State Policy Variation & Emergency Rental Assistance Take-up during the COVID-19 Pandemic

• In Event: Emergency Rental Assistance (ERA) and Eviction Prevention

Abstract

Background

Responding to fears of a wave of housing insecurity during the COVID-19 pandemic, in 2021 the federal government created the Emergency Rental Assistance Program (ERA). ERA transferred approximately $47 billion to state and local governments to implement programs aiding renters at risk of housing instability through financial assistance and other supports. Governments implementing ERA were given flexibility, within guidelines, over program rules, requirements, and administrative procedures. Consequently, programs varied on multiple dimensions. Some policy components, such as the use of self-attestation in lieu of documentation, were novel, offering lessons for future policy.

Research Questions

We merge policy data and household microdata to address two questions with regard to state-level ERA variation:

 1)    How were state-level ERA policies associated with application for and receipt of rental assistance?

 2)    Were policy components differentially related to participation among vulnerable households?

Methods

Program Data

We drew data on state-level ERA rules and procedures from the Emergency Rental Assistance Program Policy Database, a dataset documenting programs in 49 states through indicator and categorical variables. Variables were summed into indices representing policy dimensions such as generosity of benefits and accessibility.

Household Data

We merged policy data with household microdata from the Household Pulse Survey, a nationally-representative repeated cross-sectional survey of U.S. households fielded by the Census Bureau. From July 2021 to December 2023 (Pulse waves 34 to 57), Pulse asked renter households whether they had applied for rental assistance and, if so, whether they had received assistance, been rejected, or were awaiting a decision. We pooled relevant survey cohorts and restricted to renter households in states with policy data (n=265,424).

Analysis

This study examined ERA participation as a series of decision stages—whether a household applied or not, if it applied whether it received a decision, and if it received a decision whether the application was accepted or rejected—using sequential logit models. The sequential logit allows examination of the relationship between a variable and the final outcome (e.g., ERA receipt) as well as the probability of passing each transition. Standard errors were clustered by state. We estimated models examining the overall relationship between ERA policy and program participation and models interacting prior-year household income and the policy variables to consider effects on households with varying degrees of housing instability risk.

Findings & Conclusions

State ERA programs allowing more self-attestation had higher rates of rental assistance receipt (mean marginal effect 0.01), primarily through higher probability of application. Programs with larger numbers of prioritized groups (mean marginal effect 0.01) had higher rates of application but only slight differences in receipt. Policies focused on increasing access were particularly associated with program participation among individuals in the lowest-income households (e.g., the mean marginal effect on probability of receipt for the self-attestation index was 0.02 for households under $25,000 in income and extremely small for all others). Results expand on prior evidence by identifying particular policy components most effective in improving accessibility, especially for the most vulnerable households.

Authors

• Vincent A. Fusaro, Boston College
  Presenting Author

• Christopher Essem Baidoo, Boston College
  Non-Presenting Co-Author

• Kimberly Megan Collier
  Non-Presenting Co-Author

• Naoka E Carey, Boston College
  Non-Presenting Co-Author

• Rebekah Levine Coley, Boston College
  Non-Presenting Co-Author