“Gremlins” caused errors in climate change paper showing gains from global warming
The author of a controversial 2009 paper arguing that at least some amount of global warming could lead to economic gains has corrected the paper, along with a later article in a different journal. We confess to be baffled by the implications of the mix-up, although others appear to be less confused.
The 2009 article, “The Economic Effects of Climate Change,” was written by Richard Tol, of the University of Sussex, and appeared in the Journal of Economic Perspectives. It has been cited 91 times, according to Thomson Scientific’s Web of Science.
Here’s the abstract:
I review the literature on the economic impacts of climate change, an externality that is unprecedentedly large, complex, and uncertain. Only 14 estimates of the total damage cost of climate change have been published, a research effort that is in sharp contrast to the urgency of the public debate and the proposed expenditure on greenhouse gas emission reduction. These estimates show that climate change initially improves economic welfare. However, these benefits are sunk. Impacts would be predominantly negative later in the century. Global average impacts would be comparable to the welfare loss of a few percent of income, but substantially higher in poor countries. Still, the impact of climate change over a century is comparable to economic growth over a few years. There are over 200 estimates of the marginal damage cost of carbon dioxide emissions. The uncertainty about the social cost of carbon is large and right-skewed. For a standard discount rate, the expected value is $50/tC, which is much lower than the price of carbon in the European Union but much higher than the price of carbon elsewhere. Current estimates of the damage costs of climate change are incomplete, with positive and negative biases. Most important among the missing impacts are the indirect effects of climate change on economic development; large-scale biodiversity loss; low-probability, high-impact scenarios; the impact of climate change on violent conflict; and the impacts of climate change beyond 2100. From a welfare perspective, the impact of climate change is problematic because population is endogenous, and because policy analyses should separate impatience, risk aversion, and inequity aversion between and within countries.
But according to the correction notice:
Gremlins intervened in the preparation of my paper “The Economic Effects of Climate Change” published in the Spring 2009 issue of this journal. In Table 1 of that paper, titled “Estimates of the Welfare Impact of Climate Change,” minus signs were dropped from the two impact estimates, one by Plambeck and Hope (1996) and one by Hope (2006). In Figure 1 of that paper, titled “Fourteen Estimates of the Global Economic Impact of Climate Change,” and in the various analyses that support that figure, the minus sign was dropped from only one of the two estimates. The corresponding Table 1 and Figure 1 presented here correct these errors. Figure 2 titled,”Twenty-One Estimates of the Global Economic Impact of Climate Change” adds two overlooked estimates from before the time of the original 2009 paper and five more recent ones.
The second article, “Targets for global climate policy: An overview,” appeared in the May 2013 issue of the Journal of Economic Developments and Control. It attempted to correct the 2009 paper, and its abstract states:
A survey of the economic impact of climate change and the marginal damage costs shows that carbon dioxide emissions are a negative externality. The estimated Pigou tax and its growth rate are too low to justify the climate policy targets set by political leaders. A lower discount rate or greater concern for the global distribution of income would justify more stringent climate policy, but would imply an overhaul of other public policies. Catastrophic risk justifies more stringent climate policy, but only to a limited extent.
Its correction reads:
The author regrets that three errors crept into “Targets for global climate policy: An overview”. In Table 1 of that paper, the author dropped the minus sign on Chris Hope׳s 2006 estimate; his best guess is that a global warming by 2.5 °C would cause a welfare loss equivalent to losing 0.9% of income, with a confidence interval spanning −2.7% to 0.2%. This error also affects Fig. 1. Fig. 1 had a coding error in the upper bound of the confidence interval, and an error in the caption. Corrected Fig. 1 is given below.
17 Estimates of the global economic impact of climate change, expressed as the welfare-equivalent income loss, as a function of the increase in global mean temperature relative to pre-industrial times. The blue dots represent the estimates. The central line is the least squares fit to the 17 observations: D=3.99(1.39)T−1.82(0.52)T2, R2=0.67, where D denotes impact and T denotes temperature. The outer lines are the boundaries of the 95% confidence interval, where the standard deviation is the least squares fit to the five reported standard deviations or half confidence intervals: Soptimistic=0.43(0.16)T, R2=0.65 and Spessimistic=1.82(0.86)T, R2=0.53, where S is the standard deviation. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
The blog of the London School of Economics has an interesting piece on the implications of the mistake. It reads, in part:
A correction to a paper quietly published by an economics journal last week has put pressure on the Intergovernmental Panel on Climate Change (IPCC) to amend a section of its new report that claimed that moderate amounts of global warming have overall positive benefits. The correction by Richard Tol, a professor of economics at the University of Sussex, has updated a paper he published in the Journal of Economic Perspectives in 2009 which collected together the results of studies of the economic impacts of climate change.
The original paper suggested that there were net beneficial impacts up to about 2.2 centigrade degrees of warming, and has been widely cited by climate change ‘sceptics’, including Viscount Ridley who used it as the basis of an article in The Spectator in October 2013 which claimed that “climate change has done more good than harm so far and is likely to continue doing so for most of this century”.
However, after I raised concerns with Professor Tol and with the journal about a number of errors in the article, a correction was quietly posted on the journal’s website last week. The correction points out that the original paper concluded that “there were net benefits of climate change associated with warming below about 2°C”, but the updated analysis shows “impacts are always negative”.
This is significant because a more recent paper by Professor Tol, which was published by the Journal of Economic Dynamics and Control in 2013 but contained many of the same errors as the 2009 paper, was used as the basis for a section on the economic damage caused by climate change in the report of the IPCC on ‘Impacts, Adaptation and Vulnerability’.
The final version of the Summary for Policymakers of the IPCC report was published on 31 March, along with the underlying final drafts of the chapters. Section 10.9.2 of Chapter 10, on which Professor Tol is a Coordinating Lead Author, draws on his 2013 paper and states: “Climate change may be beneficial for moderate climate change but turn negative for greater warming”. The chapter also includes Figure 10-1 which plots 20 of the 21 data points included in the correction by Professor Tol to his 2009 paper.
I have now drawn the attention of the IPCC to Professor Tol’s correction and suggested that it needs to amend the text of Chapter 10 before the final version is published later this year. The amendment is also important to ensure that the error does not appear in the IPCC Synthesis Report, which is due to be published at the end of October. I have also asked Viscount Ridley, via Twitter, to amend his article in The Spectator now that Professor Tol has published a correction. Lord Ridley has so far refused.
Tol, for his part, tells us that the errors don’t undermine his conclusions:
Although the numbers have changed, the conclusions have not. The difference between the new and old results is not statistically significant. There is no qualitative change either.
Update, 12 p.m. Eastern, 5/21/14: Tol added more comments in response to the LSE post below.
Hat tip: Thomas Baguley