The Rent Eats First: Did Ending Pandemic-Era Eviction Moratoria Increase Food Insufficiency Among US Renters?

• In Event: Impacts of Eviction and Access To Opportunity in Rental Assistance Programs

Abstract

Housing insecurity and food insufficiency in the United States are fundamentally intertwined: increased rent burden is associated with increased reported food insufficiency. As these outcomes worsen in tandem for renters, more American households are faced with the realization that the ‘rent eats first’ (Desmond, 2016). Prior research demonstrates that certain housing policy interventions enacted in response to the COVID-19 pandemic effectively reduced food insufficiency among renters in the United States while in effect (e.g. An, et al., 2022). This study leverages the Supreme Court’s sudden cessation of the pandemic-era national eviction moratorium as a natural experiment, measuring the effect of the policy’s expiration on food insufficiency among renting households. Difference-in-differences analyses, in which the pre- and post-treatment periods are bifurcated by the Court’s decision (effectuated on August 27, 2021) and treatment exposure is determined by housing tenure (renting vs. home-owning households), are conducted using eight waves of nationally representative data from the U.S. Census Bureau’s Household Pulse Survey (N = 341,108). The focal multivariable linear probability model identifies a 0.80 (95% CI: 0.24 -1.4) percentage point increase in the prevalence of food insufficiency among renters after the national eviction moratorium is lifted (statistically significant at p < 0.01), adjusting for gender, race, ethnicity, age, educational attainment, employment status, income, marital status, parental status, Child Tax Credit receipt, and SNAP receipt. This causal relationship withstands a battery of tests: differential shocks (e.g. CTC and EIP roll-out) are eliminated by bounding the study period and restricting the sample, compositional balance is assessed via balance tests, a sensitivity analysis finds no significant threat of spill-over effects, and additional tests (of pre-treatment trends, seasonal trends, and spurious correlation with theoretically unrelated outcomes) mitigate confounding explanations. Secondary analyses illustrate stronger treatment effects among Black renters, female renters, low-income renters, renters with children, and renters living in states with less generous social safety nets. Furthermore, event study specifications of both main and secondary analyses illustrate that, except for households with children, these treatment effects emerged gradually over time, with effects continuing months after the program’s cessation. While limitations in causal interpretation and across forms of validity remain, restricting the generalizability of the study’s findings, this study makes three contributions to prior literature. First, it corroborates extant work assessing the relationship between eviction moratoria (state-level implementation) and food insufficiency using new data. Second, it constitutes the first evaluation of the effect of a uniform national eviction moratorium on food insufficiency. Third, it explores sub-group analyses both to corroborate prior findings about the social strata disproportionately exposed to eviction risk (e.g. in terms of race, gender, income, and the presence of children), and to determine whether higher state-level social safety net generosity attenuates the treatment effect. Tertiary analyses, currently in progress, will use variation in state-level moratoria to compare renters in states with and without protections in place, further exploring the mitigating effect of strong state-level social safety nets on the relationship between moratoria cessation and food insufficiency.

Author

• Matthew Gannon, Johns Hopkins University
  Presenting Author