PROJECT SUMMARY
The COVID-19 pandemic triggered a sudden and severe economic shock which hit low-income households
and households of color most severely. Many U.S. renter households were already housing-cost stressed,
leading to fears of a wave of housing instability and evictions and related effects on other forms of hardship
such as food insecurity. Given that housing instability and eviction are key social determinants of health and
health disparities, overflow effects on mental health were also expected. In 2020-2021 the federal government
allocated $46.6 billion in funding for the Emergency Rental Assistance Program (ERAP) to support state,
county, and municipal governments in addressing rental housing difficulties. The bulk of ERAP funds must be
used for financial assistance for rent and rent arrears, utility assistance and utility arrears, and a limited set of
other housing-related expenses arising from the pandemic (e.g., relocation expenses); states and
municipalities were otherwise given extensive flexibility over the distribution of funds. This project seeks to
provide the first systematic and rigorous evaluation of the ERAP policy.
The overarching goals of this study are to carefully document variation in state and local ERAP design,
including timing of adoption and implementation of various rules, and to evaluate the effects of different
approaches to ERAP using quasi-experimental methods that take advantage of differences across geography
and across time. This study has three key aims. First, we will create a unique, publicly-accessible, and user-
friendly dataset that tracks local, state, and national ERAP policies, the Rental Relief Policy Database (RRPD).
Second, we will merge the RRPD with household survey data from the nationally representative Household
Pulse Survey fielded by the U.S. Census Bureau throughout the pandemic. Pulse includes measures of
housing instability, food insufficiency and other material hardships, and mental health. We will analyze these
data using logistic regression models with state and year indicators which will allow us to estimate the causal
effect of ERAP policies on critical outcomes in renter households. Third, we will link the RRPD to weekly
counts of eviction filings at the Census tract or ZIP code level in 31 U.S. municipal areas. Employing
sophisticated techniques—multilevel zero-inflated negative binomial models—we will evaluate the effects of
state and local ERAP policies on the rate of eviction filings. Results from this work will both increase
understanding of COVID-19 policy effectiveness and trends in inequities, and serve as a blueprint for additional
policy and public health initiatives necessary to stabilize increasingly precarious rental housing.