Raided by the storm: How three decades of thunderstorms shaped U.S. incomes and wages

Abstract

Climate change and weather events are increasingly affecting the macroeconomic performance of countries and regions. However, their effects on income inequality are less understood. We estimate the dynamic impact of thunderstorms on income and wages and reveal a robust asymmetric effect. We leverage a comprehensive dataset covering more than 200,000 events affecting contiguous U.S. counties across three decades. Storms have caused the highest number of billion-dollar disaster events since the eighties, but they have the lowest average event cost. They are short-lived, locally confined, relatively frequent, difficult-to-predict, and hazardous albeit not fully destructive events. While such features are convenient for the identification of impacts, previous studies mostly focused on more extreme events. We document a robust negative association between storm activity, income and wages growth. While income tends to recover in the long run, wages exhibit a significantly more stubborn decline, suggesting persistent and adverse impacts on (functional) income inequality. A one standard deviation increase in wind exposure generates a loss of 0.15% (0.21%) in wages after three (nine) years; incomes fall by a larger extent initially (0.19% after three years) while fully recovering in the longer run. In addition to their notable asymmetry, such estimates are non-negligible—especially given the downward rigidity of U.S. wages. Our analyses also highlight a lack of effective adaptation and stronger negative impacts in economically disadvantaged regions. Finally, we find evidence for a sizable shock-absorbing role of federal assistance.