How do 1930s HOLC redlining maps affect modern political outcomes?
Abstract: There is substantial evidence showing that discriminatory mortgage lending practices, also known as “redlining”, have significant and long-lasting impacts on the social and economic well-being of those living in affected areas. We fill a gap in the literature by investigating whether redlining affects political outcomes independently of known socio-economic channels. Using a cross-sectional dataset of mortgage lending desirability grades assigned by the Home Owners Loan Corporation (HOLC) in the 1930s, we analyze how redlining affects people’s voting behaviour in the 2020 United States general election. We use a multiple linear regression framework to estimate how the proportion of a census tract graded C or D affects electoral outcomes in that tract while controlling for race, education, age, income, gender, and city fixed effects. Our results show that redlining has little association with modern political outcomes, in contrast with its significant and long-lasting socioeconomic consequences. Our results, however, cannot be deemed causal due to the potential for omitted variables bias and measurement error and requires additional external validation to be generalized to other time periods and other forms of credit rationing.
Jia Jun (Jacob) Li
This project is co-authored with a classmate. The sections I worked on were the following:
- Abstract
- Simple Linear Regression
- Multiple Linear Regression
- Conclusion
We contributed equally to the research idea and regression model. My co-author performed the literature review and data downloading. I coded all the analysis and figures in STATA and formatted the paper in LaTeX.