It appears to me that Joseph Romm makes four core arguments in response to my essay:
1. Manzi is wrong on the facts about IPCC projections for expected warming.
2. The consequences of the actual amount of warming that we expect to experience are horrifying, and are likely to include 80 to 250 feet of sea level rise and a dust bowl from Kansas to California.
3. Martin Weitzman has shown that traditional cost-benefit analysis is not appropriate for the case of global warming.
4. The costs of aggressive emissions abatement are very small.
I believe that each of these arguments is incorrect, and I’ll try to explain why for each one in turn.
1. Manzi is wrong on the facts about IPCC projections for expected warming
In my essay, I said that:
The current IPCC consensus forecast is that, under fairly reasonable assumptions for world population and economic growth, global temperatures will rise by about 30C by the year 2100.
Here is my evidence in support of this assertion.
The IPCC takes the sensible position that establishing a “business-as-usual” (BAU) baseline for making long-term global climate projections is a pretty tricky endeavor because it means figuring out how the population and economy of the entire world is going to develop over the next century or more. Therefore, rather than a single business-as-usual baseline, they have established a wide range of potential scenarios for how the world might develop. They have declared six of these scenarios to be “marker scenarios” that collectively provide a representative spread of realistic possibilities. Each of these scenarios assumes no implementation of the Kyoto Protocols or other climate initiatives. The Working Group I Summary for Policymakers of the most recent IPCC Assessment Report (“WG1 SPM”) provides best estimates for the expected warming through about 2100 for each of these six scenarios (See Table SPM.3 on page 13). These best estimates range from a low of 1.40C for Scenario B1 to a high of 4.00C for Scenario A1F1. Each scenario is considered equally plausible by the IPCC. The straight average of the six scenarios is 2.80C. One of these six is Scenario A1B. This scenario is characterized by “very rapid economic growth, global population that peaks in mid-century and declines thereafter…the rapid introduction of new technologies, …and the assumption that similar improvement rates apply to all energy supply and end-use technologies”. It is often used in both IPCC and external documents as an informal mid-range case. Scenario A1B also has a best estimate for warming by about 2100 of 2.80C.
Mr. Romm says that my statement is “not correct”, and then goes on to provide what he believes to be a more accurate representation of IPCC projections for expected levels of warming by 2100:
…the latest IPCC report finds that, absent a sharp reversal of BAU trends, we are headed toward atmospheric levels of carbon dioxide far exceeding 1,000 parts per million by 2100. IPCC’s “best estimate” for temperature increase is 5.5°C (10°F)…
Note that this asserted projection is much more severe than what I presented (5.50C vs. 30C), and unlike my assertion is based on a BAU trend. He supports this assertion with a general link to this same WG1 SPM document that I have referenced and discussed. I am pretty familiar with this document, and I don’t know of a single use of the term “business-as-usual” within it. Further, to my knowledge every relevant use of the term “best estimate” in the document is either in the table of scenario projections that I referenced, or in Figure SPM.5 (which is a graphical representation of these same projections), or in the footnotes and explanatory text that surround these exhibits. All of these uses of the term support the values for projected warming that I have provided. Further, to my knowledge there is no marker scenario presented in this document or anywhere in the current IPCC Assessment Report under which the best estimate for warming by 2100 is 5.50C. I do not understand the basis for Mr. Romm’s assertions.
I stand by my original statement as an accurate reflection of current IPCC consensus forecasts for expected global warming through 2100, though obviously I remain open to contradiction by relevant facts of which I am currently unaware.
2. The consequences of the actual amount of warming that we expect to experience are horrifying, and are likely to include 80 to 250 feet of sea level rise and a dust bowl from Kansas to California
Mr. Romm next provides a litany of very bad things that he predicts are likely to occur if we do not implement a set of policies to aggressively restrict emissions.
Consider Mr. Romm’s first prediction:
- Sea level rise of 80 feet to 250 feet at a rate of 6 inches a decade (or more).
A 250 foot rise in sea level would be very, very bad. It conjures up the image of massive tidal waves consuming major coastal cities. But this forecast is problematic for at least a couple of reasons.
First, I don’t understand its internal logic. At 6 inches per decade, it would take 1,600 years to get to 80 feet, and 5,000 years to get to 250 feet. I assume we are not trying to consider effects in the years 3600 AD to 7000 AD.
Second, while Mr. Romm doesn’t provide attribution for this forecast, it isn’t anything like any formal prediction I’ve ever seen from the IPCC. The IPCC is appropriately reserved about making predictions for sea level rise. They don’t provide best estimates by scenario, but just broad ranges. They also caution that the high end of any range is not an absolute maximum. The spread of IPCC projections for sea level rise through about 2100 across all six marker scenarios ranges from a low of 0.18 meters to a high of 0.59 meters. (See WG1 SPM, Table SPM.3) Crudely, these average about 0.4 meters, or about 15 inches total by about 2100. That’s quite a ways from 80 to 250 feet.
Consider Mr. Romm’s second prediction:
Mr. Romm links to a blog post about a news report about a single climate simulation model study to support this. (Read the comments section on the page to which Mr. Romm links). The IPCC certainly does make predictions that large amounts of land, especially in Latin America (See WG2 Chapter 13, pages 583 and 597) and some other equatorial areas, will be subject to “desertification processes,” and this would be a serious problem, but I have not seen a claim as dramatic as Mr. Romm’s proceed from the IPCC process. Consider this in the context of Mr. Romm’s closely related prediction, also based on a blog citation of another single study using climate simulation models, for aridity impacts for the U.S.:
…”a permanent drought by 2050 throughout the Southwest” – levels of aridity comparable to the 1930s Dust Bowl would stretch from Kansas to California. And they were only looking at a 720 ppm case!
This is a pretty bleak picture of what, for example, the U.S. will look like: a 21st century army of Joads escaping an interior apocalypse, only to arrive at the coasts to find the cities flooded under hundreds of feet of water.
In contrast, here is the complete list of “examples of projected impacts” that the IPCC provides to illustrate the effects of climate change for the continent of North America (See Synthesis Report, Summary for Policymakers, page 11):
- Warming in western mountains is projected to cause decreased snowpack, more winter flooding and reduced summer flows, exacerbating competition for over-allocated water resources.
- In the early decades of the century, moderate climate change is projected to increase aggregate yields of rain-fed agriculture by 5 to 20%, but with important variability among regions. Major challenges are projected for crops that are near the warm end of their suitable range or which depend on highly utilized water resources.
- Cities that currently experience heat waves are expected to be further challenged by an increased number, intensity and duration of heat waves during the course of the century, with potential for adverse health impacts.
- Coastal communities and habitats will be increasingly stressed by climate change impacts interacting with development and pollution.
While serious, this doesn’t sound quite so dramatic.
And so on with various other of Mr. Romm’s predictions. The negative consequences of a several degree increase in global temperatures would almost certainly be quite significant. In addition, these consequences would likely be felt disproportionately by poor people in developing countries, especially those near the equator. Exaggerating these effects, however, is not helpful. Remember that when you add up all the specific impacts of the type that we have been reviewing here, the total economic costs at 40C of warming — which we would expect to reach sometime into the 22nd century — are estimated by the IPCC to be about 1 – 5% of global GDP (See WG2 SPM, page 17)
More centrally, we should not forget that there is a trade-off involved if we have to give up economic growth to avoid some of these problems. This can sound pretty dry, but consider the untold stories on the other side of the equation that would result from many millions of people being less able to escape poverty, or at a minimum, escaping poverty more slowly. What about all of the additional children who will die from dysentery between 2030 and 2070 because their communities couldn’t afford to put in improved sanitation and drainage systems that would have been installed had economic growth not been reduced as a result of carbon rationing? What about all of the elderly patients who had earlier and more painful deaths because of the new hospital wing that didn’t get built in 2020? Such a list could go on almost indefinitely. Indur Goklany’s response to my essay provides an excellent review of many of these consequences. Certainly, it appears that the major developing countries of the world are voting with their actions about where they stand on this trade-off.
We shouldn’t let the almost cinematic rhetoric of climate change disaster stories lead us to abandon rational consideration of costs and benefits.
3. Martin Weitzman has shown that traditional cost-benefit analysis is not appropriate for the case of global warming
Or is it rational to abandon cost-benefit analysis in this case? Mr. Romm says so. He argues that this has been “shown” (note, not “argued” or “asserted”, but “shown”):
The traditional economic modeling work by Yale’s William Nordhaus cited by Manzi is quite irrelevant in the face of 1,000 ppm warming. In fact, even a 3% chance of a warming this great is enough to render useless all traditional cost-benefit analyses that argue for delay or only modest action, as Harvard economist Martin Weitzman has shown.
Professor Weitzman’s reasoning on this topic is subtle and technically ingenious. In my opinion, it is the most rigorous existing argument for a carbon tax. Addressing it in detail is beyond the scope of these comments, but I have previously responded to a slightly earlier version of it in a long online article. Alternatively, you can watch a video of Professor Weitzman presenting his paper, and then my response to it in the exact same room a few months later. I encourage anybody who is serious about the climate change debate to understand Weitzman’s logic in detail.
In very short form (recognizing that I will write somewhat loosely for purposes of brevity in this setting), Weitzman’s central claim is that the probability distribution of potential losses from global warming is “fat-tailed”, or includes high enough odds of very large amounts of warming (200C or more) to justify taking expensive action now to avoid these low probability / high severity risks.
The big problem with this argument, of course, is that the IPCC has already developed probability distributions for potential warming that include no measurable probability for warming anywhere near this level for any marker scenario. See, for example, WG1 SPM Figure SPM.6. Even the scale on these charts — never mind actual predictions for the high end of the probability distributions — doesn’t go past 80C. That is, the best available estimates for these probability distributions are not fat-tailed in the sense that Weitzman means it.
Now, one can responsibly question the probability distributions developed by the IPCC. A modest version of this is simply to recognize that we are not certain that they are correct. But this is just a sophisticated restatement of one predicate of the Precautionary Principle. That is, as I put it in my essay, we must logically accept the possibility that “even the outer edge of the probability distribution of our predictions for global-warming impacts is enormously conservative.”
Of course, this is true in principle for all probability distributions, and therefore for all risks. As Weitzman himself clearly recognizes, in order for him to distinguish climate change dangers from other dangers in this matter, he must therefore show not only that it is possible that the true probability distribution of potential levels of warming is actually much worse than believed by the IPCC, but that a reasonable observer should accept it as likely that this is the case. In order to do this, he is forced to do his own armchair climate science, and argue (as he does explicitly in the paper) that he has developed a superior probability distribution for expected levels of warming than the ones the world climate-modeling community has developed and published. As noted above, this probability distribution is radically more aggressive than anything you will find in any IPCC Assessment Report. I don’t think that it is credible to accept Professor Weitzman’s climate science in place of the IPCC’s.
4. The costs of aggressive emissions abatement are very small
Mr. Romm says of atmospheric concentration of carbon dioxide that:
…our choice is really to stay below 450 ppm or risk self-destruction.
He goes on to say:
The good news is that a host of independent analyses, including from the IPCC itself, make clear that the cost of keeping carbon dioxide concentrations at or below 450 ppm is very low indeed.
He supports this by quoting the IPCC as saying:
In 2050, global average macro-economic costs for mitigation towards stabilisation between 710 and 445 ppm CO2-eq… corresponds to slowing average annual global GDP growth by less than 0.12 percentage points. [Bold in original]
This is somewhat problematic as support for Mr. Romm’s assertions, for at least a couple of reasons.
First, as quoted, this isn’t the cost to “stay below 450 ppm,” but is the cost to end up somewhere between 710 and 445 ppm. A crude average is 578 ppm. The current atmospheric concentration of carbon dioxide is estimated to be about 385 ppm. According to Mr. Romm, the world must therefore limit emissions sufficiently to prevent an increase of more than 65 ppm in this concentration index, or else “risk self-destruction.” But this cost level would allow concentration to rise 193 ppm, or about three times this amount. Presumably, it would cost significantly more to remain at or below the 450 ppm target that Mr. Romm has put forward.
Second, the impact of differences in growth rates can be very counter-intuitive when compounded over many years. Consider a grossly simplified illustration of a society with constant population that has an average income of $10,000 per person today. If their economy grows at 2% per year for a century, average income will rise to a little over $72,000 after 100 years. Now suppose that they predicted that a by-product of this growth would be to create climate damage that by the end of a century would cause them to lose 3% of GDP by the 100th year. In this case, projected average income in year 100 would be reduced to about $70,000. Now, suppose some smart technologists offered them the option to eliminate all climate damage in return for slowing GDP growth by only 0.12 percentage points every year over the century. Should they take this deal? If they did, GDP would grow by 1.88% each year, and average income in year 100 would be a little over $64,000. Given these two alternatives, they would be much better off in economic terms by growing the economy at 2% and just taking the pain of the climate damage.
This example brings us all the way back to the start of my essay, and the central problem for advocates of aggressive emissions abatement advocates: despite the rhetoric, the projected damages from global warming just don’t appear to justify the costs of the proposed remedy.