The Spatial Hazard Events and Losses Database for the United States (SHELDUS 23), maintained by the Center for Emergency Management and Homeland Security at the University of South Carolina, provides the most comprehensive standardized repository of county-level natural hazard events and associated direct economic losses available for the United States. The database encompasses the period from January 1960 to December 2023, with systematic collection protocols ensuring temporal consistency and spatial coverage across all counties in the continental United States.
SHELDUS 23 incorporates detailed information on hazard type classification, temporal occurrence, precise geographic location, and multiple categories of economic and human impacts, including property damage, agricultural losses, injuries, and fatalities. The database covers the full spectrum of natural hazards relevant to the continental United States, encompassing meteorological events (hurricanes, tornadoes, thunderstorms), hydrological events (floods, flash floods), climatological events (droughts, extreme temperatures), and geophysical events (earthquakes, wildfires). For this analysis, this study focuses on counties within the continental United States (CONUS) for the period from 2012 to 2022.
The Geographic Wealth Inequality Database (GEOWEALTH-US), developed by Suss et al. (2024), represents the first comprehensive effort to estimate wealth levels and wealth distribution across multiple spatial scales within the United States. The database provides temporal coverage from 1960 to 2020, with econometric projections extending to 2022. This dataset addresses a critical gap in socioeconomic research by capturing accumulated financial assets and liabilities rather than annual income flows, thereby offering superior indicators of long-term economic capacity for disaster mitigation, response, and recovery activities.
The methodological innovation of GEOWEALTH-US lies in its integration of multiple data sources, including tax records, surveys, and administrative data, to construct spatially disaggregated wealth estimates. The database reveals that inter-regional wealth disparities have expanded more dramatically than income disparities over recent decades, demonstrating the growing importance of wealth-based measures for understanding economic inequality. This study utilizes the 2022 wealth estimates as the most temporally proximate cross-sectional measure of county-level economic capacity available for integration with contemporary hazard impact data.
County
County | PUMA | Tract
County | PUMA | Tract
Income / Wealth Level and Wealth Distribution
According to the common definition, income denotes the flow of economic resources including wages and investment returns, while wealth refers to the value of assets minus liabilities. Wealth is primarily accumulated through three means (Howell & Elliott, 2019): wages and other forms of income that accrue over the short term; returns on investments that accrue over the longer term; and intergenerational transfers that accrue across life times (Alvaredo et al., 2013; Charles & Hurst, 2002; Fox, 2016; Keister, 2014; Oliver & Shapiro, 2006; Volscho & Kelly, 2012).
High income level does not necessarily imply a high wealth level. For example, some young professionals in high tech regions (like San Jose) could earn the above average income but have limited net wealth due to the considerable student debt or high interest mortgages. Some retired senior people might have significant savings and living in high value properties but only have modest annual incomes. These examples show that considering both income and wealth is essential for identifying the vulnerabilities and resilience in socio-economic hazard analysis (Oliver & Shapiro, 2006).
Wealth distribution represents the concentration of economic resources across a certain amount of the population. Two communities with similar overall wealth levels can exhibit drastically different distributions. For instance, one could be characterized by a small elite holding most local real estate, while the other might have broader middle class homeownership with smaller wealth gaps among households. This variation in distributions between and within different regions (Suss et al., 2024) can shape different preparedness, experiences, and recovery patterns from hazards.
In the U.S., wealth inequality has been growing over recent decades and has detrimental effects on social outcomes, regarding educational attainment, physical health, and emotional well-being (Hansen, 2014; Keister, 2014; Shapiro, 2017). Wealth inequality has been historically associated with racial inequality, as the inequalities of the past not only play forward to influence those of the present and future, they also link with historical inequalities of race that concentrate in space as well as time (Fox, 2016; Gotham, 2014; Shapiro, 2017). The inequality returns on housing investments and disparate intergenerational transfers of assets (Alvaredo et al, 2013; Charles & Hurst, 2002; Fox, 2016; Hansen, 2014).
Wealth is relevant to Natural Hazard Damage
Unequal infrastructure investment and development (Berke, 1982; Holway & Burby, 1993; Campbell, 1996; Mileti, 1999; Vale, 2017)
Hazard insurance and financial burden (Kousky et al., 2020)
Mitigation priorities and political influence (Michael et al., 2023)
Unemployment and displacement (Kirby, 1976; Rossi, 1996; Handy & Niemeier, 1997; Kwan, 1998; Aaronson, 1999; Dietz & Haurin, 2003; Fussell & Harris, 2014; Caschili et al., 2015; Elliott & Howell, 2016; Shi et al., 2020; Papilloud & Keiler, 2021)
Post-disaster recovery and buyouts program (Fraser et al., 2003; Loughran et al., 2019; Mach et al., 2019; Elliott et al., 2020)
…….