← All projects

Statistical analysis

Housing policy & residential choice

Evaluating how ZIP code-level rent standards relate to housing voucher concentration and access to neighborhoods.

Collaborative policy-analysis project

PythonDifference-in-differencesGeoPandas

The question

How did the 2018 Small Area Fair Market Rent policy change the residential concentration of Housing Choice Voucher households?

The approach

  • Constructed a 2014–2022 census-tract panel from HUD housing data and compared Dallas County with Harris County.
  • Used difference-in-differences with tract and year fixed effects, alongside event-study checks and geospatial analysis.
  • Examined changes in voucher concentration and patterns across neighborhood racial composition and poverty.
Repository event-study chart showing estimated treatment effects by year with uncertainty intervals
Event-study estimates from the project repository. Read the notebooks for the model specification and interpretation. Source figure

What this explores

The analysis finds an association between the policy change and lower voucher concentration in the treatment area. Interpreting that pattern requires looking beyond a single estimate to pre-policy trends and the limits of the available data.

Scope & limitations

Pre-policy trend differences complicate causal interpretation. HUD suppresses small household counts, and the data do not track where individual households relocate.

Team: Lingyue Hao, He Jiang, Ammy Lin, Myla Simmons, and Anvita Suresh.

Project summary and attribution are based on the repository documentation.