Association for Public Policy Analysis & Management

Researchers have long sought to understand the drivers of disaster adaptation. While prior research has shown that recent disaster incidents often provide the window of opportunities for adjustment and motivate adaptation to reduce risks, less attention has been paid to the role of government aid. By providing financial assistance for post-disaster reconstruction and recovery, government disaster aid has the potential to encourage public and private agents to adopt risk-mitigating measures and technologies that enhance resilience to future disaster risks.

In this paper, we focus on the disaster loans provided by the Small Business Administration (SBA) and empirically examine whether the SBA disaster loans encourage private adaptation and lead to fewer disaster damage in the United States. SBA offers disaster loans to individuals and businesses at a subsidized, lower than market rates to rebuild homes after a major disaster such as flooding. The size of the loan is determined by sustained damage after all payouts from the insurance and the public aid are accounted for. In the case of flooding, the SBA mandates recipients to carry the flood insurance, but does not require them to invest in risk reduction measures (e.g., elevate homes, install shutters, and fortify the base). Given such policy designs, the SBA disaster loans could potentially affect private adaptation through multiple channels. If recipients use the SBA disaster loans to retrofit or build more resilient homes post disasters, they may better adapt to natural disasters and experience fewer damage in subsequent shocks. At the same time, if the recipients purchase flood insurance as required and rely exclusively on the risk-transfer insurance mechanism rather than reducing risk ex ante, the SBA loans may not lead to other forms of adaptation.

To empirically test the effect of SBA loans on voluntary private adaptation, we compiled a panel data set combining property losses from flooding, amount of SBA loans, and various socio-economic variables (e.g., personal income, population density, housing stock, flood insurance policy) at the county level over the period 1989-2015. We focus specifically on flooding because it is the most prevalent natural hazard in the U.S and has by far caused the largest amount of damage only second to hurricanes. Because the SBA loans are provided post a major disaster event, we also control for a county’s prior flooding condition using objective precipitation data. By employing a panel fixed effects model with instrument variables, we find limited adaptation behavior associated with the SBA loans.

Our study has important implication for disaster aid policy in the United States. The recent catastrophic disasters such as Hurricane Harvey have renewed the discussion about the lack of disaster preparedness across U.S. communities, and also raised the question of whether the federal disaster programs have been effective in mitigating disaster risks. Our research sheds light on the efficiency of disaster financing, particularly related to the provision of low-interest loans for post-disaster reconstruction.