How the marketing of real estate properties explains mortgage applicants by race and income

Bibliographic Data

ID21247446
AuthorsIsabelle Nilsson (0000-0003-0403-1309, University of North Carolina at Charlotte, corresponding author), Elizabeth C Delmelle (University of Pennsylvania), Elizabeth Delmelle (0000-0003-3735-8359, California University of Pennsylvania)
Year2025
Volume52
Issue6
Pages1407-1423
Publication date2025-07-01
Peer ReviewedYes
Open AccessYes
TypeARTICLE
VenueEnvironment and Planning B Urban Analytics and City Science (JOURNAL)
Journal identifiersISSN: 2399-8083 • E-ISSN: 2399-8091
PublisherSAGE Publications (PUBLISHER • US)
DOI10.1177/23998083241287956
OpenAlexW4403670534
LanguageEN
References cited54

In this article, we study how the marketing of single-family homes explains the racial and income makeup of mortgage applicants in a neighborhood. We use a case study of the robust housing market of Charlotte, North Carolina, and annual, longitudinal real estate listing advertisements alongside mortgage lending data, to demonstrate how the share of properties advertised a certain way in a neighborhood in 1 year explains shares of mortgage applicants by race and income the following year. We classify property advertisement text using a semi-supervised learning algorithm into five categories following a housing investment and disinvestment to renewal continuum. We find stark racial disparities in mortgage applicants by housing type, even after controlling for income. We find that Black applicants nearly exclusively apply for mortgages in neighborhoods with a high share of properties advertised as disinvested with little profit-making promise. High-income White applicants rise as the share of advertised properties becomes more homogenous

Actuarial science · Business · Insurance policy · Mortgage insurance · Real estate · Shared appreciation mortgage · Sociology · Housing Market and Economics · Urban, Neighborhood, and Segregation Studies · Urbanization and City Planning · Finance · Marketing

Tools

Open DOIOpen Access