The repo contains the code files and documents of my Master's thesis at the University of Nottingham. There is no data, however, due to the large size of the data files which exceeds the Github limits (or maybe it slows down the Git working process - I'm not entirely sure which one is the main problem). The following documents provide details on the scripts and data descriptions:
- File Code/README-code-overview.txt describes the code and data wrangling process.
- File Docs/Notes-on-variables.docx describes the variables used in the analysis.
- Files Docs/Dictionary-HMDA.xlsx and Docs/Dictionary-TRI.pdf are variable dictionary for the original data.
This research explores the impact of a long-term risk (pollution exposure) on mortgage credit costs in the U.S. housing market by utilising the Home Mortgage Disclosure Act (HMDA) and the Toxic Release Inventory (TRI) dataset. In the baseline result, the analysis identifies a modest premium in mortgage rate spreads for properties exposed to carcinogens, which is consistent with the impact of other long-term risks.
The study also examines how carcinogen exposure affects mortgage interest rates across different racial groups, lender types (bank vs. non-banks), and loan purposes (refinancing vs. new purchase). The findings reveal varying impacts among racial groups, with Black and African American applicants experiencing a higher mortgage credit premium due to carcinogen exposure, exacerbating environmental disparities in the financial sector. The study also finds that lender types and loan purposes do not significantly alter the relationship between carcinogen exposure and mortgage credit costs.