Global warming already causing crop losses of over $20 billion a year
Get the Energy newsletter
Daily energy & climate — solar, EVs, oil, the policy fights and tech bets shaping the transition. Free.
- IIASA researchers Yi Ling Hwong and Kai Kornhuber calculated that warming-driven heat extremes and drought have already cut maize, wheat, and soybean yields by 3.5 percent between 2007 and 2019 versus the 1974–2004 baseline, costing more than $20 billion a year at then-prevailing producer prices.
- Under a high-emissions scenario (SSP3-7.0), the team projects global yields could fall around 35 percent by 2100, pushing annual losses to over $161 billion and 855 million tonnes of lost production — roughly what 2 billion people consume in a year, Hwong said at the European Geosciences Union meeting in Vienna.
- Low-income African countries will suffer disproportionately because most of the population works in agriculture, a disparity Hwong warned could trigger social unrest and increased migration even as the largest absolute financial losses land on major producers such as the US.
- Columbia University's Jonas Jägermeyr countered that statistical yield models are "inherently unreliable" when extrapolated to vastly different climate regimes by 2100, and recommended process-based crop models for long-range projections.
- University of Queensland's Karine Chenu added that two widely used wheat models produce "large errors" and are especially poor at forecasting the joint effects of extreme heat and drought, citing a recent unpublished study from her team.
- The IIASA team acknowledged its estimate is likely a floor: it covers only three crops, excludes flood, storm, and rain damage, and omits price spikes already roiling other commodities such as coffee and cacao.
Why it matters: With 3.5 percent of global maize, wheat, and soybean output already lost to warming-driven extremes, the study puts a dollar figure ($20 billion today, potentially $161 billion by 2100 under SSP3-7.0) on a risk that falls hardest on farming-dependent low-income economies — and the methodology debate means even that headline may understate what's coming if process-based models prove more accurate for end-of-century conditions.
