Expiring Eviction Moratoriums and COVID-19 Incidence and Mortality

Abstract

The coronavirus disease 2019 (COVID-19) pandemic and associated economic crisis have placed millions of US households at risk of eviction. Evictions may accelerate COVID-19 transmission by decreasing individuals' ability to socially distance. We leveraged variation in the expiration of eviction moratoriums in US states to test for associations between evictions and COVID-19 incidence and mortality. The study included 44 US states that instituted eviction moratoriums, followed from March 13 to September 3, 2020. We modeled associations using a difference-in-difference approach with an event-study specification. Negative binomial regression models of cases and deaths included fixed effects for state and week and controlled for time-varying indicators of testing, stay-at-home orders, school closures, and mask mandates. COVID-19 incidence and mortality increased steadily in states after eviction moratoriums expired, and expiration was associated with a doubling of COVID-19 incidence (incidence rate ratio = 2.1; 95% confidence interval (CI): 1.1, 3.9) and a 5-fold increase in COVID-19 mortality (mortality rate ratio = 5.4; CI: 3.1, 9.3) 16 weeks after moratoriums lapsed. These results imply an estimated 433,700 excess cases (CI: 365,200, 502,200) and 10,700 excess deaths (CI: 8,900, 12,500) nationally by September 3, 2020. The expiration of eviction moratoriums was associated with increased COVID-19 incidence and mortality, supporting the public-health rationale for eviction prevention to limit COVID-19 cases and deaths.

Keywords: COVID-19; eviction; housing; social policy.

Figure 1

Eviction moratoriums in US states, March 13 to September 3, 2020. A) Map of US states indicating eviction moratorium status over the study period; B) change in state eviction moratorium status over time. Data from the COVID-19 Eviction Moratoria and Housing Policy database (13).

Figure 2

Adjusted rate ratios measuring the time-varying associations between US eviction moratorium expiration and daily COVID-19 incidence (new cases per population) (A) and mortality (deaths per population) (B), 2020. Rate ratios were modeled using negative binomial regression with fixed effects for state and calendar week, adjusting for testing rate, stay-at-home orders, school closures, and mask mandates. Event study coefficients estimate effects only in states with expiring moratoriums. States that maintained their moratoriums are included in models to control for secular trends. Data from the COVID-19 Eviction Moratoria and Housing Policy database, the Johns Hopkins Center for Systems Science and Engineering COVID-19 time series, and the COVID Tracking Project (13–15).

Figure 3

National estimates of cumulative excess cases (A) and cumulative excess deaths (B) associated with lifting of state eviction moratoriums in the United States from March 13 to September 3, 2020.