Essays on entrepreneurship and social capital in the wake of a catastrophic disaster
Date
2019
Authors
Levitt, Ryan J., author
Zahran, Sammy, advisor
Iverson, Terry, advisor
Weiler, Stephan, committee member
Manning, Dale, committee member
Journal Title
Journal ISSN
Volume Title
Abstract
Research on the economics of disasters has seen a surge in interest in recent years following a series of high-profile events, such as the 2004 Indian Ocean Tsunami, the 2005 Atlantic hurricane season, and the 2010 Haitian Earthquake (Cavallo et al., 2011). In the United States, recent hurricanes such as Harvey, Irma, and Maria have continued to draw attention to the economic consequences associated with catastrophic natural disasters. Both global climate change and the shifting of people and economic activity towards coastal areas increase the likelihood of major climatic disasters occurring in the future (Nordhaus, 2010). This dissertation utilizes Hurricane Katrina as a case study to investigate the various ways in which disasters impact a community, and the factors that both attenuate and exacerbate these impacts. The first chapter describes a framework for identifying the effects of a disaster. Using the synthetic control method, originally proposed by Abadie and Gardeazabal (2003) and Abadie et al. (2010), the chapter identifies the long-term population effects of Hurricane Katrina across the eight most damaged counties. Results highlight significant variation in terms of the magnitude of out-migration across these areas. Cross-county differences in population outcomes are largely the consequence of the severity of housing damage. Pre-existing county characteristics, such as the percent of the population with hazard insurance, were only weakly correlated with population outcomes. Results suggest a near unit-elastic relationship between the severity of housing damage and out-migration. Notable outliers include Jefferson and St. Tammany Parish, who experienced disproportionate out-migration due to the disaster. The paper argues that this is likely due to the significant amount of commuters that work in Orleans but reside in these two counties. The decision to return and rebuild one's home is dependent on whether one's neighbors plan to rebuild, if and when the government restores utilities and infrastructure, and whether surrounding businesses return. Because of the interdependent nature of these decisions, the process of disaster recovery is often characterized as a collective action problem, in which the degree of necessary coordination increases with the extent of out-migration. The second chapter seeks to test the hypothesis advanced by Storr et al. (2016) that entrepreneurs are important first movers in post-disaster environments. The chapter expands on the framework described in chapter one and applies the synthetic control method to all counties that experienced housing damages from Hurricane Katrina and/or Rita. Using these estimated effects as a dependent variable, the chapter explores the role of entrepreneurship in disaster recovery. Results indicate positive correlations between new firm formation and disaster recovery, both with respect to initial impacts, as well as throughout the recovery period. The final chapter investigates the impacts of Hurricane Katrina on social capital in the New Orleans Metropolitan Statistical Area (MSA). Previous research has found that disasters often generate "therapeutic communities," in which altruism, trust, and charity increase following an event. However, the only empirical study to examine the impacts of Hurricane Katrina on social capital finds the opposite effect, in which the concentration of community-based organizations decreased after the disaster. Building on this work, I use the synthetic control method to identify the impacts of Hurricane Katrina on a social capital index, constructed using the concentration of community-based establishments and nonprofit organizations in the New Orleans MSA. The chapter finds that this index increased significantly after Hurricane Katrina, by approximately half a standard deviation relative to the level implied by the synthetic control. Additionally, the paper shows that this increase persists through the entire sample period, even as population levels recover in the area. Decomposing the index by its various components shows that this increase was fairly uniform across included establishment sectors and nonprofit organizations.
Description
Rights Access
Subject
social capital
natural disasters