simple comparison of the costs and benefits of a global policy to reduce emissions:

Despite the uncertainty, it’s tempting to make a direct comparison between the estimated losses and the estimates of what the mitigation policies will cost: climate change will lower gross world product by 5 percent, stopping it will cost 2 percent, so let’s go ahead.

But of course, properly considered, he has the numbers exactly backward. As per the first point in this post, if we use current IPCC temperature forecasts, a better estimate for the expected global economic damages caused by human-induced climate change in 2100 is more like 2 percent, rather than the 5 percent asserted in Krugman’s article. And as per the second point, a much better estimate for the economic costs of mitigation in 2100 is about 6 percent, not the 2 percent that Krugman asserts.

While Krugman immediately (and correctly) says such a comparison would over-simplify the real decision logic, it does illustrate the crux of the problem for advocates of aggressive efforts to reduce emissions: (1) For at least the next century, the expected damages caused by global warming are nothing like the doomsday pictures presented in the popular media, but are rather a few percent of GDP, and (2) the economic costs imposed by aggressive programs designed to avoid these damages are expected to be far greater than the damages that they should expect to avoid for the next 100 years at least.

And to be a little parochial about it, from the perspective of a U.S. taxpayer, the net expected economic damages to the U.S. from global warming are expected to be materially zero through 2100, while the costs, as indicated in the Krugman article should be about 2 percent of GDP by 2050. As per the prior analysis, mitigation costs should be higher by 2100. That is, even under some very favorable assumptions, citizens of the United States are being asked to sacrifice several percent of total income (i.e., trillions of dollars of cumulative consumption) for no tangible expected benefit for the next 100 years.

Introducing the nuances of very long-term considerations, geopolitical complexity, the importance of risk and uncertainty and so forth is the subject of the balance of Krugman’s article. I’ll take these up in the next post.

Jim Manzi is CEO of Applied Predictive Technologies (APT), an applied artificial intelligence software company.


The Latest