Smart Cville Data Project: Analysis of Housing Affordability Trends in Charlottesville 2000-2019; Propagation of Algorithmic Bias and Discrimination in US Housing Policies

Nationwide urban growth and the plummeting ability to afford housing and rental prices reinforce the need to assess current housing policies and the disproportionate effect they may be having on underrepresented groups of people. My thesis and technical project address the inequalities in the housing market from two different perspectives. My STS research project details the propagation of algorithmic bias in the housing market due to current US housing policies. The Fair Housing Act’s New Rule limits the responsibility that housing providers face in ensuring that the algorithmic models they use are making unbiased decisions. Algorithms are able to exasperate historic racism through processes like housing advertisements and the rental tenant selection. We must ensure that the policies in place to protect against discrimination are actually doing so instead of perpetuating historically racist data and decision-making.

My STS research complements the technical work I completed over the course of 2020 in collaboration with local nonprofit Smart Cville. The aim of the project was to analyze the changes in Charlottesville’s real estate prices and assessment values during the past 20 years in light of the recent urban growth and lack of affordable housing. The research and a conversation with a member of the Charlottesville City Planning Commission revealed that the average sales price of real estate in Charlottesville has risen from below the national mean in 2000 to $100,000 above nationwide prices in 2020. Qualitative input from someone familiar with the Charlottesville housing market revealed that possible reasons for the decrease in housing affordability are historical zoning regulations, the prevalence of home flippers, and the rise in UVA’s enrollment.

Both my research paper and my technical project report underscore the need for policy amendments or new policies that makes housing more affordable and accessible, especially to marginalized groups. I think my STS thesis paper would have benefitted from an analysis of a concrete housing algorithm and picking apart the components of the algorithm that are vulnerable to injected bias. In terms of my technical work, an important expansion to the project would be have the code scrape directly from the database as opposed to using locally stored copies. This would ensure that the work stays current and relevant. The conversation around housing affordability and bias perpetuated through algorithmic decisions needs to continue to take place and there is important work that needs to be done in order to ensure equitable practices and a more accessible housing environment.

School of Engineering and Applied Science
Bachelor of Science in Computer Science
Technical Advisor: Daniel G. Graham
STS Advisor: Sean M. Ferguson
Technical Team Members: Allen Lang