ESSAYS ON THE IMPACTS OF CLIMATE CHANGE ON NORTH CAROLINA USING UNIQUE HOUSING AND FLOOD RISK DATA SETS
Loading...
Date
July 2024
Access
Authors
Kops, Jason
Journal Title
Journal ISSN
Volume Title
Publisher
East Carolina University
Abstract
Climate change is expected to introduce new challenges to communities in North Carolina and across the globe. However, our understanding of how and where losses from climate change will manifest is limited. This makes it difficult for policy makers to craft policies to meet these upcoming challenges. By combining novel flood risk data at the parcel level in North Carolina with transactional home sale data from Zillow, this research estimates a statistically significant, negative, and nonlinear impact of increased flood risk on a home’s value. Based on projections of future flood risk, these results are extrapolated to provide parcel-level, real-dollar estimates of the impact of a changing climate on the housing stock value in North Carolina. Additionally, these estimated costs are shown to have disproportionate effects on different segments of the state and different demographic groups within the population.
This research adds context to the cost-benefit analysis of flood risk mitigation and adaptation policy in North Carolina. Given the large majority of North Carolina communities participating in the National Flood Insurance Program, a vast number of this state’s citizens depend on its continued financial solvency. The distributional impacts of increased flood risk evidenced in this research should allow for a more robust policy regarding the allocation of resources. Furthermore, the estimated losses in property value can illuminate potential shortcomings in future property tax revenue which would affect public services such as education. Finally, the proliferation of the flood risk data in this research would provide additional price signals and allow property owners to make more well-informed decisions.
In Chapter 1, we combine a North Carolina subset of real estate transactions from the Zillow Transaction and Assessment Database (ZTRAX) with continuous flood probability metrics from the First Street Foundation Flood Lab in a hedonic model to estimate the premium of reduced flood risk beyond the impact of FEMA floodplain designation.
In Chapter 2, we extrapolate the results of our hedonic model to estimate parcel-level changes in housing stock value that are aggregated at the tract, county, and state levels using flood risk projections from the First Street Foundation for the years 2036 and 2051. We use these property value changes to estimate declines in annual property tax revenue. Additionally, we use regression analysis to investigate which socioeconomic and demographic groups will be most impacted by increases in flood risk.
In Chapter 3, we replicate the methodology from Chapters 1 and 2 except our hedonic model uses vigintile regression analysis to investigate the heterogeneous preferences between income groups using home value as a proxy.