Elevated annual average temperature has been found to impact macro-economic growth. However, various fundamental elements of the economy are affected by deviations of daily temperature from seasonal expectations which are not well reflected in annual averages. Here we show that increases in seasonally adjusted day-to-day temperature variability reduce macro-economic growth independent of and in addition to changes in annual average temperature. Combining observed day-to-day temperature variability with subnational economic data for 1,537 regions worldwide over 40 years in fixed-effects panel models, we find that an extra degree of variability results in a five percentage-point reduction in regional growth rates on average. The impact of day-to-day variability is modulated by seasonal temperature difference and income, resulting in highest vulnerability in low-latitude, low-income regions (12 percentage-point reduction). These findings illuminate a new, global-impact channel in the climate–economy relationship that demands a more comprehensive assessment in both climate and integrated assessment models.

已发现年平均气温升高会影响宏观经济增长。然而，经济的各种基本要素都受到每日气温与季节性预期的偏差的影响，而这些偏差并没有很好地反映在年平均值中。在这里，我们表明，季节性调整后的日常温度变化的增加会降低宏观经济增长，而这与年平均温度的变化无关。在固定效应面板模型中，将观察到的日常温度变化与全球 1,537 个地区 40 年来的地方经济数据相结合，我们发现额外程度的变化会导致地区增长率平均下降 5 个百分点。日常变化的影响受季节温差和收入的调节，导致低纬度、低收入地区的脆弱性最高（降低 12 个百分点）。这些发现阐明了气候与经济关系中一个新的全球影响渠道，需要对气候和综合评估模型进行更全面的评估。