Equity in Flood Mitigation Funding: Analyzing Social Vulnerability and Public Investment in Houston, Harris County, TX

Abstract

Flooding is the most frequent and financially damaging natural disaster in the United States, causing more than 8 billion (USD) in annual damages (University of Maryland & Texas A&M University, 2018). With over 30 million Americans living in high-risk flood zones, climate change is expected to further intensify the frequency and severity of flood events (US Water Alliance, 2020). While investments in flood infrastructure aim to reduce risk, mitigation efforts often fail to address deep-rooted inequalities. Research has shown that low-income communities, communities of color, and other vulnerable groups face greater risks and longer recoveries due to geographic exposure, limited resources, and exclusion from decision-making (Villa et al., 2020). In August 2017, Hurricane Harvey struck the Houston area, flooding over 200,000 homes, with the most severe impacts falling on socially vulnerable communities. In the year that followed, Harris County voters approved a $2.5 billion bond to fund flood mitigation projects. This major investment raised questions about how mitigation resources were distributed and whether funding would reach those most in need. This thesis analyzes the relationship between social vulnerability and the allocation of flood mitigation funding in Harris County. It examines whether bond and housing buyout funds were distributed equitably across watersheds relative to community need, using the Centers for Disease Control and Prevention’s Social Vulnerability Index (SVI) as a measure of vulnerability. A distributive equity framework is applied to evaluate whether funding aligned with both exposures to flooding and social disadvantage. This study analyzes funding across 22 watersheds, aggregating census tract-level SVI data to each watershed. It examines bond and buyout funding allocated using dependent variables: total funding and funding normalized by square mile, per capita, and per home flooded. Independent variables include overall SVI and four CDC-defined themes: Socioeconomic Status, Household Composition & Disability, Minority Status & Language, and Housing Type & Transportation. Regression models indicated statistically significant negative relationships between bond funding per home flooded and several SVI indicators: Socioeconomic Status (p = .003), Racial & Ethnic Minority Status (p .001), and Overall SVI (p = .011). These findings suggest that more socially vulnerable watersheds received proportionally less mitigation funding relative to the severity of their flood damage. No significant relationships were found between SVI and housing buyout allocations, indicating that investments were not systematically directed toward the most at-risk communities. These results highlight a disconnect between equity goals and implementation. Despite including SVI in project scoring, Harris County’s funding patterns did not equitably reflect vulnerability when adjusted for flood impact. These findings underscore the need for more intentional and targeted equity measures in flood mitigation planning.