Connecticut's Housing Crisis: Analyzing Eviction Disparities and Policy Impacts
In 2022, CTData and the Connecticut Fair Housing Center released a report which analyzed eviction court filings from 2017 through 2021 to better understand the eviction crisis in Connecticut. We found that:
Black and Hispanic/Latino renters have cases filed against them at the highest rates;
Eviction cases are disproportionally filed against females and even more so against Black and Hispanic/Latina females.
Renters without a lawyer were nearly twice as likely to have a removal order issued against them.
Nearly half of all eviction cases during that period were filed in just five cities: Hartford, Bridgeport, Waterbury, New Haven, and New Britain.
In this updated report, we share findings and analyses based on eviction filings data from 2017 to 2023, while exploring key policy changes in Connecticut, including changes to the UniteCT program and the introduction of a new right-to-counsel program.
Key Findings in this report:
In 2022 and 2023, eviction filings affected 1 in 20 rental households;
Black householders are overrepresented in eviction filings –accounting for 25% of the eviction filings, black householders only represent 15% of all renter households in Connecticut.
52% of Connecticut renters are female but 56% of eviction defendants are female -- meaning that for every 100 renters, there are 4 more female renters than male renters, but there are 12 additional eviction filings against female renters compared to male renters
Renters with counsel received a ruling in their favor in 45% of cases, versus 27% of renters without counsel.
9 of the 10 landlords with highest number of eviction filings were private companies, in 2022 and 2023
12% of renters were represented by a lawyer in court (2022-2023), up from 7% (2018-2019). This increase was largely due to representation increases in right-to-counsel ZIP codes.
Spotlight on Hartford
Members of the 2024 Hartford Youth Data Fellows analyzed eviction filings in Hartford from 2017 and 2024, and developed Data Snapshots to highlight their findings. Read them at the button below.
Connecticut’s Housing Landscape
Connecticut’s housing crisis is deeply embedded into a legacy of discriminatory practices and a tangled web of land use regulations. Fifty-two percent of Connecticut renters are cost-burdened, according to the 2022 American Community Survey 5-year estimates, and opportunities to rent are largely concentrated in urban areas with existing multifamily housing and where affordable housing units are often clustered. Limited rental housing drives up rent prices and puts more renters at risk of eviction. New housing units are needed to alleviate the high cost of housing, but with local control over zoning ordinances, towns limit types of new housing allowed, and much of Connecticut allows only single-family housing as-of-right.
In Connecticut, individual towns have control over local zoning ordinances, which has resulted in towns severely limiting the types of residences built within their boundaries. Development of new multifamily buildings is limited, and developers can face a time-consuming and often costly special permitting process with local land use authorities. The maps below were created by the National Zoning Atlas, which compiles local zoning regulations into one database. The map on the left shows areas where new single-family housing is permitted, the map in the center shows where new two-family housing is permitted, and the map on the right shows where new four-plus unit structures are permitted without requiring a special permit. Connecticut residents in areas that allow mostly single-family housing were found more likely to be white, to have high household incomes, and to own their homes.
Most of Connecticut does not allow for two or more housing units on one lot, resulting in renters having limited housing options. This map shows the percentage of households who rent in each ZIP code. Zip codes with a higher proportion of renters are concentrated in cities and towns that allow higher-density housing.
Guidelines from the Department of Housing and Urban Development state that no more than 30% of a household’s income should be spent on housing costs, and that households spending more than 30% of their income on housing are cost-burdened, while households spending more than 50% of their income on housing are severely cost-burdened. Renters tend to spend more of their income on housing costs than homeowners do.
According to the 2022 American Community Survey 5-year estimates, more than half of Connecticut renters were spending more than 30% of their income on rent: 25% of Connecticut renters were cost-burdened and 27% were severely cost-burdened. Homeowners were less likely to be cost-burdened.
In addition to nonpayment of rent, tenants can be evicted for a variety of reasons, including the expiration of a lease, breach in the terms of a lease, or being deemed a serious nuisance. When a housing provider files an eviction and sues a tenant for possession of their home, a record of that filing becomes public in the state’s judicial system. The presence of an eviction record can make it difficult to rent new units. Additionally, frequent moves have been linked to negative effects on school performance, mental health, and social connections. Evictions are destabilizing not only for the individual households facing eviction, but for entire neighborhoods, especially when most of the neighborhood’s housing supply is made up of rental units.
Overview of Eviction Filings
Landlords filed 118,923 eviction cases between 2017 and 2023. In the three years leading up to the Covid-19 pandemic, about 20,000 evictions were filed annually, with a drop in 2020 and 2021, followed by large increases in 2022 and 2023.
During the Covid-19 pandemic, state and national evictions moratoria were in place and emergency rental assistance was available through the UniteCT program, which distributed federal Covid-19 relief funds. The statewide evictions moratorium expired on June 30, 2021 but was followed by a modified evictions process, which was set to expire on February 15, 2022, then was extended through June 30, 2022. UniteCT continues to provide assistance through its Eviction Prevention Fund.
During the first year of the Covid-19 pandemic, between April and December 2020, 2,048 evictions were filed, about 86% lower than the 14,535 filed between April and December 2019. In 2021, filings remained lower, about half as many evictions were filed as were in 2019. Eviction filings spiked in March 2022 and have since remained higher than pre-pandemic levels. In 2022 and 2023, eviction filings affected 1 in 20 households who rent in Connecticut, according to 2022 American Community Survey (ACS) 1-year estimates.
Black renters face higher rates of eviction filings
When comparing race and ethnicity in the eviction filings data to householders in 2022 ACS data, we found that Black householders accounted for 25% of the eviction filings while representing only 15% of all renter households in Connecticut. White renters were also slightly overrepresented, accounting for 42% of eviction filings compared to 40% of renter households. Since 2017, each race/ethnicity group’s share of eviction filings has remained steady.
More than half of evictions are filed against female renters
From 2017 to 2023, 56% of eviction filings were against female renters. This proportion remained consistent across all years of data. Overall, 52% of Connecticut renters are female, meaning that for every 100 renters, there are 4 more female renters than male renters, but there are 12 additional eviction filings against female renters compared to male renters.
Eviction filings are concentrated in Connecticut’s largest towns
As discussed in the introduction, rental housing units in Connecticut are concentrated in part due to individual towns controlling zoning laws that limit new multifamily construction. Additionally, discriminatory housing and land use policies continue to deepen the concentration of rental opportunities, especially affordable rental housing, in urban centers.
This concentration results in clusters of eviction filings in Connecticut’s largest towns and areas with high proportions of renters. Hartford (16,129), Bridgeport (12,243), Waterbury (10,596), and New Haven (10,006) accounted for about 2 of every 5 statewide eviction filings between 2017 and 2023. ZIP codes with high number of eviction filings were also in New Britain, Stamford, West Haven, Bristol, and Norwich. These ZIP codes face consistent challenges community-wide due to the destabilizing effects of evictions.
Counsel and court appearances shift outcomes
The eviction process is a complex legal process that varies from state to state. The multiple steps required to access civil justice are one of the reasons navigating this process with an attorney is beneficial to a tenant who will likely struggle to navigate complex court systems.
In Connecticut, the eviction process begins with a Notice to Quit (NTQ) given to the tenant by a landlord. The Notice to Quit explains that the tenant needs to "quit possession" of the unit. NTQs are not filed with the court until a formal eviction is filed. It is understood by many housing experts that many tenants leave their units when they receive an NTQ and there is no record within court filings of these evictions.
Only after a landlord flies a summons and complaint with the court does there become a record of the eviction. Notices to Quit are not captured in this analysis. After the date on a Notice to Quit, a landlord can have a marshal serve a court summons and complaint to the tenant, which informs the tenant that the landlord has filed an eviction lawsuit against them. These filed cases are captured in this analysis. The tenant then files an appearance and an answer and prepares for trial.
In this analysis, we classify judgments as either (1) in favor of the landlord or (2) in favor of the tenant. This determination is based on Connecticut Fair Housing’s assignment of outcomes according to the dispositions identified in the data. Cases not considered judgments for the landlord are classified as in favor of the tenant. It is important to note that even if a judgement is in favor of the landlord, tenants may have secured some favorable terms, such as more time to move.
Renters with counsel are more likely to have a decision in their favor than those without counsel. Between 2017 and 2023, 59% of renters with counsel received a ruling in their favor when the landlord did not have counsel, and 44% of renters with counsel received a ruling in their favor when the landlord also had counsel. On the other hand, when both the renter and the landlord did not have counsel, 27% of rulings were in the renters’ favor. When neither party had counsel, 30% of decisions were in favor of the renter.
When disaggregating disposition by race and ethnicity, we found similar patterns across all race and ethnicity groups: before 2020, 19-27% of cases ended in favor of the renter. This increased to 40-47% in 2020 and 2021 and has since decreased back to 25-36%. Asian and Pacific Islander renters have been at the higher end of these ranges with a higher percentage of outcomes in favor of the renter, while Black tenants have been at the lower end, with more outcomes in favor of the landlord.
Right to Counsel increases percentage of represented tenants
Connecticut passed a right-to-counsel law that went into effect, which gives renters who meet certain criteria the right to a lawyer for eviction proceedings. In the first year with the right-to-counsel law in effect, it applied to renters in 15 ZIP codes.
Overall in 2022-2023, 12% of renters were represented by a lawyer in court, up from 7% in 2018-2019. Increases in representation were seen across all race groups during this period. This increase was largely due to representation increases in right-to-counsel ZIP codes. In 2018-2019, both right-to-counsel ZIP codes and non-right-to-counsel ZIP codes saw 7% of renters represented. In 2022-2023, 18% of renters in right-to-counsel zip codes were represented, and in other ZIP codes, the percentage of renters with representation was just 8%. Representation in court is a key piece in ensuring that tenants are provided access to due process.
The percentage of landlords with counsel has also been increasing. In 2023, 87% of landlords were represented by a lawyer, compared to 79% in 2017. Landlords were 8 times more likely to have a lawyer than renters were in 2023.
In right-to-counsel ZIP codes, there was not an increase in rulings in favor of the renter. Decisions in favor of the renter in right-to-counsel ZIP codes was 23% in 2018-2019 and 30% in 2022-2023, which is nearly identical to other ZIP codes’ 23% in 2018-2019 and 31% in 2022-2023. Counsel may help a renter secure more time to move out or secure other beneficial terms, even if the decision was not made in the renter’s favor by this analysis’ definition.
Post-moratorium, private landlords have filed more evictions than public housing authorities
In 2022 and 2023, 9 of the 10 landlords with highest number of eviction filings were private companies, compared to 7 of 10 in 2018-2019. One explanation for this increase in private landlords filing more evictions than public housing authorities is that private landlords can evict tenants for more reasons, including at the end of a lease, called “no-cause” evictions. According to analysis from the Connecticut Fair Housing Center, there was a sharp increase in the number of no-cause eviction filings in August-December 2021, which may have continued into 2022.
With data from the National Housing Preservation Database (NHPD), we were able to match property owners in the NHPD data with property owners in eviction filing records. This dataset includes public property owners, including housing authorities, and private property owners that receive public subsidies. Non-public entities receiving active subsidies in this dataset that matched to the eviction filing records collectively owned over 45,000 housing units and filed over 4,000 evictions between 2017 and 2023.
While eviction protections put into place during the Covid-19 pandemic worked to lower eviction filings, filings quickly returned to pre-pandemic levels or higher when protections were removed. Private landlords and private landlords who receive public subsidies filed more evictions in 2022 and 2023, respectively, than in 2017, 2018, or 2019. Public housing authorities filed similar numbers of evictions in 2023 compared to pre-pandemic years.
Change in eviction filings before and after evictions moratorium
In the 189 ZIP codes with more than 10 eviction filings in 2018-2019, most (129) either increased or decreased by 25% or less between 2018-2019 and 2022-2023. In 15% of ZIP codes (28), there was an increase of 25% or more in 2022-2023. In five of these ZIP codes, which were in Hartford, Rocky Hill, East Windsor, Farmington, and New Canaan, eviction filings increased by more than 50%. Thirty-four ZIP codes (18%) experienced a decrease in eviction filings of 25% or more.
Across the 15 right-to-counsel ZIP codes, eviction filings increased by 14%, and in one ZIP code, 06092 in Stamford, eviction filings increased by 40%. Across all other ZIP codes, eviction filings increased by 10%.
Notes about the Data
Thank you to the staff of the CT Fair Housing Center who helped make these data available.
Incomplete Totals
If the data is disaggregated, totals may not add up. For example, renters whose sex cannot be determined are included in the overall total, but not for analyses where we disaggregate by sex. We were able to assign sex probabilities for 91% and race/ethnicity for 86% of the entire data set.
Race/Ethnicity of the Principal Defendant
By applying a Bayesian prediction model that uses a person’s surname and zip code, a method developed by Elliot et al., CTData assigned a predicted probability for race and Hispanic ethnicity of the renter named as a defendant in each eviction court case. This person is normally the head of household.
Male/Female Sex of the Principal Defendant
Because court records do not include sex data of renters named in the case, CTData derived probabilities of defendants belonging to both sexes given their first name based on historic Social Security annual baby names data set, which reports the prevalence of names given to males and females born between 1945 and 2004.