In March and April, the fear and urgency surrounding the spread of Covid-19 spurred policymakers in nearly every state to institute eviction moratoriums — prohibitions on kicking people out of their homes for not paying rent.
Over the following months, more than half of these states allowed the protections to expire. Now, a new study says states that let those protections lapse saw an estimated 433,700 excess individuals contract Covid-19, and 10,700 people die from the virus.
Kathryn Leifheit, a postdoctoral researcher at the UCLA Fielding School of Public Health, and her research team found an association between the increased number of Covid-19 cases and deaths and states lifting eviction moratoriums. She and her fellow researchers looked at the 43 states (and the District of Columbia) that instituted a moratorium between March 13 and April 30. Then they looked at what happened in states where the state government lifted moratoriums, compared to those that imposed them and left them in place.
Research that attempts to isolate the effects of a single policy change is incredibly difficult — especially in a crisis like the pandemic, where dozens of policy levers are moving all at once at the federal, state, and local levels. States that lift their eviction moratoriums could be different from states that don’t and you could end up measuring something completely different, like whether a state had a mask mandate or the state’s political affiliation. Leifheit and her team worked to overcome some of these challenges by conducting sensitivity tests and by controlling for as many variables as possible.
The preprint, a research paper that is not yet peer-reviewed, was released on November 30, a month before the Centers for Disease Control and Prevention’s (CDC) national eviction moratorium is set to lapse on New Year’s Eve. It’s an early attempt to quantify the human toll of allowing families to lose their homes as the pandemic rages on. Several states also have moratoriums set to expire over the winter, if their leaders choose to let them.
The situation is dire. A potential wave of evictions is looming, and as many as 40 million Americans could suffer in the coming months. Recent reports indicate that Congress may include a federal eviction moratorium in a stimulus package before the end of the year, but if it doesn’t, the question of how evictions and the pandemic interact could become more pressing than ever.
What the study found
Researchers had previously found an association between evictions and sexually transmitted infections (more on that later) so Leifheit and her research team were looking to see if they could find a similar association between evictions and Covid-19 transmission.
“The size of the association and the number of cases and deaths that we end up attributing to these moratoriums lifting was surprising to us,” Leifheit told me.
Their model found that seven weeks after lifting their moratoriums, states saw 1.6 times as many deaths as ones that left their eviction moratoriums in place. Ten weeks after lifting their moratoriums, states saw 1.6 times as many cases. And 16 weeks after, states saw 2.1 times as many cases and 5.4 times as many deaths.
Leifheit explained the stronger effect they observed on deaths by pointing out the characteristics of people who most often face evictions: “The way that we explain the stronger effect on mortality than incidence is evictions disproportionately affect disadvantaged communities ... often people that have comorbidities and health problems prior to experiencing eviction.”
Both effects also grow larger over time — which makes sense if you think about what might happen to you if you got evicted. Right after, maybe you couch surf for a while, bouncing between a few friends and family members. You and your family might have to separate, some staying with friends and others forced to find shelter in their cars or in homeless shelters. If you’re lucky, maybe you find a permanent solution and you’re all able to come back together. Every step of the way after you are evicted, you only come into contact with more people, thereby increasing your chance of falling ill.
There are limitations to this research, which Leifheit and her co-authors are upfront about.
First, they aren’t able to track evictions because of the lack of comprehensive data, so they’re actually studying the lifting of eviction moratoriums on increased Covid-19 spread, not actual evictions. The Eviction Lab looked into whether eviction moratoriums affected eviction filings and found that while eviction moratoriums remained in place, eviction filings in 25 cities stayed well below historical averages — and when they lifted, eviction filings increased in the weeks following. So lifting a moratorium probably did lead to more evictions.
The bigger issue is that trying to isolate the effect of eviction moratoriums is extremely difficult. A state that allowed an eviction moratorium to expire could also have loosened up its public health response to Covid-19 in general, lifting restrictions on indoor dining, mask mandates, and other interventions at around the same time. For instance, Louisiana’s eviction moratorium expired on June 15, just 11 days after Gov. John Bel Edwards moved his state into phase two of reopening, “allowing churches, places of worship and many more businesses to operate at 50 percent capacity.”
The researchers worked to control for a wide range of these variables, including the number of Covid-19 tests done, the number of weeks since masks were mandated, and other underlying time trends and characteristics. But it’s near impossible to control for everything.
That said, the findings could also be undercounting the increased cases and deaths associated with lifting state moratoriums. Leifheit told Vox that the research doesn’t model spillover effects (if your state borders another state that lifted its moratorium, that probably has an effect on your Covid-19 transmission rates). The researchers were forced to rely on public health surveillance data, which requires that people actually go and get a test, something many people never do.
They also ignored local moratoriums: If Colorado lifted its moratorium, for example, but Denver still had one in place, these findings don’t account for the 700,000-plus people who are still protected from evictions.
But one thing that gave the researchers confidence in their study is that before states lifted their eviction moratoriums, the states that would go on to do so were seeing roughly the same trends in incidence and mortality as those that maintained their moratoriums. And that provides some reassurance, according to Leifheit, that the increases they observed were tied to the lifting of moratoriums rather than a preexisting trend.
The problems with studying evictions
Matthew Desmond, a professor of sociology who is perhaps best known for his 2016 Pulitzer Prize-winning book Evicted, highlighted the difficulties with studying evictions in part by pointing out that there was no nationally available data on evictions until he launched the Eviction Lab, a research group that tracks evictions and eviction-related data, at Princeton University.
In 2018, Desmond’s team found that roughly one in 40 renter households were evicted between 2000 and 2016, or “almost a million evictions against tenants every single year.” To put that in context, “at the peak of the financial crisis in 2010, estimates suggest slightly over one million foreclosures were completed nationally.” That means the rental market was seeing the rough equivalent of the 2010 housing market catastrophe every single year.
Recently, researchers have begun looking into how housing displacement and evictions relate to public health outcomes and have (unsurprisingly) found an association between the two.
In a 2019 study published in the journal Sexually Transmitted Diseases, researchers looked at the impact of eviction on rates of sexually transmitted diseases. They found higher rates of chlamydia in counties with higher rates of eviction, concluding that “county-level eviction rates are associated with chlamydia and gonorrhea rates in a significant and robust way independent of other known predictors of STI.”
In another 2018 study published in the medical journal AIDS and Behavior, researchers looked at the relationship between evictions and HIV transmission and found that evictions likely made it harder for people to adhere to the pharmaceutical drug schedule needed to keep HIV viral load low. The research team concluded that “eviction independently increased the odds of detectable” HIV viral load among study participants who experienced eviction.
There’s more research on housing displacement and public health outcomes that I won’t get into, but suffice it to say that there are existing findings linking evictions to worsening health outcomes.
Leifheit and her fellow researchers weren’t the first to study the association between evictions and Covid-19 transmission. In another study submitted to the Journal of Urban Health, law professor Emily Benfer and her co-authors write that “eviction is likely to increase COVID-19 infection rates because it results in overcrowded living environments, doubling up, transiency, limited access to healthcare, and a decreased ability to comply with pandemic mitigation strategies (e.g. social distancing, self-quarantine, and hygiene practices).” Benfer and her team cite research showing that “small differences in housing have been linked to substantial increases in transmission rates” of other infectious diseases and that “residential crowding and increased contact with others drive the spread of infectious respiratory diseases.”
And earlier this year, a preprint by researchers at Harvard University, the University of Pennsylvania, the University of Illinois at Urbana-Champaign, and Johns Hopkins University modeled the effect of evictions on the spread of Covid-19. They found evidence that evictions yielded a “significant increase in infections.”
The researchers applied their model to Philadelphia, a city with relatively high eviction rates, and found that evictions increase everyone’s risk of contracting Covid-19, not just those experiencing eviction. Importantly for policymakers, the worst effects are when evictions begin again and the epidemic is already rapidly spreading, like what would happen if the CDC’s eviction moratorium were to lift at the end of the year as cases continue to skyrocket.
While there’s reason to be cautious about the figures put out by Leifheit’s study, she and her fellow researchers’ work, as well as the existing body of research showing a positive association between evictions and decreased public health outcomes, is a clear warning to policymakers about the dangers of allowing evictions to continue as the coronavirus infects hundreds of thousands of Americans every day.
There’s an incredibly simple solution to this problem
Lots of issues surrounding the Covid-19 pandemic are difficult to solve. But stopping evictions is easy: Congress could decide to give people money, such as through more stimulus payments, and keep the eviction moratorium in place until there is widespread vaccine uptake.
“Eviction moratoriums are an incomplete solution,” said Leifheit. “At the same time that we’re talking about eviction moratoriums, we need to be talking about rent relief.”
There are real trade-offs to enacting an eviction moratorium: Small landlords without access to credit are the least capable of absorbing the loss of rent and most likely to house renters who have taken financial losses during the pandemic.
Most affordable housing in America is owned by individual investors, or “mom-and-pop” landlords. This can be surprising because many people think of rental property owners as being part of giant corporations, but 41 percent of the nation’s 48.2 million rental housing units (not just affordable housing units) are owned by these mom-and-pop property owners.
Research from the Urban Institute shows that the average rent in small rental properties is less than “the median for single-family rentals, medium-size apartment buildings, and large apartment buildings.” And in 2018, “the median income for a two-to-four-unit landlord was $67,000.” Renters of these units are predominantly Black and Hispanic, and they have the lowest median household income when compared to renters of other types of properties.
This means that renters of small properties tend to be lower-income and work in industries that were most harmed by Covid-19 — food service, retail, and construction, to name a few. On top of that, the landlords they rent from are some of the least capable of absorbing the loss of income from unpaid rent. Many of these landlords have mortgages of their own, and all of them are required to maintain their properties — which means operating costs continue even as rent payments decline.
Losing more of America’s already dwindling affordable housing stock is a looming emergency that could be exacerbated if small landlords are required to act as the social welfare state without any financial assistance.
One silver lining of Leifheit’s research is that states can independently have an impact in preventing evictions and potentially reducing cases and deaths by extending their own moratoriums.
Leifheit recommended that stronger eviction regulations, ones that prohibit landlords from even beginning the process of eviction, are likely to have a greater impact on reducing Covid-19 cases and deaths. This argument is backed up by the Eviction Lab’s work on this question as well.
But ultimately, this is a problem of the government’s making. The federal government could provide rental assistance that would allow renters to make payments and remove the burden on landlords and keep America’s affordable housing stock stable.
Mary Cunningham, vice president of metropolitan housing and communities policy at the Urban Institute, told Vox, “The CDC’s eviction moratorium saved millions of people from being evicted. Now it’s Congress’s job to actually pass a rental relief bill that really addresses the need.”
Correction, December 11: A previous version of this article referred to the study by law professor Emily Benfer and her co-authors as a preprint. The article has been peer-reviewed and is forthcoming in the Journal of Urban Health.