New research shows that the economic impacts of climate change, especially extreme temperatures, are much more widespread and costly than previously thought. The study, which uses data from Italian companies and detailed climate records, reveals significant negative impact of extreme temperatures on business productivity and efficiency of input allocation, leading to much higher overall productivity losses. These findings challenge traditional economic models that tend to underestimate the long-term costs of warming.

The study identifies two key channels through which temperature affects business productivity:

• Direct effects: These include reduced worker efficiency, poorer performance of machinery and equipment under extreme conditions, or changes in consumer demand. For example, machines like mobile phones or data centers may operate less efficiently in high temperatures.
• Indirect effects: They arise when firms face barriers to adapting inputs, such as capital or labor, to climate change. For example, if firms cannot quickly reduce capital in response to lower productivity, inefficiencies arise. These frictions in adjusting inputs, especially capital, significantly amplify the productivity costs caused by climate change.Capital is particularly susceptible to these frictions, while materials and labor exhibit greater flexibility. These indirect effects can double the estimated economic damage and account for up to half of the total aggregate productivity losses.

The research uses a micro-macro approach that integrates firm-level insights into macroeconomic models. Based on Italian data from 1999–2013, they found that there is inverted U-shaped relationship between temperature and firm productivity. This means that moderate temperatures have little impact, but extreme heat, as well as extreme cold, significantly reduce firms’ sales and input efficiency. For example, one additional day with temperatures above 40 degrees Celsius reduces sales by 0.807 %, while one day below 0 degrees Celsius reduces them by 0.094 %. While productivity is significantly affected, demand (measured by comparing tradable and non-tradable sectors) plays only a smaller, but non-zero, role.

Quantification of aggregate productivity losses

Aggregate productivity losses from warming may be significantly higher than previously thought. In the scenario an increase in the average annual temperature by 2°C reduces aggregate productivity by 1.68 %, with direct effects contributing to a decrease of 0.81 % and indirect effects to a decrease of 0.87 %. This represents an economic loss of approximately $35.37 billion of Italy's GDP in 2021. At an increase of 4°C, the drop in productivity rises to 6.81 %, which is more than four times the decrease at 2°C. These losses are about four times higher than those estimated by representative firm-based models that do not take into account frictions that distort allocative efficiency. The nonlinear relationship between temperature and aggregate productivity is due to a combination of the nonlinear effect of temperature on productivity at the firm level and the wider geographical spread of temperature extremes in the 4°C scenario.

Implications for policy and adaptation

The results suggest that existing estimates of climate damage are likely too low because they do not take into account input rigidity and their misallocation. Policies that improve firms' ability to adjust inputs – such as investments in climate-resilient infrastructure – could mitigate productivity losses. Taking into account adaptation by firms (e.g. historically adapted areas) shows that climate technologies can reduce damages by 20–30 %.

Regional heterogeneity

Study also reveals significant regional differences in productivity losses across Italian provinces. While some northern Alpine areas may see modest gains due to fewer days below 0°C, southern regions face significant negative impacts due to an increase in extremely hot days. This highlights that Climate change is likely to worsen existing regional inequalities in Italy.

In conclusion, this research highlights that a comprehensive understanding of the economic costs of climate change requires a micro-funded approach that considers not only direct impacts on firm productivity, but also the key role of frictions in input adjustment and subsequent allocative inefficiencies. Integrating these effects into future integrated assessment models (IAMs) is crucial for more accurate estimates of the economic costs of climate change and the societal costs of carbon. JRi