Mistreatment of the economic impacts of extreme events in the Stern Review Report on the Economics of Climate Change
Roger Pielke Jr. 2007. Mistreatment of the economic impacts of extreme events in the Stern Review Report on the Economics of Climate Change. Global Environmental Change 17 (2007) 302–310.
Roger Pielke Jr. is a Fellow of the Center for Science and Technology Policy Research, University of Colorado.
Full text [here]
Selected excerpts:
Abstract
The Stern Review on the Economics of Climate Change has focused debate on the costs and benefits of alternative courses of action on climate change. This refocusing has helped to move debate away from science of the climate system and on to issues of policy. However, a careful examination of the Stern Review’s treatment of the economics of extreme events in developed countries, such as floods and tropical cyclones, shows that the report is selective in its presentation of relevant impact studies and repeats a common error in impacts studies by confusing sensitivity analyses with projections of future impacts. The Stern Review’s treatment of extreme events is misleading because it overestimates the future costs of extreme weather events in developed countries by an order of magnitude. Because the Stern Report extends these findings globally, the overestimate propagates through the report’s estimate of future global losses. When extreme events are viewed more comprehensively the resulting perspective can be used to expand the scope of choice available to decision makers seeking to grapple with future disasters in the context of climate change. In particular, a more comprehensive analysis underscores the importance of adaptation in any comprehensive portfolio of responses to climate change.
Introduction: exploiting an excess of objectivity
In a provocative article titled “How Science Makes Environmental Controversies Worse” Daniel Sarewitz explains that scientific research results in an “excess of objectivity” in political debates (Sarewitz, 2004). What he means with this phrase is that in most (if not all) cases of political conflict involving science, available research is sufficiently diverse so as to provide a robust resource for political advocates to start with a conclusion and then selectively pick and choose among existing scientific studies to buttress their case. Simply put, to cherry pick, to take the best leave the rest. …
The main peril is that an advocate for a particular agenda will first decide upon a course of action and then seek science useful in justifying that course of action. Of course, the advocate’s political opponent will also settle on a (different) particular agenda and seek out their own justifying science. What then typically happens is that the political debate is transferred to the science used as justifications, rather than taking place explicitly in terms of the values or outcomes at stake that motivated the political controversy in the first place. Scientific debate then becomes a proxy for political debate, and gridlock and inaction often result because science alone cannot resolve political disputes.
In the area of climate change, there have been countless efforts to provide scientific advice to decision makers. The Stern Review Report on the Economics of Climate Change is one such effort (Stern, 2007). The Stern Review has already achieved several notable successes. Among them, it has focused attention on the challenge of climate change and helped to redirect attention away from debates over science and toward debates over the costs and benefits of alternative courses of action.
However, in making its case for the significant future economic costs of extreme weather events in developed countries the Stern Review commits two significant errors that affect its estimates. In its Chapter 5 the Stern Review concludes, “The costs of climate change for developed countries could reach several percent of GDP as higher temperatures lead to a sharp increase in extreme weather events and large-scale changes.” (Stern, 2007, p. 137). This conclusion cannot be supported by the Review’s own analysis and references to literature. One error is a serious misrepresentation of the scientific literature, and the second is more subtle, but no less significant. The serious misrepresentation takes the form of inaccurately presenting the conclusions of an unpublished paper on trends in disaster losses. The second error is more complex and involves conflating an analysis of the sensitivity of society to future changes in extreme events, assuming that society does not change, with a projection of how extreme event impacts will increase in the future under the integrated conditions of climatic and societal change. The result of the errors in the Stern Review is a significant overstatement of the future costs of extreme climate events not simply in the developed world, but globally — by an order of magnitude. …
To summarize, to justify its conclusion of large increases in future economic losses of extreme events due to climate change, the Stern Review misrepresents a single non-peer-reviewed, heavily caveated background paper to a workshop which itself resulted in conclusions counter to those presented by Stern. The Stern Review neglected to focus either on a consensus view among relevant experts (including the lead author of the study that it did cite) which asserts that it is presently not possible to quantify the role of greenhouse gas emissions in trends in disaster losses. Thus, the Stern Review provides no scientific basis for its projections for future increases in losses in developed countries (and the world) related to extreme events under conditions of future climate change. …
Conclusion: science advisors: issue advocate or honest broker?
This brief critique of a small part of the Stern Review finds that the report has dramatically misrepresented literature and understandings on the relationship of projected climate changes and future losses from extreme events in developed countries, and indeed globally. In one case this appears to be the result of the misrepresentation of a single study. This cherry picking damages the credibility of the Stern Review because it not only ignores other relevant literature with different conclusions, but it misrepresents the very study that it has used to buttress its conclusions. In a second case, Stern repeats and reinforces a common methodological mistake by presenting a sensitivity analysis as if it were a projection or prediction.
When one takes a closer look at the sensitivity analysis one finds that it in fact shows that future societal changes, and not climate changes, are the primary drivers of disaster losses and will be for the foreseeable future. The net result of these errors is that the Stern Review overestimates future losses from extreme events by one order of magnitude. …
