The Volokh Conspiracy
Mostly law professors | Sometimes contrarian | Often libertarian | Always independent
New NBER Study Finds Covid Eviction Moratoria Increased Racial Discrimination
Moratoria caused landlords to be less willing to rent to black tenants.
A new National Bureau of Economic Research study by economists
We provide evidence of intensified discriminatory behavior by landlords in the rental housing market during the eviction moratoria instituted during the COVID-19 pandemic. Using data collected from an experiment that involved more than 25,000 inquiries of landlords in the 50 largest cities in the United States in the spring and summer of 2020, our analysis shows that the implementation of an eviction moratorium significantly disadvantaged African Americans in the housing search process. A housing search model explains this result, showing that discrimination is worsened when landlords cannot evict tenants for the duration of the eviction moratorium.
The authors are likely to revise the study before final publication. But their results should not be surprising. Eviction moratoria make it difficult or impossible for landlords to evict tenants who default on the rent. That, in turn, leads property owners to be more wary of renting to people who are disproportionately likely to default, such as poor people. If blacks are, on average, poorer than whites or more likely to default for other reasons, landlords will be more reluctant to rent to them at a time when they cannot resort to eviction to deal with default. And studies do in fact suggest black tenants are, on average, poorer than white ones, and more likely to carry rental debt.
The NBER result is also consistent with previous studies showing that eviction moratoria and other policies that make it harder to evict delinquent tenants increase the cost and reduce the availability of housing. They are also likely to screen potential tenants more carefully, keeping out those who seem unusually likely to end up in default. Thus, while eviction moratoria and other similar policies benefit current tenants, they reduce the availability of housing to future ones - including current tenants wishing to move to a different location.
This, doesn't necessarily prove that eviction moratoria are unjustified. If, for example, Covid-era moratoria saved many lives, the resulting reduction in the availability of housing might have been worth it. But there is no good evidence that any such thing happened.
Similarly, eviction moratoria enacted during economic downturns might still be worth it if they save large numbers of people from poverty and homelessness. But, once again, available evidence doesn't support that theory. When the Supreme Court abruptly terminated the federal Covid eviction moratorium in August 2021 (ruling that the CDC lacked the authority to enact it), the eviction "tsunami" predicted by defenders of the policy failed to materialize.
There is much that can be done to increase the availability of housing to low-income and minority tenants. Most importantly, it can reduce or eliminate exclusionary zoning, which has a long history of blocking housing construction in ways that disproportionately harm those very groups.
If government wants to provide low-income tenants with extra support during a recession or a pandemic in order to prevent eviction, it can give them temporary rent subsidies. That can help tenants make ends meet without incentivizing landlords to exit the market, raise rents, or discriminate low-income and minority tenants. But we should avoid policies - like eviction moratoria - that tend to harm many of the very people they seek to help.
The legal and policy questions here are distinct. I have argued that the federal CDC eviction moratorium was beyond the agency's power, and that eviction moratoria also violate the Takings Clause of the Fifth Amendment, and similar provisions of state constitutions. But even those who differ with me on these legal questions should consider whether eviction moratoria really are a good strategy for helping poor tenants.
Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of Reason.com or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Comments may only be edited within 5 minutes of posting. Report abuses.
Please
to post comments
No kidding?
And did nothing to stop the spread of Covid
Oh, it did plenty to not stop the spread of COVID.
Funny, I never heard of any jurisdiction putting a moratorium landlords property taxes when they weren't collecting rent and couldn't evict tenants.
Correlation is not the same as causation.
The post hints at a reason: lower income. Is the original analysis adjusted for income? If not, then the discrimination here is against the poor, not against blacks. Although they likely are disproportionately poor.
I have several clients in the multi family space. Certain subsets of demographics of the renters have much higher deliquency rates. Thus the reason the multifamily rental industry developed the names for the various sub groups, Golfers, soccer players or basketball players. The basketball player sub group has much higher rates of property damage and deliquencies. The nicknames for the demographic subgroups have been around 30+ years for a reason.
I saw this:
"And studies do in fact suggest black tenants are ... more likely to carry rental debt."
I can't imagine why not paying your last landlord might make someone reluctant to rent to you.
Of course the landlords are trying to filter for low income (and thus, for higher likelihood of default). That, however, is not information readily available to the average landlord.
Race is a very poor proxy for likelihood of default but it's still a positive correlation and thus better than no data at all. Which, yes, means that a perfectly rational landlord will discriminate in this situation.
Agree. But this study appears to have been written by wokists, who expect us all to assume a priori that anytime a black person suffers economic failure, it's Whitey's fault.
A similar effect has been seen with regards to hiring people with disabilities. If it becomes difficult or nearly impossible to end employment for cause, the number of disabled that are hired will decrease.
Very few leftists understand many basic concepts such as the supply and demand curves. If they did, they would not be leftists.
Increased disparate impact, NOT racial discrimination. It always bears mentioning that they aren't remotely the same thing, and that often the only way to avoid disparate impact IS to racially discriminate.
According to Disparate Impact Theory, that means eviction moratoriums were systemically racist and thus the DC bureaucrats who created the rule are sickening vile racists.
"Similarly, eviction moratoria enacted during economic downturns might still be worth it if they save large numbers of people from poverty and homelessness. "
What other things do you think justify violating property rights, Ilya?
Amen. In spite of the final paragraph, property rights are getting short shrift in the argument.
Often libertarian. Not always, though.
https://reason.com/video/2024/03/08/great-moments-in-unintended-consequences-road-noise-meters-san-francisco-red-state-boycott-and-pennsylvanias-political-cartoon-ban-vol-15/
Calling MEREDITH BRAGG, AUSTIN BRAGG, AND JOHN CARTER
Increasing the supply of rental housing by reducing risks, and increasing returns for landlords is crazy talk.
The results don’t necessarily bear the conclusions Professor Somin wishes to draw.
For example, treatments that prolong life without curing tend to increase the number of people (living) with a disease. Defining a condition in a broader way – for example, requiring clearer consent to avoid rape charges – increases the number of people with the condition.
And changes can provoke negative reactions. For example, having blacks serve in WWI tended to increase the number of lynchings in the years afterwards. Similarly, black attempts to vote tended to increase lynchings.
As all these examples show, the fact that a change is associated with some sort of negative outcome don’t necessarily prove that the change is bad.
I think you have misunderstood Somin's conclusion.
As I read it, the conclusion is not that any negative consequence makes the policy per se bad.
The conclusion is: The policy is bad because it both failed to achieve it's stated goal and it carried other negative consequences.
added more information plzz.
Same reason Martha's Vineyard "Evicted" all of those illegal aliens within hours of their arrival. Barry Hussein don't want them around, anyhow.
Frank
I buy the results here, but struggle to think of a metric to optimize the policy. Right now, different jurisdictions have different barriers to eviction. We might think some states have tenancy laws that are more pro-tenant and others more pro-landlord; some jurisdictions have local police who are willing to assist a physically contested eviction while others don't etc. If it's true that a higher barrier to eviction creates opportunities for housing discrimination in the initial provision of housing (makes sense), then it's also true that relatively higher barriers on the basis of other tenancy policies create the same incentive.
I tend to think of this kind of tradeoff as being fundamentally similar to minimum wage laws (the higher the minimum wage, some elasticity to the number of people hired), right-to-work/termination without cause laws (the easier to fire employees, the easier to hire risky employees) and other sorts of protections (the union tradeoff of improving the quality of union jobs while preventing non-members from getting still-good-but-less-good jobs). It seems to me like the global optimum is probably not setting these policies to 0 or 100, it's somewhere in the middle, but this requires a principled ground for trading off the relative benefits.
The post seems to engage with this in the sense of saying the tradeoff for the eviction moratorium was the deterrence to the spread of COVID. I can agree that this did not necessarily work (though of course the context of the moratorium at the beginning of COVID was uncertainty about what kinds of income support would exist in the wake of a collapse of private employment -- instead the government provided fairly robust income support for the unemployed, fairly robust support to businesses for retaining employees, and businesses went remote fairly quickly). And legally this may have been the rationale for the policy; in effect an emergency intervention was only justified by the emergency. But I suspect many of those who supported the moratorium would have supported additional protections against eviction during ordinary circumstances. Someone might advance the argument that given spiralling rent costs in the commuter rings of American metros, additional protection against eviction might be merited across the board (in addition to, yes, YIMBY, zoning death penalty, fewer barriers on internal migration, all the other things Prof. Somin writes about that I agree with). So I would tend to think that if constitutionality is not a concern, the evaluation of the eviction moratorium is not just based on the tradeoff between its deleterious effects on new rentals and the spread of COVID, but also its positive effects on tenant security and its deleterious effects on new rentals. (This does not lead to the conclusion the moratorium was justified, just that the analysis needs to proceed a little differently)
The other sort of elephant in the room here is that in the early stages of the pandemic, the housing market experienced another serious disruption: the near collapse of AirBNB listings. One presumes at least some of the short-term rental market would have, exogeneously, become re-listed as medium or long-term renting. Now these same landlords might want to boot the tenants in a year or so when the travel market recovered, but in the mean time it seems very unlikely they'd take the loss on vacancy rather than renting, even under the consideration that they could be exploited by renters. It's notoriously difficult to disentangle two treatments applied to the same units at the same time even given the plausible exogeneity of the treatment timing due to the natural experiment qualities of COVID.
So to me it's a bit of a bummer not to hear, in the blog post at least (I'll read the NBER WP shortly) a more holistic treatment of the subject... and as always a bummer that the comments are mostly the same 10 people sharing their bumper stickers with everyone.
Eviction moratoria was part of the establishment's "not letting a good crisis go to waste" policies. I don't believe the CDC ever went on record with a targeted reduction rate in covid due to this policy. Just more collateral damage.
There was no basis for it more complicated than "renters outnumber landlords, and vote". Really, it was as simple as that: The government giving away other people's money to buy votes.
You can explain easily 90% of what the government does by assuming that polices are some mix of buying votes, and laundering money.