Federal Disaster Spending

Erda (2026) studies firm dynamics and aggregate productivity following presidentially declared floods in the U.S., and the mediating role of federal disaster spending. Under the Stafford Act, disaster declarations unlock various programs, including public assistance for infrastructure rebuilding and grants to households from the Federal Emergency Management Agency, and low-interest disaster recovery loans to households and firms in affected counties from the Small Business Administration.

Using confidential plant-level microdata from the U.S. Census Bureau and an event study design exploiting plausibly exogenous variation in flood timing, the paper compares the plant- and industry-level manufacturing productivity and dynamics in flooded regions to their pre-flood baselines over the 1977–2017 window. The findings show that the post-flood period sees accelerated creative destruction (defined as “the continual redistribution of resources from less to more productive firms that drives productivity growth in market economies”): low-productivity plants exit and their used machinery is acquired by positively selected new entrants and young plants. Survivors also upgrade their capital as they rebuild and see productivity gains.

Federal disaster spending facilitates these outcomes by expanding firms’ access to financing and crowding in bank loans, particularly for young and small firms that ultimately make outsized contributions to overall reallocation and creative destruction. For a counterfactual set of non-declared floods––cases where federal spending was not available after severe floods––creative destruction and reallocation are muted: exits are productivity-indiscriminate and entrants show no productivity advantage compared to their pre-flood counterparts. Consequently, aggregate productivity suffers, suggesting a distortionary role of financing constraints. Relative to pre-disaster baselines, per capita incomes also decline in these non-declared flood settings, whereas federal spending undoes this decline or even boosts incomes. Ultimately, the relative income gains after declared floods compared to non-declared ones generate additional tax revenue that entirely offsets the initial upfront costs of federal assistance.

The paper highlights a novel channel through which federal disaster spending supports recovery in flooded regions––efficiency-enhancing reallocation and creative destruction––and shows that the program pays for itself through tax revenues.