Seasonal temperature variability and economic cycles

In many countries, GDP varies by several percent from quarter to quarter. It has long been conjectured that temperature plays a role in these cycles, but previous research has either neglected its possible influence, focused on proximate rather than fundamental drivers of quarterly fluctuations, or concluded that the influence of temperature is small based on limited evidence and inappropriate methods. As climate change is projected to change the seasonality of temperature, the question appears very relevant. In this paper, I examine the effect of seasonal temperature variability on seasonal economic cycles using data on quarterly GDP from 98 countries and quarterly GVA by industry for 35 European economies. I first construct a dataset of seasonal temperature and seasonal economic production calculated in a consistent way across countries. This dataset reveals a much larger diversity of seasonal economic cycles around the world than previously reported. I then attribute these economic cycles to seasonal variation in temperature. For identification, I propose and apply a novel econometric approach based on seasonal differences that accounts for expectations. The results show that seasonal temperature has a statistically significantly positive effect on seasonal production, which is primarily due to countries with larger production in summer than in winter. Using GVA data for industry groups, I find significant effects of seasonal temperature on GVA only in industries with relatively high exposure to ambient temperature. Overall, the effect of temperature on seasonal economic cycles appears large, as in many countries the effect of temperature is strong enough to explain almost all of the observed seasonal economic cycle. Regarding future anthropogenic climate change, the results suggest that changes to seasonal temperatures will lead to a reallocation of economic activity across seasons.