The Uzbekistan-shaped hole in an influential climate cost paper

The retraction of a highly influential 2024 climate economics paper, originally published in *Nature*, reveals how data anomalies and flawed statistical modeling can distort global policy projections. Professor Solomon Hsiang from Stanford University identified that the study's alarming prediction of a 60% reduction in global GDP by 2100 was disproportionately driven by erroneous regional data from Uzbekistan, where reported growth figures reached an implausible 100% per year. Beyond the "Uzbekistan problem," the analysis failed to account for spatial correlations between interconnected regions, leading to understated uncertainty and statistically insignificant results once corrected. While the original authors work to revise their findings, current robust estimates suggest a more likely, though still significant, 20% decline in global GDP due to climate change, albeit with a wide uncertainty range spanning from 0% to 40%.