The Association between Housing Assistance Programs and Elementary School Outcomes in Wisconsin

Background/Purpose: The federal government makes large investments in supporting affordable housing for economically vulnerable families. Housing and Urban Development (HUD) spends over $30 billion annually to provide housing vouchers and project-based units for families and individuals with low incomes. This assistance is vital to many as housing costs have out-paced inflation in recent years, increasing nearly 20% in inflation-adjusted dollars since 2000. The housing assistance programs, including the housing choice voucher, public housing, and project-based Section 8 program, potentially provide important resources that may have ripple effects on family well-being by possibly raising disposable household income, improving the quality of housing, and increasing housing stability. Despite the importance of housing assistance for family well-being, there is little known about whether such assistance is associated with children’s educational outcomes, and to date results are mixed. Thus, we explore the association between housing assistance programs and a child’s academic achievement. Because of historical and current systemic racism in housing policies and place-based inequality, we also consider whether any associations differ across racial and ethnic groups.

Methods: we use rich administrative data from the Wisconsin Administrative Data Core (WADC) which contains information about families’ earnings and other forms of public assistance with children’s schooling outcomes. We combine this data with restricted housing data from the Department of Housing and Urban Development (HUD), which provides information on the entire population of recipients of housing assistance programs from 2007 to 2020. Our approach is to match children whose families consistently received housing assistance during early elementary school, to children in similar families who have not received assistance in their school district. The rich administrative data enables matching to be based on family income, family size and structure, prior benefit receipt, race and ethnicity, and other key family characteristics. Our sample is comprised of 56,603 children, including the treatment group who have taken housing assistance from age 5 to 9. The outcomes are measured by attendance rate and grade retention. Using Inverse Probability of Treatment Weighting (IPTW), we constructed a well-balanced sample matching children within the same school districts who reside with parents receiving housing assistance with children whose parents do not receive such support. Finally, we estimate associations using IPTW weighted regressions.

Results: Our preliminary results suggest that children whose families received housing assistance programs for five years starting in kindergarten have slightly higher rates of school attendance (0.004, p<.01) and are less likely to be retained (-0.017, p<.01). These results are concentrated within Black subgroup, which constitute half of the treatment group, demonstrating a more pronounced impact on attendance (0.010, p<.01) and grade retention (-0.024, p<.01). The effects mostly do not vary by gender and type of housing assistance, although attendance rates for boys and those receiving vouchers are not statistically significant.

Conclusions and Implications: Our findings indicate that housing assistance programs are positively associated with a child’s academic achievement. These results offer new evidence on the potential spillover effects of housing assistance programs, emphasizing their importance in reducing racial disparities in education.