What Should Policymakers and Advocates Know about Eviction?

by Maya Brennan

After years of sparse research and policy attention, evictions are a constant presence in the news and a growing topic of research. Advocates for stable housing can find locally or nationally produced eviction estimates, and research documenting eviction’s causes, consequences, and disparate impacts. Soon, the Eviction Lab at Princeton University will add more data, maps, and resources.

The rapid expansion of information has sparked efforts at policy change. State and local governments are debating and adopting new landlord-tenant laws and pilot programs, such as expanded legal representation and just-cause eviction requirements. Yet, few housing experts understand evictions well enough to channel the demand for change into clarity about specific eviction problems and potential solutions. Before Matthew Desmond’s work, the most notable article on evictions was  “Evictions: The Hidden Housing Problem,” published in Housing Policy Debate in 2003. Compared with the field’s understanding of rising housing costs and affordability challenges, knowledge about evictions remains minimal.

Now is the time for policymakers and advocates to get smart quickly. But how?

When addressing evictions, a few key principles can help.

Dig for definitions and assumptions.

Unlike many housing terms, the word eviction lacks a standardized definition. The range of what counts as an eviction is more varied than simply formal (i.e., court ordered) versus informal (i.e., by the landlord, not the court). Some eviction counts miss the most devastating unplanned moves—illegal lockouts. Other eviction counts include cases in which no renters were removed from the property.

Data and assumptions should clarify whether the eviction estimate includes

  • every unwanted move (including court-ordered removals, illegal lockouts, nonrenewed leases, and displacement because of building renovation),
  • all court actions that can lead to a tenant’s removal (including filings that went nowhere),
  • all court-authorized removals of people and possessions, and
  • other possible variations.

Talking about preventing eviction without clearly defining the term impedes the policy debate and may lead to adopting solutions that were designed to address a different problem.

The Milwaukee Area Renters Study questionnaire, which the US Census Bureau adopted for use in the 2017 American Housing Survey, defines eviction as “when your landlord forces you to move when you don’t want to” or “when you move out because a landlord makes you.” Apartment List’s 2017 renter survey left the meaning of eviction undefined. Redfin’s 2016 study counted eviction records from court filings or judgments, often known as “formal evictions.”

None of the above-mentioned counts can be easily compared with each other, and each way of defining the issue may call for different policy responses.

Don’t assume similarities from place to place.

The localized nature of landlord-tenant law means that states, counties, and independent jurisdictions have different eviction processes and legal defenses for withholding rent. These variations affect the most viable policy solutions, and they may have deeper implications for comparing eviction risks and rates from place to place.

For example, withholding rent and waiting for the landlord to sue can be a valid way to force a landlord to make repairs in some states, but it is not permissible in others.

In states with a right to redemption, a renter can pay any past-due rent and stay in the property—regardless of whether the landlord had already filed for eviction. The right of redemption could extend a few days after filing, up until the court date, or beyond. Other states do not grant any such right. In those areas, if the rent was later than the end of the grace period, the renter may not have any recourse.

Build policies to achieve a specific goal and consider the ripple effects.

Whether defined as any time a landlord forces a tenant to move or something narrower, evictions add uncertainty and instability for renters. But unless an eviction is the only way to raise the rent or sell the property for a more profitable use, landlords often face financial losses when proceeding with evictions. Smart prevention policies may attract unlikely allies, but protections that focus on just one area can inadvertently cause harm in another. Tread carefully at first, even when the need for change is urgent. If the evidence is not clear about a policy’s likely effects, build evaluation into the process.

Eviction is part of a cycle of housing vulnerabilities and job loss. Legal representation can reduce the number of evictions, but renters also need an effective safety net to weather the unexpected financial crises that give landlords cause to evict. Beyond that, renters need more affordable housing options to reduce the predictable eviction risk of severe housing cost burdens.

Because millions of households—disproportionately black female–headed households with children—have been touched by either a court eviction filing or a forced move, another policy goal could be to make sure eviction does not force renters into housing that is unfit for human habitation but preferable to homelessness. Policies that address this vulnerability and enable people to rebuild a positive rental history while addressing landlords’ legitimate concerns about future eviction risks would be groundbreaking innovations.

Adopting policy recommendations to address eviction without thinking through the goals could be worse than doing nothing. They can satisfy a community’s hunger for policy change but leave the root causes and consequences unaddressed.


From 2001 to 2003, Maya Brennan was a tenant-landlord hotline counselor serving 23 Maryland counties and Baltimore City. Most of the calls related to eviction were for nonpayment of rent.