Last year we reported on Nature’s retraction of a prominent study claiming global warming will trash the global economy, after experts finally looked at it closely and discovered that the results were distorted by phony data from, of all places, Uzbekistan, where we ourselves had been not long before, reporting on the then-latest COP boondoggle. When the data were corrected the results all fell apart, but not after mandarins like Mark Carney had embedded the fake results in the rules of the global banking system. Now h/t Roger Pielke Jr. we learn of a new paper in Science Advances in which two statisticians look carefully at a different, but equally prominent, pair of papers published in recent years claiming that greenhouse gases will cause more heatwaves, and those heatwaves will cause $28 trillion dollars in global economic damages, once again trashing the global economy. Not so fast, said statisticians Abraham Wyner and Ryan Brill of the University of Pennsylvania. In this case it wasn’t junk data from Uzbekistan. It was junk data from Mongolia and Kenya. Oh, and bad methodology. Both the Wyner and Brill paper and Pielke Jr.’s own reanalysis show that instead of a $28 trillion dollar loss, the corrected numbers show the effect of heat on the economy to be close to zero, and possibly even positive. Blast.

The newer papers were published by Christopher Callahan and Justin Mankin. And Wyner and Brill go through a lot of technical reanalysis of their methodology and argue that Callahan and Mankin overstated the certainty of their findings by ignoring, or taking inadequate notice of, the fact that while they had thousands of data points, they were all highly correlated with each other so their sample didn’t have as much detail as they claimed. Applying corrections for that caused the uncertainty estimates to expand widely. They also noted that the economic measure used, annual income growth rates, are correlated over time. Awkwardly, taking these points into account, along with removing a few dodgy observations from Kenya and Mongolia, caused the main conclusions to reverse: heat is no longer a threat to growth, instead if anything it will increase it.

Pielke Jr., inspired by the Wyner and Brill paper, did a further analysis of his own and found a further litany of errors in the original studies. So much for peer review, you may well cry. And rightly, because the results change dramatically by switching from Celsius to Fahrenheit, which makes no sense. Also, the economic effects are measured in dollars and don’t distinguish between an actual change in output and a mere currency fluctuation. Furthermore Honduras changed the way it measured economic activity in 2001 causing a one-time spurious jump in output, and removing that one distorting effect alone cut the global effects by a third. Plus they mangled the way they handled missing years. Pielke Jr. argues:

“A reproduction of their methodology with all four errors corrected results in totals from a $5 trillion gain to a $2 trillion loss, scattered around zero and nowhere near the published $28 trillion loss.”

Then, referring to the retraction we mentioned above, Pielke Jr. points out that the Callahan and Mankin paper has at least as many fatal problems and likewise ought to be retracted. Maybe in future we could just save time and whenever a paper comes out predicting economic doom due to tiny temperature changes we just assume it’s wrong and will eventually be withdrawn.