By Hélia Costa, Guido Franco, Filiz Unsal, Sarath Mudigonda, Maria Paula Caldas.
The pace of temperature increase has been steadily accelerating over the past decades (IPCC, 2021). The increasing frequency and intensity of heat stress episodes due to climate change poses significant threats to the global economy, including through its effects on productivity. Despite recent more stringent mitigation efforts and the expectation that climate targets will be met, temperatures are still projected to rise, raising concerns about the economic costs associated with climate change.
One key channel through which temperature affects economic outcomes is labour productivity. As temperatures increase, both the cognitive and physical capacity of workers decrease, and extreme temperatures can also increase absenteeism due to heightened health issues and transport disturbances. Beyond its direct effect, heat stress can further impact productivity through disruptions to infrastructure (such as energy), increased production costs, or disruptions to supply.
Against this backdrop, our new paper (Costa et al., 2024) presents novel cross-country firm-level evidence on the effect of heat stress – both slow onset events (gradual temperature increases) and extreme weather events (heatwaves) – on labour productivity. The analysis builds on a unique dataset gathering detailed weather and financial information for more than 2.7 million manufacturing and services firms across 23 advanced economies between 2000 and 2021, complemented with country-level information on adaptation investment. The newly constructed dataset reveals that the number of warm days and the incidence of heatwaves present an increasing pattern in the period of analysis in most locations (Figure 1).
Figure 1. The number of warm days increased in most locations in the sample period
Note: The maps show the change in the average number of days in the year where the daily maximum temperature rose above 30°C, in the last five years of the analysis period (2016-2021) relative to the first five years (2000-2004).
Source: Costa et al. (2024) based on data from Orbis and ERA-5 reanalysis data (Copernicus Climate Change Service).
How do extreme temperatures affect firm productivity?
We find that both more frequent high-temperature days and the occurrence of heatwaves lead to substantially reduced labour productivity (Figure 2). Ten extra days above a temperature of 35 degrees Celsius in a year result in a 0.3% reduction in firms’ annual labour productivity. This effect is comparable, for example, to the decrease in productivity following a 5% rise in energy prices (André et al., 2023). One additional heat wave lasting at least five days, in turn, causes a 0.2% reduction in firms’ annual labour productivity.
Figure 2. Higher temperature negatively affects productivity
Note: Bars represent estimated coefficients and vertical lines the respective confidence intervals. Each bar is a different estimation. In Panel A each estimation differs with respect to the definition of the temperature variable, which is either the number of days above 30ºC, or above 35ºC or above 40ºC. In Panel B, each estimation differs with respect to the definition of heat wave, varying both the temperature threshold above which temperature has to rise for a heat wave to have occurred (90th and 95th percentile of the local historic mean) and the minimum number of consecutive days this temperature needs to have occurred.
Source: Costa et al. (2024)
This effect is more pronounced in less productive and smaller firms, and exacerbated by longer heat waves, high humidity, and low wind speeds. We also find that the negative productivity impacts exhibit a non-linear relationship with rising temperatures and persist for two years before fading. The heterogeneity of the impact across firms suggests differences in not only exposure but also vulnerability to heat stress. Larger firms, for instance, may have greater resilience to rising temperatures thanks to better financial resources, access to advanced technology, and knowledge of behavioural adaptation practices.
The analysis suggests some degree of adaptation may have already taken place. Firms in warmer locations and those more used to experiencing heatwaves exhibit lower productivity losses under similar temperature extremes. Additionally, both the implementation of National Adaptation Plans and firm-level investment in adaptation are also linked to reduced adverse effects of heat stress on productivity. However, the extent of current adaptation remains limited: higher temperatures relative to an already warm average result in more significant productivity losses, and there is no evidence of adaptation to severe extreme temperatures.
Directions for policy
In highlighting the substantial economic impacts of temperature-related changes, our analysis underscores important productivity and growth challenges posed by both gradual and disaster-driven climate impacts, providing valuable insights for policy making. First, sustained efforts in climate change mitigation are key to attenuating the increase in temperatures and the intensity and frequency of heat waves. This is particularly important given the non-linearity of costs to extreme temperatures and the limits to the effectiveness of adaptation suggested by our analysis.
Second, our results stress the urgent need to limit the economic impacts of heat stress through enhanced adaptation measures, tailored to different national and regional contexts to account for relevant heterogeneities in impacts and capacity. Where barriers to effective private sector engagement exist, like information and knowledge gaps, financial constraints, or coordination failures, policy efforts could prioritise promoting private sector adaptation. For example, while more than 60% of firms in the European Union report being impacted by the physical risks of climate change, only slightly over one-third have taken concrete steps to build resilience (EIB, 2023).
Policymakers can support firms, particularly small and medium-sized enterprises, through targeted economic incentives such as subsidies or tax breaks to encourage investments in heat-resilient infrastructure and technical measures, like green roofs or advanced cooling systems. Additionally, providing information to firms can drive behavioural changes, such as adjusting work schedules to avoid peak heat periods. Complementary direct public investment may be necessary, for example in changing urban structure, climate-proofing transport systems, or investing in adaptation technology R&D.
Lastly, heat stress is just one of many climate-related challenges confronting economies. Other slow onset and extreme weather events can pose significant risks for firm-level performance and broader macroeconomic outcomes. Our upcoming research dives into these risks from two angles: we explore how flooding impacts firm performance – focusing on output, capital, and investment – and how regional macroeconomic results are impacted by extreme weather events. Together, these efforts aim to provide a more comprehensive understanding of the economic risks posed by climate change and inform the development of robust macroeconomic structural models.
References
André, C. et al. (2023), “Rising energy prices and productivity: short-run pain, long-term gain?”, OECD Economics Department Working Papers, No. 1755, OECD Publishing, Paris, https://www.oecd.org/en/publications/rising-energy-prices-and-productivity-short-run-pain-long-term-gain_2ce493f0-en.html.
Costa, H. et al. (2024), “The heat is on: Heat stress, productivity and adaptation among firms”, OECD Economics Department Working Papers, No. 1828, OECD Publishing, Paris, https://doi.org/10.1787/19d94638-en.
EIB (2023), “European Overview – EIB Investment Survey”, European Investment Bank (EIB), Vol. ISBN: 978-92-861-5609-0, https://www.eib.org/en/publications/20230285-econ-eibis-2023-eu.
IPCC, 2021. Climate Change 2021: The Physical Science Basis, Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change, https://www.ipcc.ch/report/ar6/wg1.