Climate Damage Functions for Productivity: Evidence from Enterprise Survey Microdata
Using firm-level microdata from 125 countries and a long-difference framework to address spatial correlation concerns, this study reveals a robust, convex temperature damage function for productivity that imposes increasingly severe growth losses in hotter baseline climates, with significant regional heterogeneity and smaller projected losses when accounting for precipitation and climate extremes.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Climate change is often discussed in terms of rising sea levels or melting ice caps, but its most immediate economic impact is felt in the quiet hum of factories, the output of farms, and the daily operations of businesses. Economists have long tried to measure how much heat and weather extremes slow down economic growth, but their tools have often been blunt. They typically look at entire countries, comparing national income to average temperatures. This approach has a significant flaw: it treats a nation as a single, uniform unit, ignoring the fact that a business in a hot desert city faces a very different reality than one in a cool coastal town. Furthermore, short-term weather shocks, like a single hot summer, can be confused with long-term climate trends, making it difficult to see the true, lasting damage heat inflicts on productivity. To understand the real cost of a warming world, researchers need to look closer, at the individual enterprise, and track changes over longer periods to separate fleeting weather events from permanent shifts in the climate.
A new study by Matthew Ndubuisi Uba from Ca' Foscari University of Venice takes this closer look. By analyzing data from over 200,000 businesses across 125 countries, the research maps out how rising temperatures affect the ability of firms to produce value. The study uses a method that compares the average performance of businesses in a specific region at the beginning of a survey period against their performance at the end, rather than looking at year-to-year fluctuations. This "long-difference" approach allows the researcher to isolate the persistent effects of climate change, filtering out the noise of temporary weather variations. The data comes from the World Bank Enterprise Surveys, which ask companies about their sales and costs, combined with high-resolution climate records that track temperature, rainfall, floods, and droughts.
The results reveal a clear and concerning pattern: the damage heat does to business productivity is not the same everywhere, and it gets worse the hotter the starting conditions are. When the researcher looked at the global average, the signal was weak and hard to pin down. However, once the data was broken down by region, a distinct story emerged. In cooler regions, such as parts of North and Central America and South Asia, a slight warming appeared to have a neutral or even slightly positive effect on productivity. But in regions that are already hot, like Sub-Saharan Africa, Latin America, and parts of Europe and Central Asia, rising temperatures caused a sharp decline in business growth. The relationship is not a straight line; it is curved. This means that for every additional degree of warming, the loss in productivity becomes increasingly severe in already hot climates. A business in a hot region suffering from further warming loses more ground than a business in a cool region suffering the same amount of warming.
The study also examined other climate factors, such as heavy rainfall, floods, and long periods without rain. While these extreme events matter, they did not change the fundamental finding that temperature is the dominant driver of productivity loss. In fact, when the model included these other weather variables, the projected losses from warming actually appeared slightly smaller. This suggests that the damage caused by heat might be partly mediated through other channels, such as how heat interacts with water availability, rather than heat acting as a simple, isolated force. The research also tested whether these findings were driven by just one specific part of the world. By removing one region at a time from the analysis, the researcher confirmed that the pattern holds true across the globe and is not an artifact of a single influential dataset.
The implications of these findings are significant for how we understand economic risk. The idea that climate damage can be summarized by a single global number is misleading. Instead, the risk is highly uneven. A business operating in a hot baseline climate faces a steep, accelerating curve of productivity loss as the planet warms, while a business in a cooler climate may face a much flatter curve. This suggests that financial stress tests and policy planning need to account for local climate conditions rather than relying on broad averages. The study does not claim to solve the problem of climate adaptation, but it provides a clearer map of where the economic pain will be felt most acutely. It confirms that as the world warms, the gap between the economic performance of firms in hot regions and those in cooler regions is likely to widen, driven by a nonlinear relationship where heat becomes increasingly destructive the hotter it gets.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.