Heat and the productivity of work
Extreme heat and heat waves appear on 47 recent papers, and temperature keywords on 208, but only 29 of those carry a productivity, labour, worker or firm keyword. As summers get hotter, how much output is lost on hot days, in which industries, and does adaptation close the gap?
What prompted this
Since October 2025, 208 papers carry an extreme-heat, heat-wave, heat-stress or temperature keyword (105 in economics journals, 69 in AI): "extreme heat" on 34, "heat waves" on 13. Only 29 of the 208 also carry a productivity, labour, worker, output, firm or absenteeism keyword; "labor productivity" alone appears on 94 papers, mostly about other causes.
Questions
- Using daily plant-level data, how much does output fall on days above 35°C in manufacturing, construction and agriculture, and does the loss persist into the following days?
- Do firms in hotter regions adapt (shifts, cooling, automation) so that the marginal effect of heat falls over time?
- Who bears the loss: workers through pay, firms through profits, or customers through delays?
Data that could help
Daily production or output records from firms, weather station and reanalysis data (ERA5), and worker-level time and attendance data.
Next step
Match daily output data from a multi-plant manufacturer to local temperatures and estimate the output loss per hot day by plant type.