Eviction Lab updatesNovember 15, 2020

Preliminary Analysis: Eviction Filings During and After Local Eviction Moratoria

  • Peter Hepburn and Renee Louis
  • The Eviction Lab

In response to the COVID-19 pandemic, actors at multiple levels of government enacted policies meant to protect renters from the threat of eviction. The federal government issued an eviction moratorium on federally assisted units and properties with federal mortgages through the CARES Act, as well as a more far-reaching moratorium in private rental housing units ordered by the Centers for Disease Control and Prevention (CDC).

At the same time, many cities, counties, and states across the country adopted moratoria, some of which exceeded the protections offered by the federal government by freezing multiple stages of the eviction process. The eviction process usually involves five stages:

  1. Landlord provides notice to the tenant that they intend to evict them
  2. Landlord files an eviction case in court
  3. The court holds a hearing
  4. The judge issues a writ of possession
  5. The sheriff executes the writ of possession and removes the tenant.

State and local moratoria varied in which tenants were protected, what types of eviction were forestalled, what steps of the eviction process were halted, under what conditions, and for how long. In some cities, courts held hearings and even handed down eviction orders, but did not allow sheriffs to enforce the order of eviction. The most protective moratoria suspended all stages of eviction, followed by those that stopped the initiation of the eviction process (i.e. notice and filing).

Early analyses of eviction moratoria have focused mostly on the federal efforts.1 There is evidence that the CARES Act eviction moratorium reduced eviction filings from covered properties. When CARES Act protections expired, new filings spiked in the weeks before enactment of the CDC moratorium. In this brief, we examine effects of state and local moratoria. Specifically, we ask two questions. First, how far below historical average were eviction filings when these moratoria were in place? Second, what happened when they were lifted?

To answer these questions, we draw on data from the Eviction Tracking System (ETS). The ETS includes eviction filing data from 25 cities across the country, both in 2020 and for a range of previous years.2 All of these cities were covered by some form of state- or local-level eviction moratorium during the onset of the COVID-19 pandemic. In many sites, moratoria restricted notice to tenants or filing with the courts. In others, landlords were allowed to proceed with notice and filing, but moratoria suspended the hearing stage of eviction. In one city—Phoenix—hearings were allowed to proceed and writs of possession were issued by the court, but sheriff’s enforcement was halted. At the start of November, the vast majority of ETS cities (22 out of 25) no longer offered renters any protections from eviction.

Cities saw an increase in eviction filings after moratoria were lifted, and the size of that increase depended on what stage of the eviction process was suspended.

When state and local moratoria were in place, eviction filings were well below historical average. In Figure 1 we compare eviction filings during the effective dates of moratoria to eviction filings for the same dates and locations in previous years. We split this figure into four panels, grouping cities together according to the earliest stage of the eviction process that was suspended under the most restrictive applicable state, county, or city moratorium.3

In many cities, state or local moratoria cut eviction filings to zero (or near-zero). This is especially true of the cities that suspended the first and second stages of the eviction process—notice and filings. In these cities, new eviction filings were at or below 21% of historical average while their moratoria were in place. In Cleveland, for instance, the courts only accepted eviction filings under exceptional circumstances when their local moratorium was in place. They handled 148 new eviction filings between March 16th and June 15th, compared to 1,192 on average during this period between 2012 and 2016. Due to Connecticut’s state-wide eviction moratorium, which froze notice and filing on April 10th and will remain in place until at least the end of 2020, neither Bridgeport nor Hartford have seen more than 9% of typical filings since April 23rd.

Figure 1.1. Eviction filings during the local moratorium period, relative to historical average for the same period.
Fig 1.1 Created with Sketch. 0% 25% 50% Tampa, FL Gainesville, FL Jacksonville, FL Greenville, SC Cleveland, OH Charleston, SC South Bend, IN Austin, TX Bridgeport, CT Hartford, CT Wilmington, DE Philadelphia, PA Pittsburgh, PA Indianapolis, IN Boston, MA Milwaukee, WI Notice Filings

By contrast, cities that allowed filings but suspended hearings were less successful in reducing new eviction filings. On average across the eight cities that fell in this category, new filings were at 25% of historical average. In Phoenix, which allowed eviction filings and hearings but froze enforcement of eviction orders, total filings rose to 37.6% of average.

Figure 1.2. Eviction filings during the local moratorium period, relative to historical average for the same period.
Fig 1.2 Created with Sketch. 0% 25% 50% Phoenix, AZ Cincinnati, OH Columbus, OH Memphis, TN Kansas City, MO St Louis, MO Richmond, VA Houston, TX Fort Worth, TX Hearings Enforcement

As these moratoria lift, new filings have increased, in some cases dramatically. In Figure 2 we track total new eviction filings relative to historical average in the six weeks before and after expiration of the state or local moratorium.4 In the figure:

  • We split cities depending on the type of moratorium in place. We drop the “Enforcement Only” category because we do not yet have enough data from Phoenix about what happened following the expiration of Arizona’s state-wide moratorium on October 31.
  • The bold red line represents the weighted average across the included cities in that category.
  • The lighter grey lines represent each of the cities, revealing that some cities were more affected by lifting moratoria than others.
Figure 2. Eviction filings relative to historical average in the six weeks prior to and following the end of the local moratorium.
Fig 2 -100% -50% 0% 50% 100% Filings -100% -50% 0 % 50% 100% Notice -100% -50% 0% 50% 100% -6 -3 0 3 6 Week Relative To End Of Moratorium Percent Change Relative to Average Hearings

The top panel includes ETS sites in which notice was suspended, the middle panel sites in which filings were suspended, and the bottom panel includes sites in which hearings were suspended.

Figure 2 yields two important findings. First, cities saw an increase in eviction filings after moratoria were lifted, and the size of that increase depended on what stage of the eviction process was suspended. In the top two panels of Figure 2, we see that cities in which notice or filings were suspended saw more dramatic jumps in filings when local moratoria were lifted. On average, in cities in which notice was suspended, new filings jumped from 4.7% of historical average two weeks before the end of a moratorium to 64.7% two weeks after. By contrast, in the bottom panel we see that new filings increased much less in cities that had only suspended hearings.

Second, filings typically remained well below historical averages in the six weeks following the lifting of local eviction moratoria. On average, six weeks after a moratorium lifts, new eviction filings were still only 46.1% of the historical average.

This brief contributes to a growing body evidence demonstrating the efficacy of eviction moratoria in preventing forced displacement of renters amidst the pandemic. When state and local moratoria were in place, eviction filings were well below half of historical average in all ETS sites. This is true especially of those sites that suspended the earliest stages of the eviction process: notice and filing of eviction cases. Under normal circumstances, many tenants facing the threat of eviction never make it to court, leaving their homes when their landlord provides notice or when they receive the court filing. As such, suspending these early stages of the eviction process is particularly important minimizing displacement.

State and local moratoria were not solely responsible for the reduction in eviction filings: federal moratoria, stimulus payments, unemployment insurance expansion, and uncertainty in the rental housing market also served to reduce filings. Even when these local measures were lifted, eviction filings generally remained below historical average. But they did increase, which speaks to the critical importance of moratoria in reducing eviction and supporting residential stability. Our analysis suggests that moratoria work, and they work best when they halt the first stages of the eviction process.


  1. Work by Rebecca Cowin, Hal Martin, and Clare Stevens at the Federal Reserve Bank of Cleveland offers an important exception.
  2. ETS sites represent a convenience sample of cities in which both current and historical eviction filing microdata are available. Discussion of site selection and ETS methods can be found here. ETS sites are not representative of all U.S. cities.
  3. If a city was covered by both a state and a local moratorium, we categorize it according to the policy that suspended the earlier stage of the eviction process. Austin, TX, for instance, was initially covered by the same state-wide moratorium as Houston and Fort Worth, suspending hearings. Travis County then instituted a more-restrictive policy suspending notice to tenants. Accordingly, we categorize Austin in the “Notice Suspended” group. We do not evaluate the efficacy of state moratoria relative to local moratoria.
  4. So as to have sufficient observations after a given moratorium ended, we include those sites in which moratoria lapsed before October 1. There are 20 ETS sites that meet this criteria. The omitted sites are Phoenix, Austin, Bridgeport, Hartford, and Boston.
Filed under research
rendered by = themes/THEMENAME/layouts/updates/single.html