Housing Vouchers and School Discipline: Evidence from Houston

• In Event: Promoting Housing Mobility for Families in the Housing Choice Voucher Program

Abstract

The Housing Choice Voucher (HCV) program is the largest federal low-income housing subsidy in the United States, providing rental assistance to enable families to access neighborhoods they otherwise could not afford. Because housing shapes the neighborhoods, schools, and peer environments children experience, understanding the educational consequences of housing assistance has become a critical question at the intersection of housing and education policy. Yet existing evidence on vouchers and student outcomes remains mixed—the Moving to Opportunity experiment found long-term earnings gains but null short-term educational impacts, while more recent work from New York City documented improvements in test scores. Importantly, nearly all causal studies focus on northeastern and midwestern metropolitan areas, limiting applicability to the rapidly growing Sun Belt cities that now serve a disproportionate share of low-income renters and voucher holders. These cities differ substantially in housing market structure, racial composition, and policy environment—most notably, landlords in Texas may legally refuse voucher holders. This study provides the first causal evidence on HCV’s educational impacts in Houston, Texas, the fourth-largest city in the nation. We address two research questions. First, what is the causal impact of HCV receipt on students’ academic and behavioral outcomes? Second, do effects vary by age at voucher receipt, with earlier receipt producing larger gains? We use comprehensive administrative data linking records from the Houston Housing Authority and the Texas Education Agency, spanning 2000–2022. These data track households from three major waitlist openings (2006, 2012, and 2016) and include student enrollment, demographics, standardized test scores (grades 3–8), disciplinary records, high school graduation, college enrollment, and annual earnings. Our identification strategy leverages the quasi-random timing of voucher receipt among waitlisted families. Conditional on waitlist entry, the timing of voucher allocation depends primarily on funding availability and waitlist position rather than family characteristics. For time-varying outcomes, we implement the Callaway and Sant’Anna (2021) difference-in-differences estimator using not-yet-treated students as the comparison group. For one-time outcomes such as high school graduation, we estimate models measuring the effect of cumulative voucher exposure before age 18. Preliminary results reveal null effects on math and reading test scores in grades 3–8. However, we find meaningful impacts on behavioral and long-term attainment outcomes. Voucher recipients are 2.7 percentage points less likely to receive disciplinary action (relative to a base rate of 22.3%), with reductions emerging in the first year and growing over two to three years. We also find that an additional year of voucher exposure increases high school graduation rates by 0.4 percentage points, and students who received vouchers before age 12 show a 4.6 percentage point increase in graduation rates compared to those receiving vouchers later. These findings suggest that in Houston’s deregulated housing market, HCV receipt improves long-term educational attainment primarily by reducing disciplinary problems rather than boosting short-run academic achievement. The stronger effects for earlier recipients raise important questions about whether benefits are driven by increased residential stability or mobility into better-resourced neighborhoods and schools.

Authors

• Hojung Lee, Rice University
  Presenting Author

• Anna Rhodes, Rice University
  Non-Presenting Co-Author

• Jeremy E Fiel, Rice University
  Non-Presenting Co-Author

• Jiaxin Li, Rice University
  Non-Presenting Co-Author

• Rui Zeng, Rice University
  Non-Presenting Co-Author