Center Home Science Policy Photos University of Colorado spacer
CIRES CU
Location: > Prometheus: Of Blinders and Innumeracy Archives

September 13, 2005

Of Blinders and Innumeracy


Posted to Author: Pielke Jr., R. | Climate Change

Elizabeth Kolbert has an article in the New Yorker on everyone’s favorite topic these days, hurricanes and global warming. The article is amazing because even though the data is staring Kolbert right in the face, she apparently cannot bring herself to grasp its implications for her argument.

Kolbert describes the problem: “In June, the Association of British Insurers issued a report forecasting that, owing to climate change, losses from hurricanes in the U.S., typhoons in Japan, and windstorms in Europe were likely to increase by more than sixty per cent in the coming decades. (The report calculated that insured losses from extreme storms—those expected to occur only once every hundred to two hundred and fifty years—could rise to as much as a hundred and fifty billion dollars.) The figures did not take into account the expected increase in the number and wealth of people living in storm-prone areas; correcting for such increases, the losses are likely to be several hundred per cent higher.”

Let’s first do some fact checking and take a look at what the Association of British Insurers report (PDF) actually says. Table 6.4 on p. 25 indicates that the ABI analysis shows a current total loss baseline of $16.5 billion for U.S. hurricanes, Japanese typhoons and European windstorms. And on p. 23 it notes that “If carbon dioxide concentrations doubled, total average annual damages from US hurricanes, Japanese typhoons and European windstorms combined could increase by up to $10.5 bn (¥1140 bn, €8.5 bn) from a baseline of about $16.5 bn today, representing an increase of around 65%.” So this is where she gets the more than sixty percent figure. So far so good, although the projected date for these changes in into the 2080s, a little bit further on than “coming decades.” But let’s move on.

But where does the number come from for the effects of societal changes, which she describes as likely to be “several hundred percent higher”? One has to go to the supplementary information to the report (PDF) to find this information. At p. 41 tables 3.8 and 3.9 one finds the data on projected wealth and population growth. Based on the numbers here, the combined effects of population and wealth on storm damages, independent of climate changes, would result in 2085 in total damages of $81.8 billion for U.S. hurricanes, Japanese typhoons and European windstorms, or an increase of $65.3 billion over the 2004 baseline. This is an increase of more than 600%, quite a bit larger than the “several hundred percent” reported by Kolbert.

Now, I do have some questions about the ABI study, which would be worth revisiting in the future, such as why it projects a combined annual rate of increase in wealth and population of only 2% for these regions. (The ABI seems unfazed by the apparent contradictory information it presents: “If Hurricane Andrew had hit Florida in 2002 rather than 1992, the losses would have been double, due to increased coastal development and rising asset values.” This represents about an 8% annual increase.) Also, The ABI’s estimates for increasing damages related to wind storms is roughly an order of magnitude larger than projected by the IPCC in its SAR. Reports such as those produced by the ABI can be extremely valuable, but they’d be more readily understood and placed into context if their analyses were submitted to the peer-reviewed literature.

But let’s press ahead. Kolbert’s column understates what the report actually says about the effects of growing population and wealth by a factor of 3. The ABI finds that future losses to wind storms are about 10 times more sensitive to societal factors than to climate factors. Yet even using the wrong numbers Kolbert writes a paragraph that clearly acknowledges the dominant role of societal changes in shaping future windstorm damages. And yet, when it comes to discussing policy options she focuses exclusively on the less significant factor:

“As the rest of the world has adopted Kyoto—earlier this year, the treaty became binding on the hundred and forty nations that had ratified it—these arguments have become increasingly indefensible, and the President has fallen back on what one suspects was his real objection all along: complying with the agreement would be expensive. “The Kyoto treaty didn’t suit our needs,” Bush blurted out during a British-television interview a couple of months ago. As Katrina indicates, this argument, too, is empty. It’s not acting to curb greenhouse-gas emissions that’s likely to prove too costly; it’s doing nothing.”

I do think that there is more than just political opportunism at work here. I simply don’t think that people are prepared to see the numbers right before their eyes. Part of this dynamic must be the effects of starting with a solution (greenhouse gas reductions) and searching for justifications, but part also must be related to the lure of silver bullet policy solutions rather than the difficult, yet mundane challenges of addressing the day-to-day decisions that cumulatively result in a tremendous increase in vulnerability to disasters. But if we are to address the ever-growing vulnerability of society to weather disasters we cannot continue to ignore the obvious.

Posted on September 13, 2005 07:41 PM

Comments

You nail it in the last paragraph. Working backwards from a conclusion is always what gets you into trouble.

But it's not always the case. Sometimes people just lie, I'm afraid.

Posted by: Dylan Otto Krider at September 13, 2005 10:39 PM


Dr. Pielke,

I agree with your point that "when it comes to discussing policy options [Mrs. Kolbert] focuses exclusively on the less significant factor." IMO, that is the only point that I would consider worth making, since Kolbert is not an expert, she is not pretending to be an expert and she is not writing to an expert audience. With respect, the rest of your post strikes me as a little bit nit-picky.

E.g.: "This is an increase of more than 600%, quite a bit larger than the "several hundred percent" reported by Kolbert."

Six is quite a bit larger than several? The American Heritage dictionary says: "several: Being of a number more than two or three but not many."

You also write:
"Kolbert's column understates what the report actually says about the effects of growing population and wealth by a factor of 3."

Where does she do this?

Posted by: James Bradbury at September 14, 2005 02:21 PM


That last paragraph's reference to "silver bullet solutions" is a little odd since the policy changes needed to limit coastal development are way simpler and easier to implement than those needed to stop AGW. How many more Katrinas will it take to change that?

Posted by: Steve Bloom at September 14, 2005 09:49 PM


James- Thanks. I pretty much agree with you. While Kolbert may have the luxury of being imprecise (though I'd like to think columnists/journalists should be held to high standards of accuracy), I want my commentary to be as solid as possible. So view the fact checking as my own investigation of the data. To answer your question, Kolbert finds ("several hundred percent"/65% = ~3/1, and the real ratio from the ABI report is actually 10/1, 10/3 is where I got the "factor of 3"). I don't think this is very important in the broader context -- I think that the main points stand irrespective of these details.

Posted by: Roger Pielke Jr. at September 14, 2005 09:57 PM


Roger, I think you've missed the point of what Kolbert was saying. She wasn't saying that the effects of societal change would be several hundred percent higher. She said "The figures did not take into account the expected increase in the number and wealth of people living in storm-prone areas; correcting for such increases, the losses are likely to be several hundred per cent higher."

i.e. because more people will be living in vulnerable areas, and because property there will be more valuable, the effects of increased destructiveness in the future will be greater than the effects of the same amount of increased destructiveness applied to the current demographic situation.

So you can imagine a situation: AGW and no societal changes = 65% greater cost. No AGW but societal change = 600% greater cost. AGW and societal change = 1.65 x 6 = 1000% greater cost. OK these are pretty naive calculations but you can see that the effect of AGW is indeed to increase the cost by several hundred percent.

Posted by: Tom Rees at September 15, 2005 04:40 AM


Tom, thanks, I see what you are saying. Though if this is indeed what Kolbert was saying then she is even further off than I thought, as the ABI report she draws this information from does not say this, and when one talks of something being X percent larger this is usually a multiplier not added on. But even so, the larger point on her discussion of policy options remains. Thanks.

Posted by: Roger Pielke Jr. at September 15, 2005 05:19 AM


Roger, the percentages are additional but the effect relative to baseline is multiplicative. e.g. a storm that costs $1 billion today will cost $6 billion in the future due to societal changes, then you add 65% for the effects of AGW. True there's nothing in the report about this, but I guess it's where she was coming from.

You're right that this has no impact on policy regarding the relative costs of AGW vs societal changes (if you eliminated societal changes, then the effect would be a whopping reduction in the cost of my hypothetical storm from $10 billion to $1.65 billion - even without addressing the AGW issue).

But it does have major implications for the cost-effectiveness of any measure (including AGW emissions reductions) to reduce risk.

Posted by: Tom Rees at September 15, 2005 09:31 AM


> ... I simply don’t think that people are prepared
> to see the numbers right before their eyes. ...
> But if we are to address the ever-growing
> vulnerability of society to weather disasters
> we cannot continue to ignore the obvious.

Do you mean things like the exponential implications of "steady growth"? It's a global phenomenon, not just weather-related.

Posted by: Steve Sturgill at September 16, 2005 12:56 PM




Sitemap | Contact | Find us | Email webmaster