Climate change study predicting dire economic harm is retracted over Uzbekistan numbers
Prestigious journal Nature claimed the world’s economic output would decline 62 per cent by 2100 – a more severe estimate than other forecasts – but that information now appears to be skewed
A widely cited study on economic damage from climate change has been retracted following criticism from peers.
The research, published last year in the prestigious journal Nature, projected that the world’s economic output would decline 62 per cent by 2100 under a high-carbon emissions scenario. The estimate was much more severe than other forecasts, prompting scrutiny of the underlying data.
“We broadly agree with the issues raised, and have made corrections to the underlying economic data and to our methodology to address them,” said study author Leonie Wenz, from the Potsdam Institute for Climate Impact Research in Germany.
“These changes are too substantial for a correction of the original article in Nature.”
The study examined historical data from 1600 regions worldwide over the past four decades to project how changes in temperature and precipitation would affect economic growth, including factors like agricultural yields, labour productivity and infrastructure.
However, after the study was published, other researchers found that economic data from one country – Uzbekistan – during a short time from 1995 to 1999 had skewed the results. Without Uzbekistan, the 2100 damage forecast fell to 23 per cent, not 62 per cent. The researchers published their critique in Nature in August.
Another researcher who wasn’t involved in the original work, Christof Schotz, said the results were more uncertain than the study suggested and published a separate critique in Nature in August.
The study had been cited by the US Congressional Budget Office, the World Bank and the Network for Greening the Financial System, or NGFS, a coalition of central banks from which the US Federal Reserve withdrew this year.
The coalition, which has members in about 90 countries, incorporated the damage forecast in risk tools that banks can use, for instance, to stress-test portfolios to meet European regulations. The NGFS damage estimate still incorporates the study – with a disclaimer flagging the retraction – pending its next update.
Central bank models of the economic impacts of climate change could have far-reaching implications. If these models show impacts are going to be much worse than previously thought, regulators could make banks set aside more capital to buffer against potential losses associated with assets exposed to climate change, according to Ely Sandler, a researcher at the Harvard Kennedy School.
Scientists noted that despite problems with the data in this study, there is still scientific agreement that climate change hurts the world economy.
“There is a broad scientific consensus regarding the severe negative economic effects of climate change,” said Mr Schotz, who added the retraction “does not alter that reality”.
Ms Wenz noted that, despite the retraction, the conclusions of her group’s study were consistent with other research showing climate change has substantial effects on macroeconomic productivity, and that those impacts outweigh the costs of climate change mitigation.
The group plans to submit a revised version of the study for peer review. Nature said it followed the journal’s process for handling concerns about a published paper, which involves consulting the original authors of the paper and seeking advice from independent, external peer reviewers.
“Our priority is always to uphold the integrity of the scientific record, ” said Karl Ziemelis, chief applied and physical sciences editor for Nature.
The Wall Street Journal
