GOVERNMENT ACQUISITION OF HOMES TO REDUCE FLOOD DAMAGE, HOUSEHOLD WILLINGNESS TO PARTICIPATE, AND IMPLICATIONS FOR ACQUISITION POLICY

Studies show that a relatively small proportion of properties are responsible for a large proportion of the National Flood Insurance Program's claims. While several household-level flood risk mitigation measures exist to address the pervasive losses, government acquisition of at-risk homes is the most effective household-level flood mitigation approach to eliminate the flood risk to properties, but also the most expensive. The program, despite its potential, is arguably the most controversial and thus under-utilized by homeowners. To encourage participation, flood risk mitigation officials and policy makers want to know the factors that affect homeowners' willingness to participate in the program. So, to better inform policy makers, in this dissertation I examine empirically the factors that affect homeowner participation, including acquisition contract attributes, future insurance pricing, and availability of alternative mitigation efforts. Thus, this dissertation contributes to the growing literature on natural hazard risk mitigation and decision-making by providing evidence on the factors that influence homeowner participation in the government acquisition program. Also, this work provides valuable information on the potential effect of proposed programmatic changes on participation. In addition to attributes of the offered program, this dissertation explores how observables, like homeowner demographics and flood risk to the home, and latent characteristics, like homeowners' perceptions of flood risk, also impact willingness to participate. The dissertation consists of an introductory chapter and three self-contained papers organized into three chapters. In the first paper, which is presented in chapter 2, I combine homeowners' stated preference, socioeconomic, and flood risk data to examine acquisition price effects on participation and provide willingness to accept estimates. This paper has been published in the Southern Economic Journal. The second paper builds on the first paper in several ways. I use data from a national survey that elicits homeowners' stated preference for household-level flood risk mitigation, to examine how proposed changes to the buyout program, coupled with availability of alternative mitigation options like home elevation and future insurance pricing, will affect buyout participation. I estimate a conditional logit model and a random parameter logit model and provide willingness to accept estimates. In the final paper, which is presented in chapter 4, I evaluate the policy implications of utilizing subjective or objective risk indicators in hazard risk management tools. In the paper, I derive a single index each for subjective and objective risk using factor analysis, then integrate these indices with the discrete choice data from chapter 3. I then estimate conditional logit models to derive willingness to accept estimates to better understand how different risk indicator types affect hazard policy recommendations