The enactment of the warranty of habitability in the early 1970s was hailed as a revolution in tenants’ rights. Reversing centuries of legal precedent, the doctrine established that a tenant’s obligation to pay rent is contingent upon the landlord’s obligation to maintain the premises in good repair. Today, nearly fifty years later, scholars and advocates frequently observe that the law has not lived up to the potential originally envisioned. Yet these observations have been based on weak empirical evidence. This Article presents the results of the first rigorous empirical study on the effectiveness of the warranty of habitability. Based on statistical analysis of over twelve hundred eviction case files and unit-level data matching of these files to Housing Code enforcement records, the study finds that the overwhelming majority of tenants with meritorious warranty of habitability claims do not benefit from the law at all.

The Article makes two significant contributions to the literature on the warranty of habitability. First, it establishes that an operationalization gap exists in the law. While prior studies have observed that the warranty appears to be less effective than originally envisioned, all suffered from methodological limitations. These studies were either based on small, nonrepresentative samples or measured the use of the warranty against the entire population of tenants facing nonpayment of rent eviction. No study has been able to rigorously assess the use of the warranty of habitability in cases where it should be used: those in which the tenant has a meritorious claim. This study does so.

Second, the Article upends the leading theories for why the warranty of habitability is ineffective. These theories posit that tenants are unable to benefit from the warranty of habitability because they lack access to legal representation and/or because strict requirements exist for assertion of the claim. The findings of this study show that neither theory withstands empirical scrutiny. Specifically, the data reveal that although legal representation significantly affects a tenant’s likelihood of benefiting from the warranty of habitability, most represented tenants with meritorious claims still do not benefit from the law at all. The findings also demonstrate that the strict procedural requirements cannot explain the law’s ineffectiveness—even where the requirements are absent, the law rarely protects tenants.

Introduction

Ms. J’s apartment in the South Bronx had become truly unlivable. The bathroom ceiling had collapsed, the walls were covered in mold, and the entire place was infested with mice.1 There were leaks in the bedroom and bathroom that had become so severe that, on multiple occasions, water flooded not only Ms. J’s apartment, but also the hallways of the building and neighboring units.2 Ms. J had called the City to report the problems, and inspectors had cited the landlord for violations of the Housing Code, but still no repairs had been made.3 Eventually, Ms. J stopped paying rent, as was her legal right to do. Since the early 1970s, the warranty of habitability has established that a tenant’s obligation to pay rent is contingent upon the landlord’s obligation to maintain the premises in good repair.4 The law states that where a landlord fails to maintain the property, the tenant is entitled to a rent abatement—a reduction in the amount of rent owed.5 Rather than fix the conditions in Ms. J’s apartment, however, the landlord filed an eviction action against her for nonpayment of rent.6 The law contemplates this response and allows the warranty of habitability to be asserted as a defense and counterclaim to the eviction complaint.7

The two sides came into Housing Court in July 2016, and the judge ordered the landlord to correct the defective conditions.8 The order required the landlord to make the repairs on two specific dates in August.9 Yet Ms. J waited at home all day both days, and no one ever showed.10 The parties went back into court in early September, and the court again ordered the landlord to make the repairs—this time, a few weeks later.11 The landlord again did not comply.12 This series of events repeated itself six more times throughout the fall and winter of 2016, and even into the spring and summer of 2017.13 Each time, the court ordered the landlord to make the exact same repairs, and each time, the landlord ignored the order.14 Eventually, the case settled.15 The landlord still had not made any of the repairs, but Ms. J agreed to repay the full amount of the back rent.16 The letter of the law had proven meaningless. Despite spending over a year in court, Ms. J was unable to effectively invoke her right to a rent abatement, nor was she able to use the law to secure performance of the repairs. And Ms. J had an attorney.17

The warranty of habitability was hailed as a “revolution” in landlord-tenant law;18 it was expected to provide a “powerful new remed[y] with which the urban poor could compel landlords to maintain their buildings adequately.”19 Yet nearly fifty years after the warranty’s enactment, to what extent is Ms. J’s experience typical, and to what extent is it an outlier? This Article presents the results of the first large-scale empirical study rigorously assessing the extent to which there is a warranty of habitability operationalization gap—a gap between the number of tenants with meritorious claims20 and the number of tenants who receive some benefit from the claim. Determining that there is a large gap, the study explores the reasons underlying it through further empirical analysis. The results upend the leading theories on why the warranty of habitability is underenforced.

The study was conducted in the largest rental market in the country, New York City,21 looking specifically at nonpayment of rent eviction cases.22 Data was collected and analyzed to determine: (1) the overall rate of rent abatements in cases in which the tenant has a meritorious warranty of habitability claim; (2) whether and to what extent tenants with meritorious warranty claims receive other benefits from the claim, such as longer periods of time to repay rental arrears or avoidance of possessory judgments;23 (3) whether and to what extent the warranty functions as a tool within eviction proceedings to secure repairs; and (4) whether and to what extent legal representation affects a tenant’s ability to benefit from the warranty where he or she has a meritorious claim.

The study was conducted using two unique datasets of nonpayment of rent eviction cases from 2016. The first dataset is a statistically significant sample of all nonpayment of rent eviction cases in which the tenant appeared. The second dataset is a statistically significant sample of nonpayment of rent eviction cases in which the tenant appeared and there were open “hazardous” or “immediately hazardous” Housing Code violations at the unit at the time the case was filed.24 This dataset was constructed based on a unique unit-level matching of eviction case data with Housing Code violation data. In total, over twelve hundred nonpayment of rent eviction case files were collected, reviewed, and coded.25

The study found that very few tenants with meritorious warranty of habitability claims actually benefited from the law. Overall, less than 2 percent of tenants who had meritorious claims received rent abatements. Perhaps even more astonishing, only 7 percent of tenants whose landlords have been cited by the City for hazardous or immediately hazardous Housing Code violations—a subset of those who had meritorious claims—received abatements. The findings also rule out the possibility that tenants with meritorious claims are reaping other types of benefits from their claims. Tenants with meritorious claims are no more likely to avoid possessory judgments or to receive longer periods of time to repay arrears as compared with tenants without meritorious warranty claims.26 The study also found that although tenants are more likely to benefit from the warranty of habitability when they have legal representation, the lack of access to counsel does not sufficiently account for the operationalization gap. The significant majority—at least 70 percent—of tenants who were represented by counsel and had meritorious warranty of habitability claims still did not receive a rent abatement. Finally, the findings showed that while eviction proceedings are indeed functioning as a forum to order landlords to perform needed repairs, the forum lacks accountability. Specifically, in 72 percent of cases in which the landlord agreed to make repairs in a court-ordered settlement agreement and there was a subsequent settlement agreement in the case, the tenant reported that those repairs were still outstanding in a subsequent court appearance.

These findings make two broad sets of contributions to the scholarly literature on the warranty of habitability. First, the findings provide rigorous evidence of the existence of an operationalization gap in the warranty of habitability. While much research has pointed to problems with the warranty’s implementation, prior empirical studies have consistently taken one of two forms. One set of studies has examined the overall frequency with which tenants assert warranty of habitability claims in court or receive rent abatements, without distinguishing between tenants who do and do not have meritorious claims.27 A second set of studies has taken the form of nonrepresentative observational or case studies that have looked at outcomes among small groups of tenants with meritorious claims.28 This study is the first thus far to rigorously examine on a large, representative scale the extent to which tenants benefit from the warranty of habitability when they have meritorious claims. It is also the first study to assess the possibility that tenants use the warranty of habitability to obtain beneficial outcomes in their cases other than rent abatements. That is, prior studies have not examined whether tenants use their entitlement to a rent abatement as leverage to achieve other desired case outcomes, such as longer repayment periods or the avoidance of possessory judgments. This study does so.

Second, the findings of this study debunk the conventional wisdom on the reasons for the ineffectiveness of the warranty of habitability. Since the warranty’s initial enactment nearly fifty years ago, scholars have tried to explain why tenants have not appeared to benefit from the law to the extent originally envisioned. The existing scholarship reflects a general consensus around two explanations: (1) tenants lack access to counsel, and (2) there are onerous legal requirements for asserting a claim.29 Recent scholarship has also hypothesized that the warranty is underutilized in part because judges lack ready access to Housing Code violation records.30 The findings of this study upend all of these existing theories.

First, the study finds that legal representation accounts for only a small fraction of the overall operationalization gap. While many previous studies have analyzed whether tenants who have access to counsel are more likely to receive rent abatements or raise warranty of habitability claims in court, none has measured the impact of legal representation specifically where the tenant had a meritorious claim.31 This is the first study thus far to do so, and the finding shows that while representation matters, the vast majority of represented tenants who have meritorious claims still do not benefit from the warranty. Second, the study finds the existence of a large operationalization gap even though New York lacks any of the onerous legal requirements for assertion of a claim. Thus, while these requirements may impose meaningful barriers where they exist, the findings of this study demonstrate that they do not sufficiently explain the warranty’s lack of implementation. And finally, the findings also refute the theory that providing judges easy access to Housing Code violation records, without more, will serve as a meaningful solution to the warranty’s operationalization failures. Code enforcement records are readily available to Housing Court judges in New York City,32 yet the data show that judges rarely take advantage of the opportunity to access them.

The Article proceeds as follows. Part I delves into the history of the warranty of habitability and explains the policy goals that drove its widespread enactment in the 1970s. Part II reviews the existing theoretical and empirical scholarship on the law’s usage. Part III describes the objectives, data, and methodology of the study conducted. Part IV presents and analyzes the results. Part V describes the significance of these findings for our understanding of the warranty’s implementation and the reasons for its ineffectiveness. The Conclusion points to directions for future research.

I.  Evolution of the Warranty of Habitability

The implied warranty of habitability has a nearly fifty-year history. First articulated in 1970, the doctrine was adopted with the expectation that it would bring transformative change to the landlord-tenant relationship. Advocates and scholars believed that the law would level the playing field in eviction cases, compensate for ineffectual code enforcement systems, and serve as a strong deterrent mechanism against landlord property neglect. These expectations were widely shared by advocates, legislators, and jurists across the country.33 Following the warranty’s initial adoption in the District of Columbia, forty-nine states embraced it in an extraordinarily short period of time.34 This Part describes the social, political, and legal concerns that motivated the creation of the warranty of habitability, and then traces its judicial and legislative adoption.

A.    Motivations for the Warranty of Habitability

Prior to the enactment of the implied warranty of habitability, the doctrine of caveat emptor—buyer beware—applied to residential rental agreements.35 Landlords had limited obligations to maintain their units, and thus tenants were largely left to their own devices when conditions fell into disrepair. This doctrine was rooted in nineteenth-century law that conceived of the lease as merely a possessory interest in land.36 A landlord fulfilled his or her obligations under the lease simply by conveying the land.37 The tenant then had complete control over the land and was responsible for maintaining any structures on it, while also assuming unconditional liability for the rent.38 The lease contained no implied promises regarding the state of the premises being conveyed.39 This scheme developed in an agrarian context in which the typical lease had a lengthy term and the tenant farmer was as well positioned to make the repairs as the landlord.40

As demographic shifts occurred in the twentieth century, it became increasingly clear that caveat emptor was ill-suited to the realities of modern landlord-tenant relationships.41 By the 1960s and 1970s, overcrowded slums with dilapidated housing had come to characterize urban centers.42 Poor tenants faced egregious and unsafe living conditions, and extremely few had the resources necessary to make the repairs.43 The nature of contemporary landlord-tenant relationships also created different expectations.44 A tenant renting an apartment usually held a short-term lease and expected to receive more than the land itself. The tenant instead sought to rent a dwelling equipped with utilities and functioning amenities.45 There was a growing movement among legal advocates and scholars to modernize residential landlord-tenant law to conform to these expectations and needs.46

Housing codes had been enacted in many jurisdictions by this time, allowing for landlords to be held civilly and criminally liable for substandard conditions in their properties.47 However, there was strong consensus that enforcement was lacking.48 The costs associated with prosecuting landlords were high, and as commentators noted at the time, only “extreme violation[s] [ ] ha[d] any chance of being remedied in the major city setting, where large numbers of old buildings [we]re deteriorating rapidly.”49 Code enforcement agencies were underfunded and overwhelmed, and most lacked sufficient adjudicatory resources to pursue aggressive litigation.50 The agencies were also reluctant to seek criminal sanctions.51Civil liability, meanwhile, was proving an ineffective deterrent mechanism because fines were too low.52 It was often cheaper for a landlord to pay a court-ordered fine than to make repairs.53 Thus, as a mechanism for holding landlords accountable for making repairs, code enforcement was broadly considered “inefficient and unworkable.”54 It became widely understood that the modern realities of rental housing demanded a stronger legal tool.55

Public outrage was also growing at the law’s toleration of slum conditions, particularly in urban centers.56 The civil and welfare rights movements had swept the nation, generating a broad set of demands to expand the rights of poor and marginalized groups. As housing conditions were deteriorating and the size of urban slums was expanding, this context helped fuel a broad tenants’ rights movement.57 Organized tenants held rent strikes, picketed, and engaged in other forms of protest to demand improved housing quality and affordability, while also standing behind litigation and lobbying efforts oriented toward the same goals.58

The grassroots activism and legal reform efforts for better housing conditions coalesced around the goal of establishing an implied warranty of habitability in residential leases.59 The warranty would make the tenant’s covenant to pay rent mutual with the landlord’s covenant to make repairs.60 Thus, where landlords did not keep premises in good repair, tenants would be relieved of all or a part of their rental obligations.61 Tenants would be “deputize[d]” to act as “private attorney[s] general,” empowered to impose automatic financial consequences on their landlords whenever they failed to address known disrepair.62 Advocates believed that this scheme of financial liability would serve as a much-needed accountability and deterrence mechanism.63 Whereas landlords realistically perceived the threat of financial penalties for code violations or damages imposed by affirmative litigation to be minor, it was expected that landlords would take the threat of losing all rent revenues—imposed without the need for bureaucratic intervention or a drawn out court proceeding—much more seriously.64

B.    Establishment of the Warranty of Habitability

In 1970, the US Court of Appeals for the District of Columbia Circuit became the first court to recognize the warranty of habitability.65 In Javins v First National Realty Corp,66 the court held that “a warranty of habitability . . . is implied by operation of law into leases of urban dwelling units . . . and that breach of this warranty gives rise to the usual remedies for breach of contract.”67 The issue came before the court in the context of an eviction action for nonpayment of rent.68 The tenants had failed to pay rent, and when the landlord brought an eviction case seeking possession on that basis, they asserted as a defense that they were relieved of their rental obligations because the landlord had failed to make needed repairs.69

Before Javins, “the only nonprocedural defenses [to nonpayment of rent eviction] had been payment of the rent claimed and constructive eviction.”70 The court in Javins, however, both recognized the implied warranty of habitability as a legal doctrine and held that it could be invoked as a substantive defense in a nonpayment of rent eviction. The court declared that “the tenant’s obligation to pay rent is dependent upon the landlord’s performance of his obligations, including his warranty to maintain the premises in habitable condition.”71 It explained that in adjudicating whether the landlord had a right to possession of the apartment for nonpayment of rent, the lower court must first determine whether the tenants were relieved of all or a part of their rental obligations as a result of the landlord’s failure to repair.72 The reduction in the amount of rent owed, known as a rent abatement, is typically described as a percentage of the total rent owed and is based on the severity of the substandard conditions and the length of time for which they persisted.73 The court further held that if the defective conditions extinguished the tenants’ rental liability, the tenants were entitled to retain possession of the apartment.74

A wave of similar judicial opinions followed. By the late 1970s, courts in California, Hawaii, Massachusetts, New Hampshire, New Jersey, Washington, and Wisconsin, among others, had recognized the implied warranty of habitability.75 Legislatures also acted swiftly.76 By the time New York passed its warranty of habitability statute in 1975, the warranty of habitability had already been recognized by legislatures in Rhode Island (1970), Arizona (1974), and Delaware (1974).77 The doctrine was eventually adopted in some form in every state except Arkansas.78 The specific contours of the laws varied, but in its most progressive iterations, including in New York, the warranty of habitability relieved tenants of all or a part of their rental obligations so long as (1) the landlord had notice of the defective conditions, either constructively,79 orally, or in writing from the tenant or from a public agency (with no requirement that a housing code violation be issued); (2) the defective conditions affected the habitability of the premises; and (3) the landlord had failed to make repairs.80 Most jurisdictions also adopted accompanying laws protecting tenants from retaliatory eviction when they invoked their right to withhold rent as permitted by the warranty.81

While courts and legislatures cited numerous reasons for adoption of the warranty of habitability,82 they overwhelmingly emphasized that the law would act as a tool for improving the rental housing stock occupied by low- and moderate-income families.83 The Javins court noted that the “inequality in bargaining power” between landlords and tenants left tenants with “little leverage to enforce demands for better housing.”84 Among other barriers, tenants were prevented from successfully negotiating for improved conditions because “racial and class discrimination and standardized form leases . . . [left] tenants in a take it or leave it situation.”85 Severe shortages in affordable rental housing further exacerbated the inequalities in bargaining power, which, as the California Supreme Court observed, meant that “even when defects are apparent the low income tenant frequently has no realistic alternative but to accept such housing.”86 Mirroring the views of activists and commentators, courts also emphasized that the resource constraints faced by housing code enforcement agencies made a private remedy and right of action for tenants facing substandard housing conditions all the more necessary.87 These concerns were echoed repeatedly throughout the country by courts and legislatures as they ushered in one of the most revolutionary changes to landlord-tenant law in modern history.88

C.    Developments in Warranty of Habitability Laws

In recent years, many jurisdictions have narrowed the circumstances in which the warranty of habitability can be invoked. They have done so by adopting three types of limiting rules. First, “good faith” laws require tenants to demonstrate genuine withholding of rent for bad conditions.89 Under these laws, tenants cannot assert the warranty as a defense unless they can show that their motive for not paying rent was the landlord’s failure to repair.90 The laws effectively excuse landlords’ noncompliance with their obligations by removing the financial consequences the warranty imposes whenever the failure to repair coincides with other events that cause the tenant to fall behind in rent. The laws also practically diminish the availability of the warranty by increasing the burden of proof; some tenants who genuinely intended to withhold rent for defective conditions may simply have insufficient evidence to make out a good faith showing.

Second, many legislatures and courts have imposed landlords’ protective orders, also known as “rent escrow” laws, requiring tenants to deposit unpaid rent with the court as a condition of asserting the warranty of habitability.91 Some versions of rent escrow laws require tenants to deposit their rent at the time of the withholding, whereas others impose the requirement upon the tenant’s assertion of the warranty defense in the eviction case.92 Most commentators consider rent escrow requirements to severely limit the warranty’s effectiveness.93 Many tenants are unaware of the requirements and fail to comply with them during the appropriate time period. Thus by the time they appear in court, they have already effectively waived their right to assert the warranty of habitability as a defense.94 Additionally, many tenants are unable to comply with the requirements because they are using withheld rent to cope with the disrepair.95 Tenants spend money to make repairs on their own, to pay for temporary fixes such as space heaters when the heat is out or hot plates when the stove is not working, and to replace damaged possessions.96 Commentators have pointed out that the result of rent escrow laws is often that the tenants who need the protections of the warranty of habitability the most become the least likely to benefit from it.97

Third, some jurisdictions have imposed onerous notice requirements for assertion of a warranty claim.98 In their most burdensome iterations, these rules require that notice to the landlord of defective conditions be established through an official housing code violation report.99 Thus, if a tenant calls the landlord about the condition of disrepair, talks to the landlord in person, or even sends a letter describing the problem and the landlord fails to make repairs, the landlord cannot be held liable. This requirement engrafts the same problems faced by code enforcement systems onto the warranty of habitability. Where code enforcement agencies are ineffectual and underresourced, a warranty of habitability scheme tied to this system will face the exact same limitations. Commentators have also remarked that such requirements are misaligned with how tenants communicate with their landlords in practice.100

Multiple factors have motivated the enactment of these restrictive doctrines. To some extent, the doctrines reflect underlying hesitation about the establishment of the warranty of habitability.101 In some jurisdictions, legislatures and the public were never fully supportive of establishing such an impactful set of rights for tenants, and these doctrines were a way of limiting their breadth. According to Professor David Super, rent escrow requirements in particular may be a way of “appeas[ing]” landlords unhappy with the recognition of the warranty.102 He observes that where courts have recognized the warranty as a matter of common rather than statutory law, courts have been vulnerable to landlords’ criticism of judicial overreach and therefore are more willing to adopt restrictive doctrines.103 Courts and legislatures have also enacted the restrictive doctrines as a mechanism to protect against perceived tenant abuse of the warranty.104 By imposing strict notice requirements, forcing tenants to escrow their rent, or requiring a showing of good faith, courts and legislatures believe that they are ensuring that only tenants who genuinely withhold rent for bad conditions are benefiting from the warranty. According to these courts and legislatures, tenants who have failed to pay rent for a reason other than defective conditions should not be able to reap financial rewards from the establishment of a right to rent abatement if they also happen to satisfy the law’s requirements.105

II.  Existing Research on the Warranty’s Effectiveness and Theories for Tenant Underuse

Since the warranty of habitability was enacted nearly fifty years ago, scholars have tried to understand whether the law has lived up to the potential that advocates and proponents originally envisioned, and if it has not, why not.106 Multiple studies show that tenants rarely assert the warranty as a defense in nonpayment of rent eviction cases. Other studies show that very few tenants receive rent abatements. These studies, however, have limitations. The large-scale studies do not isolate cases of tenants with meritorious claims, and thus leave unknown the extent to which the outcomes constitute an operationalization gap. The only study thus far that has measured outcomes among cases with meritorious claims was conducted using a small sample size that does not purport to be representative. No study has yet determined the size of the gap between the number of tenants with meritorious warranty claims and the number who benefit from the law.

Leading scholarship on the warranty of habitability has consistently attributed the apparent ineffectiveness of the law to two factors: the lack of access to counsel, and onerous substantive doctrines (such as good faith withholding, rent escrow, and strict notice requirements that restrict the claim’s use). Yet these theories have not been subject to rigorous empirical scrutiny. The existing studies show that tenants who are represented by counsel are more likely to receive rent abatements, but these studies have not controlled for whether tenants who are represented are more likely to have meritorious claims. The scholarship on the substantive doctrines, meanwhile, has been largely theoretical in nature.

This Part provides an overview of the scholarship on the warranty of habitability, describing (a) the existing empirical studies on the law’s overall usage and effectiveness, (b) the research findings regarding the impact of legal counsel, and (c) current explanations for the law’s apparent ineffectiveness.

A.    Use and Effectiveness of the Warranty of Habitability

Marilyn Mosier and Richard Soble pioneered the empirical scholarship on the warranty of habitability in the early 1970s with a study of the Detroit landlord-tenant court in the years immediately following Michigan’s enactment of the law.107 Through case file review and in-court observations, Mosier and Soble found that rent abatements were awarded in an extremely small percentage of the total number of nonpayment of rent eviction cases.108 Specifically, they found that at most, rent abatements were awarded in 2 percent of all nonpayment of rent cases.109 Shortly after Mosier and Soble’s research was published, a team of Illinois-based researchers conducted a similar study of Chicago’s eviction court and found that zero tenants in the sample of cases they studied received rent abatements, even though 41 percent of tenants had raised the warranty of habitability as a defense.110

Two more recent studies produced findings similar to those in Mosier and Soble’s research. The first study was an observation-based study conducted by Professor Barbara Bezdek of a sample of nonpayment of rent eviction cases in Baltimore in the early 1990s.111 Bezdek found that rent abatements were ordered in only 1.75 percent of all cases she observed.112 The second study reviewed court records of all nonpayment of rent eviction cases in Essex County, New Jersey, in 2014.113 The authors, Professor Paula Franzese, Abbott Gorin, and David Guzik, calculated the overall frequency with which tenants formally raised the warranty as a defense.114 They found that the warranty was asserted in the tenant’s answer in only 0.2 percent of all cases (80 out of 40,000).115 Based on these findings, Franzese and her colleagues concluded that the warranty was significantly underutilized.116

These four studies measured the frequency with which the warranty of habitability was asserted or won (in the form of a rent abatement) within the total population of nonpayment of rent cases. None measured this frequency against the population of cases with meritorious warranty claims. Thus, the studies’ conclusions that the warranty is ineffective rest on the assumption that more tenants could have asserted or won the claim than actually did so. It is unknown whether that assumption was valid. Moreover, even if it was valid, the findings tell us little about the size of the gap between the number of tenants with meritorious claims and the number who benefited from the law.

The only study thus far that has sought to determine a tenant’s likelihood of benefiting from the warranty of habitability when he or she has a meritorious claim is Professor Michele Cotton’s “multi-case study” of fifty-nine rent escrow actions in Baltimore.117 In these actions, tenants petition the court to have their rent deposited into the court’s escrow account rather than paid to the landlord based on violations of the warranty of habitability.118 Cotton found that less than half—42 percent—of tenants who had established entitlement to a rent abatement actually received one.119 However, Cotton’s study was based on a small sample of cases that did not claim to be statistically representative of the population as a whole;120 thus, the conclusions that may be drawn from the findings are limited.

These studies leave two significant gaps in our knowledge about the use and effectiveness of the warranty of habitability. First, no large-scale study has yet compared the number of cases in which tenants benefit from the warranty against the number of cases in which tenants have meritorious claims.121 Thus, we do not know the extent to which the low usage rates reflect the law’s ineffectiveness, or simply reflect low rates at which tenants have meritorious claims. No one has yet determined the size of the operationalization gap. Second, the existing studies leave open the possibility that tenants may benefit from the warranty of habitability through outcomes other than rent abatements.122 Tenants who settle their cases may elect to leverage their right to a rent abatement to negotiate a longer repayment period or avoid a possessory judgment in favor of the landlord. No studies have accounted for this possibility. Without research that fills these gaps, we cannot properly reach a conclusion about the extent to which tenants benefit from the warranty of habitability.

B.    Impact of Legal Representation

Very limited research exists on the impact of legal representation on the use of the implied warranty of habitability. Mosier and Soble’s study of the Detroit landlord-tenant court found that tenants who were represented by counsel were more likely than unrepresented tenants to raise the warranty as a defense.123 They also found that represented tenants achieved overall better outcomes in their cases as compared to unrepresented tenants.124 However, this study did not identify the extent to which the represented tenants were more likely to have warranty of habitability claims. It is possible, in other words, that lawyers chose tenants for representation because they had meritorious claims, and thus that the higher usage of the claim and stronger outcomes simply reflect this selection bias.

The only other research that exists on the effect of counsel has been embedded within two studies on the overall impact of access to counsel in eviction cases.125 The first, a 1992 study on the impact of counsel in eviction cases in New York City, found that rent abatements were awarded in 18.8 percent of cases in which the tenant was represented by counsel, compared with only 3.3 percent of cases in which the tenant was unrepresented.126 Tenants were randomly assigned to the treatment (offer of representation) and control (no offer of representation) groups to eliminate selection bias.127 However, there was no specific control for whether the tenants in each group had meritorious warranty of habitability claims at the same rate.

The second study, a more recent assessment of the impact of access to counsel in eviction cases in Massachusetts, found that monetary outcomes were significantly more favorable to the tenant where the tenant was represented.128 These monetary outcomes reflected rent abatements resulting from the warranty of habitability, but also could reflect monetary damages awarded based on other claims129 or reductions in the rent owed due to miscalculations or partial payment by the tenant.130 Like in the 1992 study, it was also unknown whether the treated (offer of representation) and control (no offer of legal representation) groups had meritorious warranty of habitability claims at the same rate. No research has rigorously assessed the impact of counsel on the use of the warranty of habitability while controlling for whether the tenant had a meritorious claim.

C.    Explanations for the Law’s Ineffectiveness

There is a general consensus among scholars who have studied the warranty of habitability that the law’s ineffectiveness is attributable to two main factors. First, scholars claim that the ineffectiveness is a function of tenants’ lack of access to counsel.131 Nearly all tenants in eviction proceedings are unrepresented; in some jurisdictions, as many as 94 percent of tenants appear in court without counsel.132 Pointing to the research described in Part II.B, commentators argue that the overall lack of access to counsel is responsible for the claim’s underuse.133 They posit that unrepresented tenants do not have the knowledge, wherewithal, or resources required to effectively navigate the legal process in order to benefit from the warranty of habitability.134

Second, commentators argue that restrictive substantive doctrines, namely rent escrow, good faith withholding, and onerous notice requirements, limit the claim’s usage.135 These doctrines are not universal, but are becoming increasingly common across jurisdictions.136 Professor Super, a leading scholar on the warranty of habitability, attributes the “fall” of the warranty of habitability primarily to the spread of these rules.137 Writing in the California Law Review in 2011, Super finds that these “procedural obstacles have rendered the implied warranty of habitability almost irrelevant in practice.”138 He argues that the requirements are costly for tenants to comply with, are vulnerable to landlord abuse, and encourage tenants to move rather than pursue their claims.139 While he acknowledges that data on their impacts is lacking, he contends that these substantive limitations “likely are a significant contributor to the low rate of relief granted [for violations of the warranty of habitability] to low-income tenants.”140 Franzese has likewise blamed these rules for the ineffectiveness of the warranty, describing them as a “practical bar to aggrieved tenants’ very assertion of the defense of breach of the warranty.”141

Scholars have also put forward other explanations for the law’s apparent ineffectiveness.142 Super argues that tenants factor fears of retaliation into the “costs” of litigation; thus, to the extent tenants anticipate landlord retaliation, they will be unlikely to assert their rights under the law.143 Most recently, Franzese argued that the lack of centralized and accessible housing code record databases prevents judges from effectively enforcing the warranty.144 Franzese posited that the availability of code enforcement data through such a database would both inform the court’s analysis of the law and “would be a tool for [the] government to reduce or withhold any rent subsidies until the premises are restored to an inhabitable condition.”145 She explicitly pointed to New York City’s centralized code violation database as a model for other jurisdictions to follow.146

III.  Study Background and Design

This study sought to use rigorous methodological analysis to assess the extent to which tenants who have meritorious warranty of habitability claims received benefits from the claim.147 It also sought to rigorously evaluate the existing theories regarding the apparent ineffectiveness of the law, including the extent to which legal representation affects tenants’ likelihood of receiving the law’s benefits. New York City was chosen as the site for this study because, in addition to being the nation’s largest rental market, it is located in a jurisdiction that lacks the substantive doctrines often blamed for the law’s failures. This legal backdrop is ideal because it allows for disentanglement of the various contributors to the claim’s underuse. This Part describes the study’s objectives, context, data, and methodology.

A.    Objectives

The overarching objectives of this study were twofold. First, the study aimed to properly assess the effectiveness of the warranty of habitability through rigorous methods and statistical analysis. While prior large-scale studies measured the overall frequency with which tenants asserted the warranty of habitability as a claim or received rent abatements in nonpayment of rent eviction cases, this study measured what I call the “operationalization gap”—the difference between the number of cases in which the tenant has a meritorious warranty of habitability claim and the number of cases in which the tenant receives some benefit from that claim.148 It did so by identifying the cases in which the tenant appears to have a meritorious claim based on evidence of defective conditions in the unit.149 Moreover, while prior studies have focused nearly exclusively on tenants’ use of the warranty of habitability to achieve rent abatements, this study also considered the possible use of the law to achieve other beneficial case outcomes or to secure repairs.

Second, the study set out to rigorously evaluate the existing theories regarding the warranty of habitability’s ineffectiveness. As described previously,150 scholarship has consistently attributed the doctrine’s apparent failures to two factors: lack of access to counsel and restrictive substantive doctrines. The scholarship, however, has been largely theoretical in nature; no studies have yet subjected these factors to rigorous empirical scrutiny.151 This is the first study to do so. To understand the impact of counsel, I compared outcomes of cases with meritorious claims where tenants were and were not represented. To understand the significance of the restrictive substantive doctrines, I assessed the extent to which tenants benefited from the warranty of habitability in a jurisdiction (New York City) in which these doctrines are absent. While this assessment does not allow for a precise determination of the impact of the doctrines, it indicates the extent to which we can properly attribute the warranty of habitability’s ineffectiveness to them. In other words, the existing literature would predict that in jurisdictions where the restrictive doctrines do not exist, the warranty of habitability would be widely used. I assess whether this prediction is accurate. I also used the available data to glean insights into the extent to which an accessible and centralized Housing Code records database aids in judicial enforcement of the law.

These objectives translated into four specific research questions that drove the analysis of the quantitative data:

(1) How often do tenants with meritorious warranty of habitability claims receive rent abatements?

(2) To what extent do tenants with meritorious warranty of habitability claims receive other benefits as a result of the claim, such as a longer time period to pay rental arrears or the avoidance of a possessory judgment?152

(3) To what extent is the warranty of habitability serving as an effective tool to hold landlords accountable for making necessary repairs?

(4) To the extent it exists, is the warranty of habitability’s operationalization gap primarily a function of the lack of legal representation?

B.    Study Context

New York City was an optimal site for this study for multiple reasons. For one, New York’s warranty of habitability laws lack the restrictive rules that previous scholarship has blamed for the law’s ineffectiveness. Specifically, tenants are not required to deposit their unpaid rent into escrow, nor are they required to demonstrate that the reason for the nonpayment was withholding of rent for defective conditions.153 Notice requirements are also liberal: tenants are never required to provide notice in writing, let alone through the Code enforcement agency.154 New York City also has a centralized and publicly accessible Housing Code record database that judges can easily reference, which Professor Franzese predicts would aid in the law’s enforcement.155 Analysis of the effectiveness of the warranty in this context provides crucial insight into whether the barriers traditionally cited to are in fact the primary culprits for the law’s apparent ineffectiveness, or whether there are other, perhaps less well-understood factors contributing to the outcomes commentators have observed. Additionally, the data available in New York City allow for an assessment of the impact of counsel while controlling for the strength of the tenant’s warranty of habitability claim. This assessment more accurately indicates the impact of legal representation on the use of the claim than any of the studies conducted previously.

A brief overview of New York’s warranty of habitability laws and eviction procedures is necessary to contextualize the study design and results. New York enacted the warranty of habitability through legislation in 1975.156 The statute, New York Real Property Law § 235-b, provides that all residential leases, whether written or oral, contain an implied covenant that the premises be “fit for human habitation,” and that the tenants “shall not be subjected to any conditions which would be dangerous, hazardous or detrimental to their life, health or safety.”157 As in most jurisdictions, it further provides that any attempt to waive these obligations is void as contrary to public policy, and that no expert testimony is needed to establish damages.158 A landlord must have had actual or constructive notice of the conditions in order for a tenant to recover for breach of the warranty. Written notice can never be required, however, regardless of what is provided in the lease.159 As stated previously, New York has no rent escrow or good faith requirements for the assertion of the warranty of habitability.160 While the warranty can be asserted affirmatively, most tenants assert the claim as a defense and/or counterclaim once a nonpayment of rent case is commenced against them.161

In recent years, approximately 200,000 nonpayment of rent eviction cases have been filed annually in New York City Housing Court.162 Consistent with the eviction case resolution processes nationwide, the overwhelming majority of such cases are resolved through settlement agreements.163 Nearly all settlements take the form of repayment agreements in which the tenant agrees to pay the rental arrears owed within a stated period of time.164 There are three key outcomes negotiated in a repayment agreement. First, the parties negotiate the amount of money that must be repaid. Any rent abatement granted to the tenant will be incorporated into this amount.165 Where a rent abatement is granted, the agreement will reference the abatement explicitly.166 Second, the parties negotiate the length of time for repayment. If the tenant repays the amount owed by the deadline, the tenancy will be reinstated. Third, the parties negotiate whether the agreement will include a judgment for the landlord.167 What occurs if the tenant misses a payment under the agreement depends on whether the agreement contained a judgment for the landlord. If the agreement includes a judgment, the landlord is authorized to evict the tenant immediately upon the tenant’s breach of the agreement terms. If the agreement does not include a judgment, the landlord must file a motion seeking the court’s permission to go forward with the eviction.

Oftentimes, cases will include multiple settlement agreements. Where the tenant fails to pay the arrears by the deadline in the first agreement, either the tenant or the landlord can bring the case back to court. The tenant most likely would do so to seek an extension of time to pay. The tenant can also do so where the landlord has failed to comply with orders to make repairs. The landlord would bring the case back to court to seek authority for an eviction where a judgment was not awarded in the initial settlement agreement and the tenant failed to pay by the required deadline.168 Although parties have the option to have a hearing before the judge in all of these scenarios, the result will most frequently be a subsequent repayment agreement with a new deadline.169

The eviction case procedures provide numerous opportunities for tenants to assert that repairs are needed in their units and for judges to order those repairs. The pro se answer form, used by virtually all tenants who submit an answer, provides as one of the standardized response options that repairs or services are or were needed in the unit.170 Judges also ask tenants whether repairs are needed as part of the judge’s review of the settlement agreement.171 Wherever the tenant states that repairs are needed, the judge will require that the agreement include a provision obligating their performance. The agreement will enumerate the specific defective conditions and will provide “access dates” on which the repairs will be made. This process is repeated for each settlement agreement in the case.

Judges also have tools to verify the presence of defective conditions in the tenant’s unit. The Housing Code enforcement database, maintained by the New York City Department of Housing Preservation and Development (“HPD” or “the Code enforcement agency”), is publicly accessible online and is searchable by unit. This database includes a multiyear history of the complaints made, inspections performed, and violations issued for each unit. All judicial benches are equipped with desktop computers and wireless Internet, allowing judges to easily access the available data. Judges also have the authority to order the Code enforcement agency to perform Housing Code inspections.172

C.    Data

Two distinct datasets were constructed for this study. The first dataset was a statistically significant random sample of all nonpayment of rent eviction cases filed in 2016173 in which the tenant appeared.174 The dataset was built using the New York Office of Court Administration’s comprehensive database of all eviction case filings.175 This Office of Court Administration database identified the index number, case type (nonpayment of rent or “holdover”176), and whether the tenant appeared or defaulted for each case filed.177 Approximately ninety-seven thousand cases satisfied the inclusion criteria.178 From these 97,000 cases, 746 index numbers were randomly selected using a data randomization generation tool. The selection was stratified in order to account for borough-level differences in the data.179 Seven hundred and forty-six cases is a representative sample of the total study population at a 90 percent confidence interval, with a margin of error of 3 percent and a response distribution of 50 percent.180 The files for all 746 cases were retrieved from the Housing Court, scanned, and coded according to criteria and guidelines described below. The unit-level addresses for these cases were also matched with the HPD Housing Code enforcement database. This matching allowed each case to be linked to the unit’s Housing Code complaint and violation history.

The second dataset was a random sample of all nonpayment of rent eviction cases filed in 2016 in which the tenant appeared and in which one or more “hazardous” or “immediately hazardous” Housing Code violations were open at the unit at the time the case was filed.181 This dataset was constructed by matching the Office of Court Administration database with the HPD Housing Code violation database at the unit level.182 The matching identified 1,553 cases. From these 1,553 cases, 507 case index numbers were randomly selected using a data randomization generation tool. The selection was stratified in order to account for any borough-level differences in the data. Five hundred and seven cases is a representative sample of the total study population at a 90 percent confidence interval, with a margin of error of 3 percent and a response distribution of 50 percent. The files for all 507 cases were retrieved from the New York City Housing Court, scanned, and coded according to the same criteria and guidelines described below.

D.   Methodology

The case files in both datasets were coded across seventeen different criteria. A detailed description of the coding guidelines is provided in the Appendix. The criteria included whether the tenant was represented;183 whether the Answer asserted needed repairs; the outcomes of the first settlement agreement, including whether a possessory judgment entered, whether a rent abatement was awarded, and the length of time provided to the tenant to repay the arrears;184 whether the first and any subsequent settlement agreements required the landlord to perform “substantial repairs”; whether the judge ordered a Housing Code inspection; and whether the judge had accessed the Housing Code enforcement records of the unit.185 “Substantial repairs” were defined as repairs of a condition sufficiently serious to constitute a violation of the warranty of habitability.186

1.   All nonpayment cases dataset.

The first dataset—which I will refer to as the “all nonpayment cases” dataset—constituted a representative sample of all nonpayment of rent eviction cases in which the tenant had the ability to pursue claims and defenses.187 Within this dataset, cases were grouped based on whether the tenant had a meritorious warranty of habitability claim.188 Cases were assigned to the comparison group wherenre all available information indicated that the tenant had not experienced conditions of disrepair sufficient to establish a warranty of habitability claim.189 Specifically, cases were assigned to the comparison group where the tenant did not assert repairs in the Answer, there were no substantial repairs included in the settlement agreement, and there were no open “hazardous” (Class B) or “immediately hazardous” (Class C) code violations at the unit at the time the case was filed.190 Thirty-four percent of all nonpayment of rent cases met these conditions. I refer to this group as the “no meritorious claim” group.

Cases were assigned to the meritorious claim group based on the presence of factors indicating that the tenant had experienced serious conditions of disrepair, and thus likely could have established a warranty of habitability claim. These factors included (1) the assertion that repairs were needed in the tenant’s Answer; (2) the inclusion of substantial repairs in the initial settlement agreement; and (3) the inclusion of substantial repairs in multiple settlement agreements. Some evidence of conditions of disrepair was present in the majority of nonpayment of rent cases. In half (50 percent) of all nonpayment of rent cases, tenants asserted that repairs were needed in their Answer to the complaint.191 Slightly over half (51 percent) of cases included substantial repairs in the initial settlement agreement. There was not perfect overlap between cases in which repairs were asserted in the Answer and imposed in the settlement agreement—only 36 percent of cases met both conditions. There are two potential explanations for this finding. First, the Answer does not specify which repairs are needed, and thus in a certain percentage of cases the repairs claimed were likely insubstantial. Second, new repair needs may have arisen between the filing of the Answer and the settlement agreement, and thus some settlement agreements may have included substantial repairs that were not needed at the time of the Answer.192 Overall, 10 percent of cases had repairs asserted in the Answer and substantial repairs included in multiple settlement agreements.193

Table 1: Descriptive Statistics of All Nonpayment of Rent Eviction Cases

Evidence of conditions of disrepair

Percentage of nonpayment of rent eviction cases

 

Need for repairs asserted in Answer

50%

Substantial repairs in settlement agreement*

51%

Repairs asserted in Answer and substantial repairs in settlement agreement 

36%

Repairs in Answer and substantial repairs in multiple settlement agreements*

10%

No evidence of conditions of disrepair** 

34%

* One of two “meritorious claim” groups

** “No meritorious claim” group

The group of cases with meritorious warranty of habitability claims—which I will refer to as the “meritorious claim” group—was configured and tested using two different definitions: (1) cases in which the settlement agreement required the landlord to make substantial repairs (Definition 1), and (2) cases in which multiple settlement agreements required the landlord to make substantial repairs and the tenant asserted that repairs were needed in his or her Answer (Definition 2).194 The criteria included in Definition 1 were more inclusive but less confident indicators of a meritorious warranty of habitability claim, whereas the criteria used in the second definition were less inclusive but more confident indicators.195 In the Definition 2 group, cases were included only if two or more settlement agreements required repairs of the same conditions and the access dates in the first agreement had passed by the date of the second agreement.196

2.   Violation dataset.

The second dataset—which I will refer to as the “violation dataset”—constitutes a representative sample of cases in which one or more “hazardous” (Class B) or “immediately hazardous” (Class C) Housing Code violations were open at the unit at the time of filing. These are cases in which there was an even stronger indication that the tenant had a meritorious warranty of habitability claim. Conditions of disrepair that constitute Class B or Class C violations nearly always affect habitability,197 and the open status of the violation indicates both that the landlord had notice of the condition of disrepair and that the landlord likely had not yet completed repairs.198 This dataset thus comprised a third meritorious claim group. At times, subsets of the violation dataset were also used to test results among groups of cases with even stronger evidence of a meritorious warranty of habitability claim. Thus, outcomes were analyzed for subgroups of violation cases where the tenant had also asserted that repairs were needed in the Answer, substantial repairs were included in the settlement agreement, and/or substantial repairs were included in multiple settlement agreements.

The purpose of the violation dataset was primarily supplemental, as the cases included likely comprise only a small fraction of all nonpayment of rent cases in which the tenant had a meritorious warranty of habitability claim. Many tenants do not report defective conditions to the City, or do so only once their landlord has repeatedly failed to make repairs.199 Thus, the “all nonpayment cases” dataset provides a more comprehensive representation of the use of the warranty of habitability across all nonpayment of rent eviction cases. The violation dataset is included to respond to potential concerns that the methodology used to identify cases with meritorious warranty of habitability claims in the first dataset are overly inclusive, and thus that the findings are diluted. Each case included in the violation dataset had on average 3.7 Class C violations, 0.5 Class B violations, and 1.3 Class A violations open at the time of case filing, totaling 5.5 open violations per case.200 Ninety-five percent of cases in the dataset had one or more open Class C violation.

Using both datasets, Welch’s two-sample t-tests and Pearson’s chi-squared tests were performed to compare case outcomes among the three “meritorious claim” groups and the “no meritorious claim” group. As described in more detail below, outcomes compared included rent abatements, the rate of possessory judgments, the length of the repayment period, and orders to perform repairs.

IV.  Results and Discussion

This Part provides the results of the statistical analysis and discusses the answers they provide to the four specific research questions. The analysis revealed that many more tenants had meritorious warranty of habitability claims than received any benefit from the claim. A small percentage of tenants with meritorious claims received rent abatements; no tenants, however, received other benefits, such as longer repayment periods or avoidance of a possessory judgment, as a result of having a meritorious claim. And while settlement agreements very frequently imposed repair obligations, it appears that those obligations most often went unfulfilled and unenforced. The lack of legal representation accounted somewhat for the findings but was insufficient to fully explain them.

Parts IV.A and IV.B provide the results of the statistical analysis for the three types of case outcomes studied: rent abatements, possessory judgments, and length of time for payment of the arrears. Part IV.C provides the same for the data related to the enforcement of repair obligations, and Part IV.D provides the results of the analyses regarding legal representation.

A.    Question 1: To What Extent Do Tenants Who Have Meritorious Warranty of Habitability Claims Receive Rent Abatements?

The data analysis revealed that tenants who had meritorious warranty of habitability claims rarely received rent abatements. Rent abatements were granted in only 1.75 percent of all nonpayment of rent eviction cases, even though between 36 and 51 percent of the tenants in the study had meritorious claims. Put differently, a tenant with a meritorious warranty of habitability claim had between a 2.35 and 3.29 percent chance of receiving a rent abatement generally, and a 9 percent chance if there were open code violations in the unit. Even using the most conservative set of indicators to identify cases with meritorious warranty claims—cases in which there were open code violations, the tenant asserted repairs in the Answer, and substantial repairs were included in multiple settlement agreements—only 15 percent received rent abatements. In sum, the overwhelming majority of tenants who were entitled to rent abatements did not receive them. A detailed description of the statistical findings is provided below.

1.   All nonpayment of rent cases.

Rent abatements were awarded in 1.75 percent of all nonpayment of rent cases (13 out of 745).201 The percentage rose only slightly when calculated within cases with evidence of conditions of disrepair. Tenants were awarded rent abatements in 3.5 percent of cases with repairs asserted in the Answer. Of cases in which substantial repairs were included in the first settlement agreement, 2.35 percent were awarded rent abatements, and of cases in which substantial repairs were included in the settlement agreement and repairs were asserted in the Answer, 3.29 percent were awarded abatements. Abatements were granted in 2.76 percent of cases in which repairs were asserted in the Answer and substantial repairs were included in multiple settlement agreements. No abatements were awarded in the control group. The average abatement amount was $1,955. These results are presented in Table 2 below.

Table 2: Rent Abatements in All Nonpayment of Rent Cases

Case Classification

Abatement Rate

All cases

1.75%

Repairs in Answer

3.5%

Substantial repairs in settlement agreement*

2.35%

Repairs in Answer and substantial repairs in settlement agreement

3.29%

Repairs in Answer and substantial repairs in multiple settlement agreements*

2.76%

No conditions of disrepair**

0%

       * One of two “meritorious warranty claim” groups

** “No meritorious warranty claim” group

2.   Violation cases.

Rent abatements were awarded in 9 percent of all violation cases, even though the tenants in all such cases had meritorious claims. The rate of rent abatements did not increase substantially even where additional evidence existed of conditions of disrepair. Tenants were awarded rent abatements in 10 percent of cases in which the tenant has asserted that repairs were needed in his or her Answer. Of cases in which substantial repairs were included in the first settlement agreement, 13 percent were awarded rent abatements, and of cases in which substantial repairs were included in multiple settlement agreements, the same share—13 percent—were granted abatements. Abatements were awarded in 15 percent of cases in which repairs were asserted in the Answer and substantial repairs were included in multiple settlement agreements. The average abatement amount in the violation dataset was $2,275. These results are presented in Table 3 below.

Table 3: Rent Abatements in Violation Cases

Case Classification

Abatement Rate

All violation cases

9%

Repairs in Answer

10%

Substantial repairs in settlement agreement

13%

Repairs in Answer and substantial repairs in settlement agreement

13%

Repairs in Answer and multiple settlement agreements

15%

3.   Discussion.

The data revealed that tenants received rent abatements at very low rates even where there were multiple indicators that they had meritorious warranty of habitability claims. The findings showed a large operationalization gap as measured by the award of a rent abatement: only between 2.35 and 9 percent of tenants who had a meritorious warranty of habitability claim actually benefited from that claim. At minimum, these findings show that the warranty of habitability is not operating in practice as it is designed on paper: to condition rental obligations on repairs. Instead, most tenants—approximately ninety-eight out of one hundred—are being held to their full rental obligations regardless of defective conditions. The result is that landlords are rarely facing financial consequences for neglecting their properties.

The data also showed that tenants were most likely to receive rent abatements when there were open code violations in the unit. Tenants were substantially less likely (approximately one-half to one-quarter as likely) to receive abatements when there was other evidence of conditions of disrepair but no code violations. This finding is striking. Although code violations provide proof of the existence of conditions of disrepair, a primary motivation for enacting the warranty of habitability was to provide an alternative to code enforcement for holding landlords accountable for conditions of disrepair. Courts, advocates, and legislators believed that by giving tenants the power to act as “private attorney[s] general” to enforce habitability standards, the warranty would function as an important work-around to often inefficient and poorly resourced housing code enforcement systems.202 But to the extent the warranty of habitability provides meaningful relief only where the Code enforcement system has been activated, as is indicated by this data, the law is not serving this purpose.

B.    Question 2: To What Extent Do Tenants with Meritorious Warranty of Habitability Claims Receive Other Benefits from the Claim, Such as a Longer Time Period to Repay Rental Arrears or the Avoidance of a Possessory Judgment?

The data also ruled out the possibility that tenants with meritorious warranty of habitability claims receive benefits from the claim other than rent abatements. As described above, the other key outcomes negotiated in a nonpayment of rent eviction case are (1) whether a possessory judgment is awarded to the landlord,203 and (2) the length of the repayment period afforded to the tenant. The analyses of both datasets showed that there was no statistically significant difference in either of these case outcomes between cases with and without meritorious warranty of habitability claims. Tenants with meritorious warranty claims were statistically just as likely to receive a possessory judgment as tenants without warranty claims.204 In cases in which possessory judgments were awarded, there was no statistically significant difference in the length of the repayment period. Similarly, in cases in which no possessory judgment was awarded, there was no statistically significant difference in the length of repayment period.205 Thus, tenants did not appear to be “trading” the opportunity for a rent abatement for other types of desirable outcomes in their cases. A detailed description of the statistical findings is provided below.

1.   All nonpayment of rent cases.

Among “no meritorious claim” cases, 74 percent had possessory judgments and the average length of time for repayment of arrears was 37.6 days. Where a case had a possessory judgment, the average length of time for repayment was forty-two days, whereas when the case did not have a possessory judgment, the average repayment period was twenty-four days. As described in Part III.D.1, two different sets of criteria were used to identify the “meritorious warranty claim” group within the “all nonpayment of rent cases” dataset: (1) cases with substantial repairs in the settlement agreement (Definition 1), and (2) cases with substantial repairs in multiple settlement agreements and repairs asserted in the Answer (Definition 2). Among cases satisfying the criteria under Definition 1, 73 percent had possessory judgments and the average length of time for the repayment of the arrears was 39.3 days. Where a case had a possessory judgment, the average length of time for repayment was forty-four days, whereas when a case did not have a possessory judgment, the average repayment period was twenty-six days. Among cases satisfying the criteria under Definition 2, 75 percent had possessory judgments and the average length of time for repayment of arrears was forty days. Where a case had a possessory judgment, the average length of time for repayment was forty-four days, whereas when a case did not have a possessory judgment, the average repayment period was twenty-nine days.206

Welch’s two-sample t-tests and Pearson’s chi-squared tests were performed to test for statistical significance in the difference in outcomes between the “no meritorious claim” cases and each of the two “meritorious warranty claim” case groups. There was no statistically significant difference in outcomes between the “no meritorious claim” comparison group and either of the two “meritorious warranty claim” group. The full statistical results are reported in Tables 4 and 5 below.

2.   Violation cases.

Sixty-four percent of violation cases had possessory judgments. The average length of time for repayment of arrears among all violation cases was 36.4 days. The average repayment period was 42 days for cases with possessory judgments, and 26 days for cases without possessory judgments.

Pearson’s chi-squared and Welch’s two-sample t-tests were performed to test for statistical significance in the difference in outcomes between the violation cases and the “no meritorious claim” cases (in the “all nonpayment cases” dataset). The results showed no statistical significance in the average length of repayment period or in the rate of possessory judgments. The average length of the repayment period also did not differ at a level of statistical significance when the issuance of a possessory judgment was held constant. Specifically, the repayment period was the same in violation cases with possessory judgments and “no meritorious claim” cases with possessory judgments. There was also no statistically significant difference between violation cases without possessory judgments and “no meritorious claim” cases without possessory judgments. The full statistical results are reported in Tables 4 and 5 below.

Table 4: Possessory Judgment Rate in All Nonpayment of Rent and Violation Cases

Case classification

Percentage of cases with possessory judgment for landlord

P-value207based on difference with no meritorious claim group

Substantial repairs in settlement agreement*

73%

0.78

Repairs in Answer and multiple settlement agreements*

75%

0.91

Violation cases

64%

0.09

No conditions of disrepair**

74%

* One of two “meritorious claim” groups among all nonpayment of rent cases

** “No meritorious claim” group

Table 5: Average Length of Repayment Period in All Nonpayment of Rent and Violation Cases

* One of two “meritorious claim” groups among all nonpayment of rent cases

** “No meritorious claim” group

3.   Discussion.

This research is the first to address the possibility that tenants with meritorious warranty of habitability claims are benefiting from the claim by achieving favorable case outcomes other than rent abatements. It effectively rules out this possibility. While tenants with open code violations at their units were slightly more likely to avoid possessory judgments as compared with tenants without warranty claims, this difference was small and not statistically significant. Moreover, such tenants still “paid” for this avoidance of the judgment with a shorter repayment period, equal to that awarded to tenants without warranty claims who also avoided a possessory judgment.208 The achieved benefit was therefore minimal.

These findings, together with the rent abatement findings, indicate that the vast majority of tenants with meritorious warranty of habitability claims did not receive any material benefit from the claim. The small percentage of tenants who received rent abatements indeed comprised the only tenants with likely meritorious warranty claims who benefited from the law at all. In other words, between 2.35 and 9 percent of all tenants who should have been able to invoke the law were able to successfully do so. The warranty of habitability did not provide any benefit at all to approximately 91 to 97 percent of tenants who appeared to satisfy the elements of the claim.

C.    Question 3: Does the Warranty of Habitability Serve as an Effective Tool to Hold Landlords Accountable for Making Needed Repairs?

It is possible that although most tenants are unable to successfully invoke the warranty to achieve rent abatements or other beneficial outcomes in their eviction cases, they are effectively using the law as a tool to compel landlords to perform needed repairs. The settlement agreements in slightly over half of all nonpayment of rent cases included an order obligating the landlord to make substantial repairs, which would seem to indicate that the law is being used in this way.209 Yet the fact that the settlement agreement included such an obligation does not necessarily mean that the landlord complied with it and made the repairs.

The data do not allow for conclusions to be drawn regarding the extent to which repairs were ever completed once they were ordered in settlement agreements. However, cases that have multiple settlement agreements provide insight into the extent to which repair orders are followed. Cases result in more than one settlement agreement when a tenant fails to comply with the repayment terms set forth in the initial settlement agreement.210 The landlord then takes the next step toward eviction, and either the landlord or the tenant will bring the case back to court.211 The parties will then enter into a new settlement agreement, typically a repayment agreement.212 If applicable, that agreement will again include an order for the landlord to make any necessary repairs.

Among cases that have (1) repairs ordered in the initial settlement agreement, and (2) a subsequent settlement agreement entered into after the “access dates” included in the initial settlement agreement, the frequency with which the same repairs are included in a subsequent settlement agreement provides some indication of the extent to which repair orders are followed. Specifically, where a case has two or more settlement agreements and the first agreement included an order for the landlord to make repairs, the fact that the same repairs are ordered in a subsequent settlement agreement (entered into after the access dates for repairs in the first agreement have passed) strongly suggests that the landlord did not comply with the initial repair order.213 Conversely, where a case has two or more settlement agreements and the first agreement included an order for the landlord to make repairs, the fact that the same repairs are not ordered in a subsequent agreement (entered into after the access dates for repairs in the first agreement have passed) strongly suggests that the landlord complied with the initial repair order. Thus, the frequency of each outcome was calculated to determine the extent to which landlords comply with repair orders included in settlement agreements. The findings indicate that repair orders were not complied with in nearly three-quarters of all cases where the data allow for this analysis.

Two other case activities serve as additional indicators of the extent to which the warranty of habitability is effectively used to improve housing quality within eviction cases: the frequency with which judges order Housing Code inspections, and the frequency with which judges access Housing Code enforcement records.214 As described in Part III.B, judges presiding over nonpayment of rent eviction cases have broad authority to order the Housing Code enforcement agency to perform an inspection of the unit.215 This authority is significant because it allows judges to use the information they gather through eviction cases regarding conditions to trigger a parallel enforcement system. Where a tenant reports that she does not have heat, for example, the judge’s order of a Housing Code inspection means that if the tenant’s report is accurate, the Housing Code enforcement agency will initiate its own action against the landlord to ensure the repair is made. The landlord’s obligation to repair thus will no longer be tied to the eviction case, nor will it depend on the tenant’s ability or willingness to enforce the judge’s repair order.216

Judges also have the ability to access Code enforcement records, which include the history of complaints, inspections, and violations issued within the prior year. The ability to obtain these records is significant because it means that the judge has access to external, objective information about the conditions of the tenant’s unit, which, as Professor Franzese argues, should help promote enforcement of the warranty of habitability.217 The availability of the database also means that a judge can easily know whether the Housing Code enforcement agency is already aware of or involved in the conditions in the tenant’s unit. A judge who is concerned about a tenant’s report of serious conditions of disrepair can know whether it is worth ordering a Code inspection, or whether doing so would be duplicative because the agency is already involved. In other words, this integration should help encourage judges’ appropriate use of their authority to order Housing Code inspections.

Despite the integration of the Code enforcement and Housing Court systems, the data show that judges rarely use these tools to enforce the warranty and promote repair issues in the tenant’s unit. The full results of the analyses are reported and described below.

1.   All nonpayment of rent cases.

In nonpayment of rent cases in which substantial repair orders were included in the original settlement agreement and the parties entered into a subsequent settlement agreement after the access dates in the original settlement agreement had passed, the subsequent agreement included the same repair obligations 72 percent of the time.218 Judges invoked their authority to order a Housing Code inspection in only 1.2 percent of all nonpayment of rent cases. Perhaps even more striking, such an inspection was ordered in only 0.4 percent of cases in which substantial repairs were included in the settlement agreement where there were no open Housing Code violations at the time of case filing or complaints made to the Code enforcement agency within six months prior to the filing.

2.   Violation cases.

In violation cases in which substantial repair orders were included in the original settlement agreement and the parties entered into a subsequent settlement agreement after the access dates in the original agreement had passed, the subsequent agreement included the same repair obligations 80 percent of the time. Judges invoked their authority to order a Housing Code inspection in only 1.8 percent of all violation cases. At the same time, there is little evidence that judges were aware of open Housing Code violations in the unit. A printout of the online record of the Code enforcement history of the unit was included in the case file in only 5.7 percent of cases, even though there were open Code violations in every case included in this dataset.219

While judges may have accessed the Code enforcement database and not printed out a paper copy of the record for the file, circumstances suggest that such behavior would be unlikely. For one, it is typically court attorneys (attorneys who assist the judge in the courtroom) who access online records, and a printout of the record would be the most likely method of presenting the record to the judge. Second, it makes logical sense that judges (through their court attorneys) would print out and preserve the record once they have accessed it. Complete eviction case file records exist only in hard paper copy, rather than in any electronic database, and thus in the context of this system, the practical action for judges to take upon accessing an online record related to a case would be to add it to the paper file. Moreover, cases tend to involve multiple court appearances, and thus judges who accessed this record would likely want to remind themselves of the record in a later court appearance. Thus, the finding that a paper copy of the Code enforcement record was in the file in only 5.7 percent of cases likely reflects the frequency with which the judge indeed accessed the record.220

3.   Discussion.

These findings strongly suggest that the warranty of habitability is not serving as an effective tool to compel the performance of needed repairs. In the overwhelming majority of cases in which repairs were ordered in settlement agreements, it appears that landlords did not in fact follow through on their obligations. To be sure, it is unknown to what extent landlords later complied with their obligations even though they did not comply on the scheduled access dates. However, the fact that between 72 and 80 percent of repairs appeared to have not been performed on the scheduled access dates strongly suggests that landlord’s repair obligations are not being effectively enforced in the course of nonpayment of rent eviction cases.

The findings also indicate that judges rarely utilized the tools available to them to hold landlords accountable for needed repairs. Judges invoked their authority to order Housing Code inspections in only a tiny share of cases, despite tenants’ frequent reporting of serious conditions of disrepair. Had they done so, they would have triggered an overlapping enforcement system that should have then provided an additional layer of landlord accountability. Thus, even if the Housing Court judges were not able to unilaterally enforce habitability laws, they would have activated a system that perhaps could do so more effectively. However, judges did not follow this path.

Judges also rarely took advantage of the opportunity to learn the Housing Code enforcement history at the unit. In the violation dataset, judges accessed the Code enforcement history only 5.7 percent of the time. Thus, nearly 95 percent of the time that there were code violations at the unit, the judge was likely unaware of this fact (or did not have the full information regarding which violations were still outstanding and which had been cleared). This finding further indicates that judges’ failure to frequently order Housing Code inspections was not simply a response to their awareness that the Code enforcement agency was already involved with the unit. Rather, the finding suggests that judges generally are not aware of code violations that exist in tenants’ units, and yet still decline to order code inspections when tenants report defective conditions.

D.   Question 4: To the Extent It Exists, Is the Warranty of Habitability Operationalization Gap Simply a Result of the Lack of Legal Representation?

The data showed that legal representation substantially affected tenants’ ability to benefit from the warranty of habitability.221 Represented tenants with meritorious warranty of habitability claims were at least nine times more likely than unrepresented tenants with meritorious claims to receive a rent abatement.222 Except where there were open code violations in the unit, unrepresented tenants virtually never received abatements when they had meritorious claims. Approximately one in four represented tenants, meanwhile, received abatements when they had meritorious claims, whether identified based on either of the two sets of criteria in the “all nonpayment cases” dataset or the presence of open code violations. These findings strongly suggest that the lack of legal representation is an important contributor to the operationalization gap that has been detected.

However, the findings also show that the lack of legal representation does not fully account for the operationalization gap. Although rent abatements were much more frequent where tenants had legal counsel, rent abatements were not the norm in meritorious claim cases even among cases in which the tenant was represented. Most represented tenants—approximately three-quarters—with meritorious warranty of habitability claims did not receive rent abatements, even when they had open code violations in their units. These findings suggest that factors beyond the lack of access to counsel are also responsible for the operationalization gap.

As a preliminary matter, Pearson’s chi-squared tests were performed to test for selection bias in representation—that is, whether lawyers were choosing cases for representation based on the strength of the warranty of habitability claim.223 In the “all nonpayment cases” dataset, the tenant was unrepresented by counsel in 91 percent of all nonpayment cases, and represented by counsel in 9 percent of cases. To test for selection bias, I first looked at whether represented cases were more likely to include substantial repairs in the settlement agreement. The results showed that there was no statistically significant difference among the rate at which repairs were included in settlement agreements between represented and unrepresented cases.224 Next, I looked at whether represented cases were more likely to assert needed repairs in the Answer and/or to include substantial repairs in multiple settlement agreements. The results showed that the incidence was exactly the same—11 percent—where the tenants were represented and unrepresented. Pearson’s chi-squared tests again showed that there was no statistically significant difference between these rates. Thus, these results indicate that it is unlikely that lawyers were selecting cases for representation based on the presence of a meritorious warranty of habitability claim; overall, tenants had meritorious warranty claims at the same rate whether they were or were not represented. The full results are reported in Table 6 below.

Table 6: Presence of Conditions of Disrepair in Represented Versus Unrepresented Cases

Evidence of conditions of disrepair

Incidence in represented cases

Incidence in unrepresented cases

P-value

Substantial repairs in settlement agreement

53%

51%

0.61

Substantial repairs in multiple settlement agreements and repairs in Answer

11%

11%

0.84

No conditions of disrepair

31%

23%

A similar analysis was performed to test for selection bias in the violation dataset. In this dataset, the tenant was unrepresented by counsel in 79 percent of cases and represented by counsel in 21 percent of cases. To test for selection bias in representation, Welch’s two-sample t-tests compared the number of open violations in unrepresented versus represented cases. Cases in which the tenant was represented had an average of 1.5 Class A, 0.6 Class B, and 4.3 Class C violations open at the time of case filing. Cases in which the tenant was unrepresented had an average of 1.3 Class A, 0.4 Class B, and 3.6 Class C violations open at the time of case filing. The differences between these two groups, compared separately for each code violation class level, also were not statistically significant. These findings strongly suggest that counsel did not select cases for representation based on the number or severity of open code violations in the unit at the time of case filing. The full statistical results are reported in Table 7 below.

Table 7: Open Code Violations in Represented Versus Unrepresented Cases

 

Number of open violations in represented cases

Number of open violations in unrepresented cases

P-value

Class A

1.5

1.3

0.43

Class B 

0.6

0.4

0.51

Class C

4.3

3.6

0.07

1.   All nonpayment cases.

Pearson’s chi-squared tests were performed to test for differences in the rate of rent abatements among represented and unrepresented tenants with meritorious warranty claims. The results revealed that for tenants with the same evidence of conditions of disrepair, there were substantial and statistically significant differences in abatement outcomes based on representation status. Where substantial repairs were included in the first settlement agreement, the abatement rate was 27 percent for represented tenants compared with 0 percent for unrepresented tenants. Where substantial repairs were included in multiple settlement agreements and repairs were asserted in the Answer, the abatement rate was 30 percent for represented tenants compared with 0 percent for unrepresented tenants.225 The full statistical results are reported in Table 8 below.

2.   Violation cases.

Pearson’s chi-squared tests were also performed to test for differences in the rate of rent abatements among represented and unrepresented tenants with meritorious warranty claims, where merit is indicated by open code violations. The results showed that where there were open Class B or Class C violations at the unit at the time of case filing, the abatement rate was 27 percent for represented tenants compared with 3 percent for unrepresented tenants, and that this difference was statistically significant. Thus, legal representation had a demonstrated positive effect on the ability of tenants to successfully invoke the warranty of habitability. This finding is consistent with the finding in the “all nonpayment cases” dataset, which likewise showed that representation affected tenants’ likelihood of benefiting from the warranty. The full statistical results are reported in Table 8 below.

Table 8: Abatement Rates in Represented Versus Unrepresented Cases

Evidence of conditions of disrepairs

Abatement rate in

represented cases

Abatement rate in unrepresented cases

P-value

Substantial repairs in settlement agreement*

27%

0%

0.003

Substantial repairs in multiple settlement agreements and repairs in 

Answer*

30%

0%

0.003

Violation cases

27%

3%

0.003

No conditions of disrepair**

0%

0%

* One of two “meritorious claim” groups among all nonpayment of rent cases

** “No meritorious claim” group

3.   Discussion.

The findings show that legal representation substantially affects a tenant’s likelihood of receiving a rent abatement when he or she has a meritorious warranty of habitability claim. Strikingly, they demonstrate that the warranty of habitability is all but inaccessible to tenants without counsel who appear to satisfy the elements of the claim but who do not have open code violations at their units. Tenants are simply unable to reap the benefit of the claim prescribed by the law on paper—a rent abatement—when they are unrepresented. Represented tenants with the same evidence of conditions of disrepair have a one-in-four or one-in-three chance of receiving a rent abatement. The warranty is slightly more useful to unrepresented tenants where there are open code violations in the unit, with 3 percent receiving rent abatements. However, the impact of representation is still extremely significant. Represented tenants are nine times as likely to receive a rent abatement as compared to unrepresented tenants who have the same number and class levels of open code violations at their units. Representation, in short, dramatically affects the ability of tenants to benefit from the warranty of habitability.

At the same time, these findings indicate that representation does not fully account for the operationalization gap in the warranty of habitability. At most, between one-quarter and one-third of represented tenants with meritorious warranty of habitability claims receive rent abatements. This means that at least two-thirds of tenants with meritorious warranty claims do not benefit from the claim despite having legal representation.

V.  Implications of the Findings for Our Understanding of the Warranty of Habitability and Access to Justice

The findings of this study reshape our understanding of the effectiveness of the warranty of habitability. The findings provide the most conclusive evidence to date that there is a large operationalization gap in the law. All prior large-scale empirical studies on the warranty have measured the rate at which the claim was asserted or won within the overall population of nonpayment of rent eviction cases, without distinguishing between cases of tenants with and without meritorious claims. This prior research sounded the alarm that the law was likely ineffective, but left open the possibility that the low usage rate simply reflected a low rate of tenants with meritorious claims. This study addressed these methodological shortcomings by specifically measuring the size of the gap between tenants who have meritorious warranty claims and those who benefit from the law. It also took into account the possibility that tenants with meritorious claims were forgoing rent abatements—the relief explicitly provided under the law—in favor of other benefits in their cases. The results together showed that more than 90 percent of tenants with meritorious claims did not benefit from the warranty at all. The results further revealed that tenants were unable to use the law as a tool to secure needed repairs. While judges often ordered landlords to perform repairs, the data shows that landlords evaded compliance with the orders nearly three-quarters of the time. These findings strongly indicate that the warranty of habitability suffers from a major operationalization gap.

The results of this study are especially significant because they upend the traditional wisdom about the driving forces behind the warranty’s ineffectiveness. Almost all of the existing scholarship on the warranty of habitability to date has attributed its failures to the barriers imposed by restrictive substantive doctrines and the lack of access to counsel. The findings here show that those explanations are inadequate. First, the study found that tenants’ claims have a low rate of effectiveness even where the law is unencumbered by restrictive substantive doctrines. New York’s warranty of habitability laws lack onerous notice, good faith withholding, or rent escrow requirements—indeed, tenants face few formal hurdles to assertion of the claim. Existing scholarship would suggest that this backdrop would translate into widespread use of the claim.226 Yet the study found the opposite: very few tenants with meritorious claims actually benefited from the law.

It certainly may be the case that even fewer tenants benefit from the warranty of habitability where restrictive doctrines exist. However, the findings of this study demonstrate that these doctrines cannot, without more, explain the low usage rates of the law. This result has serious implications for policy. Proposals for legal reforms to the warranty of habitability, particularly those put forth by scholars and advocates in recent years, have focused primarily on the rollback of these restrictive doctrines.227 The findings suggest that those reforms are unlikely to result in widespread effectiveness of the law.228

The study’s findings also disrupt our understandings and assumptions about the role of access to counsel in the effectiveness of the warranty of habitability. While the data showed unambiguously that representation mattered, it also revealed that the lack of access to counsel did not account for the majority of the warranty of habitability’s operationalization gap. This finding has important implications for future research and policy. In 2017, shortly after the period for which the data in this study was collected, New York City became the first jurisdiction in the United States to enact legislation establishing universal access to counsel for low-income tenants in eviction proceedings.229 The legislation is being phased in over a five-year period such that all income-eligible tenants will be offered free legal counsel by 2022.230 Other jurisdictions quickly followed suit: in 2018, a San Francisco ballot initiative established the right to counsel for all tenants in eviction cases, and Newark, New Jersey passed an ordinance guaranteeing representation to tenants under 200 percent of the federal poverty line.231 A number of motivations underlie these initiatives, among them that the provision of counsel would lead to stronger outcomes for tenants and greater enforcement of existing protections.232

While only a comprehensive and rigorous evaluation of the implementation of the laws will show their effects, the findings in this study suggest that they will likely enhance usage of the warranty of habitability for tenants with meritorious claims. In this regard, the study’s findings lend support to scholars’ contentions that the lack of access to counsel acts as a barrier to the effectiveness of the warranty of habitability.233 They also bolster existing views that expanded access to counsel will improve outcomes for tenants.234 However, the results also indicate that the provision of legal representation likely will not, on its own, be enough to expand the benefits of the warranty of habitability to all—or even most—tenants with meritorious claims. The study showed that among tenants with meritorious claims who had legal representation, 75 percent did not benefit from the claim. Thus, while universal access to counsel is likely to improve the effectiveness of the warranty, it is unlikely to serve as a cure-all.

The findings also cast doubt on the argument that the warranty’s ineffectiveness is attributable in part to the inaccessibility of Housing Code records. Professor Franzese has argued that in many jurisdictions, judges are without the tools to effectively enforce the warranty of habitability because there is no centralized and publicly available code violation database.235 Franzese has hypothesized that the availability of those records to judges through a centralized database would promote enforcement of the warranty.236 Unfortunately, findings here strongly indicate that the mere existence of such a system is not, without more, a cure-all for improving the usage of the warranty. Judges in New York City have precisely the tools Franzese identified—indeed, Franzese points to New York City’s integrated system as a model for other jurisdictions to follow—but the data show that judges rarely took advantage of them. Moreover, few tenants benefited from the warranty of habitability despite the existence of this integrated system.237

These conclusions signal that current understandings of the barriers to use of the warranty of habitability are incomplete. None of the existing theories for the law’s ineffectiveness withstands empirical scrutiny. While the data show that some of the identified barriers, such as lack of access to counsel, certainly contribute to the claim’s underuse, they also show that these barriers cannot account for the scope of the underuse.

Conclusion

Nearly fifty years ago, the US Court of Appeals for the District of Columbia Circuit declared that the warranty of habitability was implied in all residential leases. Proponents hailed the development as a revolution in tenants’ rights. Professor Myron Moskovitz, writing in the California Law Review shortly after the first jurisdictions adopted the implied warranty of habitability, predicted that by giving tenants the power to enforce laws prohibiting substandard housing, the courts’ rulings would spur improvements to the quality of housing, particularly that enjoyed by low-income tenants in urban settings.238 The law would do so “not merely by adding to the number of enforcers,” but by allowing enforcement to be driven by those most affected.239 This Article presents the results of the first rigorous empirical study assessing the effectiveness of the warranty of habitability. The study demonstrates that tenants overwhelmingly do not benefit from the warranty even when they are likely to have meritorious claims. Specifically, the study found that a mere 2.35 to 9 percent of tenants with meritorious warranty of habitability claims receive rent abatements. The findings also ruled out the possibility that tenants are receiving other types of benefits from the claim, such as a longer repayment period or avoidance of a possessory judgment. And further, the findings indicate that the warranty of habitability does not serve as an effective tool within eviction cases to compel landlords to perform repairs—although the court often orders landlords to complete repairs, the data strongly suggest that landlords rarely comply with these orders.

This study was also the first to rigorously evaluate whether and to what extent legal representation affects a tenant’s likelihood of benefiting from the warranty of habitability. It found that representation mattered significantly—tenants with meritorious warranty of habitability claims had between a 0 and 3 percent chance of obtaining an abatement when they were unrepresented, compared with an approximately 27 percent chance when they had representation. This finding strongly supports providing increased access to counsel as one way to improve usage of the claim. Yet the findings should also sober expectations that a right to counsel will eliminate the warranty of habitability operationalization gap. Approximately 73 percent of tenants who had meritorious claims and were represented by counsel still did not benefit from the law.

The findings of the study also caution against an overfocus on the onerous substantive doctrines that exist in some jurisdictions. While those doctrines may very well impose additional barriers to the implementation of the warranty where they exist, the results here show that their existence does not fully—or even primarily—account for the operationalization gap. Even where the warranty of habitability is unencumbered by these doctrines, it is still not widely enforced.

These conclusions signal strongly that more quantitative and qualitative research is needed to identify other procedural and/or substantive legal barriers to the claim’s usage beyond those identified by the existing scholarship. Preliminary qualitative and legal research conducted in conjunction with the quantitative research presented here suggests that nondiscretionary cure period rules severely restrict the use of the warranty. Until 2019,240 New York had a nondiscretionary cure period rule, codified at New York RPAPL § 747-a but commonly known as the Five-Day Rule, which provided that if a landlord has obtained a judgment in a nonpayment eviction proceeding and “more than five days has elapsed,” then “the court shall not grant a stay of the issuance or execution of any warrant of eviction” until the tenant has paid the amount of the judgment.241 In the context of the warranty of habitability, the effect of this statute was that where a tenant is awarded a rent abatement at trial due to the landlord’s breach of the warranty of habitability, unless the rent abatement was for 100 percent of the arrears, the tenant would be required to pay the balance of the rent owed within five days in order to avoid eviction.

Statutes like the Five-Day Rule are quite common across jurisdictions, yet have received virtually no scholarly attention in discussions of the effectiveness of the warranty of habitability.242 At least seven other states have equivalent rules providing for very short, nondiscretionary cure periods upon a finding of rent owed to the landlord.243 The cure periods established in these states range from three to ten days.244 Even worse, at least thirty states provide tenants no cure rights at all.245 Thus, in these jurisdictions, if at the conclusion of trial the judge determines that the tenant owes one dollar of rent or more, the tenant has no opportunity to satisfy the balance and will face near-immediate eviction. This outcome is the same regardless of whether the tenant has withheld rent for defective conditions in good faith and/or whether the court has awarded the tenant a partial rent abatement for the landlord’s violation of the warranty of habitability.

These rules significantly increase the risks of taking a nonpayment of rent case to trial for the purpose of securing a rent abatement. As shown by the data here, repayment agreements will almost always provide tenants more than ten days to repay the arrears.246 That longer period of time is often necessary for tenants to save up enough money to pay down the balance, or to seek out and obtain charitable assistance. Thus, in jurisdictions with nondiscretionary cure periods, tenants are unwise to take a case to trial unless they are in possession of or have ready access to the balance of the arrears (whatever the amount of that balance may be, as determined by the judge). In jurisdictions with no cure rights, tenants who take their case to trial must be confident that the amount of the rent abatement will exceed the amount of rent owed.

Because cure period restrictions affect tenants’ risks of taking a case to trial for the purpose of achieving a rent abatement, they also heavily influence tenants’ ability to negotiate a rent abatement in a settlement agreement.247 In jurisdictions with nondiscretionary cure rules, tenants who do not possess the amount of money likely to represent the remainder of the arrears found owed are unable to successfully negotiate a rent abatement because they cannot make good on the threat of taking the case to trial. Similarly, in jurisdictions with no cure rights, tenants have little leverage to negotiate a rent abatement because landlords know that tenants are unlikely to take their case to trial: if any amount of rent is found to be owed—that is, if the rent abatement is any less than the full value of the arrears—the tenant will be evicted.

Additional research should also explore whether the ineffectiveness of the warranty of habitability is attributable to nondoctrinal factors such as court culture or imbalances of power. The preliminary qualitative research conducted in conjunction with the quantitative research presented here suggests that a debt collection culture of the housing courts may play a significant role.248 Some tenants described the Housing Court culture as treating landlords’ rights to collect rent more seriously than tenants’ rights to adequate housing. Tenants reported numerous instances of failed efforts to hold their landlords accountable for property conditions, which occurred simultaneously while they were being held responsible for their rental obligations. According to tenants’ accounts, their efforts failed not because their claims were invalid or because they were unfamiliar with the proper legal procedures, but because judges did not want to entertain them.

Further research should be conducted into both of these—as well as many other—possible explanations for the limits of the law.

Appendix: Case File Coding Guidelines

Background Information About the Case

Criteria

Coding Guideline

Rent-regulation 

status of the unit

This information was coded based on the landlord’s assertion of the rent-regulation status in the petition.249 Units were classified in one of three categories: rent-regulated status, market status, and nonprofit or government-owned.

Legal representation 

Tenants were coded as either represented or unrepresented based on whether they had representation at the time they entered into the first settlement agreement. 

 

Answer

Criteria

Coding Guideline

Date of Answer

The date the tenant completed the Answer was marked.

Whether repairs are asserted in the Answer

Cases were coded either “yes” or “no.” The pro se Answer form includes a checkbox option which states, “There are or were conditions in the apartment and/or the building/house which the Petitioner did not repair and/or services which the Petitioner did not provide.” “Yes” was marked when the box was checked, and “No” was marked when the box was blank, unless the tenant later received leave of court to amend the Answer and in the Amended Answer included a similar claim asserting conditions of disrepair (including an express claim for breach of the warranty of habitability). 

Whether a service defect is asserted

The pro se Answer form includes two checkbox options related to service defects: (1) “I did not receive the Notice of Petition and Petition” and (2) “I received the Notice of Petition and Petition, but service was not correct as required by law.” “Yes” was marked when either of the boxes was checked, and “No” was marked when both boxes were blank, unless the tenant later received leave of court to amend the Answer and in the Amended Answer asserted a service defect.

Case Outcomes

Criteria

Coding Guideline

Date of first settlement agreement

The date of the first settlement agreement resulting in a case disposition was marked.

Whether the settlement agreement includes a judgment for the landlord

This was coded as “yes” or “no” based on whether the first settlement agreement granted a judgment for possession to the landlord. Where the case was discontinued, this outcome was coded as “DISCON.” Cases that went to trial were marked “TRIAL.”

Number of days for payment of the arrears

The number of days between the date the settlement agreement was entered and the date the arrears were due. Where the settlement agreement set a schedule for incremental repayments over a period of time longer than sixty days, the outcome was coded as “pay agreement.” Where the settlement agreement set a schedule for incremental repayments over a period of time shorter than sixty days, the length of time was calculated based on the final date on which repayment would be due. Where the case was discontinued, this outcome was coded as “DISCON.” All coding was based on the first settlement agreement. Cases that went to trial were marked “TRIAL.” 

Amount of arrears owed 

The amount owed was as stated on the settlement agreement as the amount of arrears due and owing. Where a rent abatement was awarded, the abatement amount was not reflected. Ongoing use and occupancy also was not included. “DISCON” was coded for discontinued cases. All coding was based on the first settlement agreement. 

Whether the settlement agreement requires the landlord to make substantial repairs

This was coded as “yes”/“no.” “Yes” was coded where the settlement agreement included repairs of conditions that qualify as rent impairing pursuant to NY Multiple Dwelling Law § 302-a. “Yes” was also coded where the agreement included repairs of conditions that have been found to constitute a violation of the warranty of habitability, which includes, inter alia: lack of heat and/or hot water; flooding; fumes and/or smoke, leaking gas; lead paint; bedbugs; mold; broken appliances (for example, refrigerator or stove); cockroaches; secondhand smoke; mice and/or rats; noise and/or dust; failure to install kitchen facilities, and broken locks.

Whether there are multiple settlement agreements

This was coded as “yes” or “no” only if the first settlement agreement included substantial repairs. “NA” was marked if the first settlement agreement did not include substantial repairs or was a discontinuance. 

Same repairs in multiple settlement agreements

Coded as “yes” if there are multiple settlement agreements and a subsequent settlement agreement requires the landlord to make the same repairs as required by the first settlement agreement and the access dates in the first settlement agreement have passed. Marked as “no” if there are substantial repairs in the first settlement agreement and the same repairs are not included in a subsequent settlement agreement but the access dates have passed. Marked as “NA” if the access dates have not passed or if there were not substantial repairs included in the first settlement agreement. 

Abatement

Coded as “yes” or “no” based on whether the settlement agreement expressly stated that the landlord granted the tenant a rent abatement. 

Abatement amount

The dollar amount of the abatement awarded.

Amount of arrears owed of $7,000, $9,000, or $11,000

Coded as “yes” or “no” based on whether the amount of arrears owed as stated on the settlement agreement was either $7,000, $9,000, or $11,000.

Prospective setting of rent

Coded as “yes” or “no” based on whether the settlement agreement prospectively sets the tenant’s rent. 

 

Housing Code Enforcement

Criteria

Coding Guideline

HPD record in file

This was coded as “yes” or “no” based on whether there was a printed-out record of the Code enforcement history of the unit in the case file. 

Housing Code inspection order/request

This was coded as “yes” or “no” based on whether the judge submitted a standardized form titled “Judicial Request/Order for Housing Inspection.” 

 

  • 1. Request for and Report of Resource Assistant Premises Visit, Beaumont Management Group, LLC v Jackson, LT-021832-16/BX, *1–2 (NY City Civ filed Dec 21, 2016).
  • 2. Id.
  • 3. Open Violation Summary Report, Beaumont, LT-021832-16/BX, *1 (NY City Civ filed Aug 8, 2017).
  • 4. See Javins v First National Realty Corp, 428 F2d 1071, 1072–73 (DC Cir 1970). The warranty of habitability is often referred to as the “implied warranty of habitability” because it is implied in every residential rental agreement. See, for example, id at 1080 (“[T]he District’s housing code requires that a warranty of habitability be implied in the leases of all housing that it covers.”). In New York, the doctrine is typically referred to as the “warranty of habitability” because it was enacted by statute. See NY Real Prop Law § 235-b. I use the term “warranty of habitability” or simply “warranty” to reflect this local usage and for simplicity of language.
  • 5. See, for example, Javins, 428 F2d at 1072–73 & n 3.
  • 6. Petition Non-Payment Dwelling, Beaumont, LT-021832-16/BX, *1 (NY City Civ filed Apr 1, 2016).
  • 7. See, for example, Park West Management Corp v Mitchell, 391 NE2d 1288, 1295 (NY 1979).
  • 8. Stipulation of Settlement, Beaumont, LT-021832-16/BX, *1 (NY City Civ filed July 27, 2016).
  • 9. Id.
  • 10. Affidavit of Respondent, Beaumont, LT-021832-16/BX, *2 (NY City Civ filed July 14, 2017).
  • 11. Stipulation of Settlement, Beaumont, LT-021832-16/BX, *1 (NY City Civ filed Sept 7, 2016).
  • 12. Stipulation of Settlement, Beaumont, LT-021832-16/BX, *1 (NY City Civ filed Oct 26, 2016).
  • 13. See Affirmation in Opposition to Petitioner’s Motion to Restore, Beaumont, LT-021832-16/BX, *1–5 (NY City Civ filed July 18, 2017).
  • 14. Id.
  • 15. Abatement Hearing, Beaumont, LT-021832-16/BX, *1 (NY City Civ filed Jan 12, 2018).
  • 16. Id.
  • 17. Affirmation in Opposition to Petitioner’s Motion to Restore, Beaumont, LT-021832-16/BX, *1 (NY City Civ filed July 18, 2017).
  • 18. Edward H. Rabin, The Revolution in Residential Landlord-Tenant Law: Causes and Consequences, 69 Cornell L Rev 517, 521 (1984).
  • 19. David A. Super, The Rise and Fall of the Implied Warranty of Habitability, 99 Cal L Rev 389, 394 (2011). See also Paula A. Franzese, Abbott Gorin, and David J. Guzik, The Implied Warranty of Habitability Lives: Making Real the Promise of Landlord-Tenant Reform, 69 Rutgers L Rev 1, 12 (2016) (“Social justice reformers and tenants’ advocates heralded the advent of the implied warranty of habitability with great hopefulness.”).
  • 20. Cases with meritorious claims were identified based on evidence of conditions of disrepair in the unit. For a detailed description of the methodology, see Part III.D.
  • 21. According to the 2010 US Census, New York City had 2,146,892 renter households. See US Census Bureau, New York City Profile of General Population and Housing Characteristics: 2010 (2010), archived at https://perma.cc/N8WR-9Y73. Los Angeles, the next largest city in the United States, has only 814,305 renter households. See US Census Bureau, Los Angeles City Profile of General Population and Housing Characteristics: 2010 (2010), archived at https://perma.cc/8PCK-4SJS. New York City was also selected as the site for this study for a number of other reasons. See Part III.B.
  • 22. Although the warranty of habitability may be asserted by tenants affirmatively, it is generally understood that the potential of the claim lies in its use as a defense and counterclaim in nonpayment of rent eviction cases. Affirmative cases tend to involve complicated and lengthy procedural requirements, and access to counsel is limited, as legal services providers prioritize representation of tenants who are at risk of eviction. See Jessica K. Steinberg, Informal, Inquisitorial, and Accurate: An Empirical Look at a Problem-Solving Housing Court, 42 L & Soc Inquiry 1058, 1065 (2017). In eviction cases, by contrast, tenants are already in court and can simply assert the claim as a defense or counterclaim in the case. See Franzese, Gorin, and Guzik, 69 Rutgers L Rev at 2–3 n 1 (cited in note 19). See also Steinberg, 42 L & Soc Inquiry at 1064–65 (cited in note 22) (describing the problems involved in pursuing habitability claims both affirmatively and defensively).
  • 23. A possessory judgment is a judgment that grants a legal right to possession of the unit in favor of the landlord. Judgments in Nonpayment Cases (New York State Unified Court System), archived at https://perma.cc/G586-F45F. In order for a landlord to regain physical possession of a unit, the landlord must obtain a possessory judgment and must be entitled to issuance and execution of the warrant of eviction. Id. Typically, settlement agreements that include a possessory judgment also authorize the issuance of the warrant of eviction, but stay execution of the warrant to provide the tenant time to pay the rental arrears. If a tenant does not pay the arrears by the deadline included in the settlement agreement, the landlord is authorized to execute the warrant of eviction (in other words, physically evict the tenant) without appearing before a judge. New York City Landlords & Owners: Questions & Answers About Housing Court *22 (Access to Justice NY State Courts, July 2012), archived at https://perma.cc/RL65-A2GJ. Where a settlement agreement does not include a possessory judgment, the landlord must file a motion for issuance of judgment upon the tenant’s breach, and the judge must allow that motion before the landlord can proceed with the physical eviction. Id. Thus, tenants have stronger procedural protections against physical eviction where they are able to avoid possessory judgments.
  • 24. The Housing Code system in New York City has three classifications of violations: “Class A” for nonhazardous violations, such as a bathroom door that needs refitting or painting that needs to be done; “Class B” for hazardous violations, such as a defective carbon monoxide detector; and “Class C” for immediately hazardous violations, such as the lack of heat or hot water. See NYC Admin Code § 27-2001 et seq. Class A violations must be repaired within ninety days, Class B within thirty days, and Class C within twenty-four hours. NYC Admin Code § 27-2001 et seq.
  • 25. A more detailed description of the study’s methodology is provided in Part III.D.
  • 26. For a detailed description of the meaning and significance of a possessory judgment, see note 23.
  • 27. See Franzese, Gorin, and Guzik, 69 Rutgers L Rev at 22–23 (cited in note 19); Barbara Bezdek, Silence in the Court: Participation and Subordination of Poor Tenants’ Voices in Legal Process, 20 Hofstra L Rev 533, 547–48 nn 52–54 (1992); Anthony J. Fusco Jr, Nancy B. Collins, and Julian R. Birnbaum, Chicago’s Eviction Court: A Tenants’ Court of No Resort, 17 Urban L Ann 93, 109–11 (1979); Marilyn Miller Mosier and Richard A. Soble, Modern Legislation, Metropolitan Court, Miniscule Results: A Study of Detroit’s Landlord-Tenant Court, 7 U Mich J L Ref 8, 42 (1973).
  • 28. See Michele Cotton, When Judges Don’t Follow the Law: Research and Recommendations, 19 CUNY L Rev 57, 67–69 (2015) (noting that many of the fifty-nine cases studied involved serious housing code violations recorded in city inspections); Franzese, Gorin, and Guzik, 69 Rutgers L Rev at 23 n 97 (cited in note 19) (finding that among a sample of thirty-one cases studied in which the warranty of habitability was raised, it successfully led to repairs in approximately half).
  • 29. See Part II.
  • 30. Id.
  • 31. See D. James Greiner, Cassandra Wolos Pattanayak, and Jonathan Hennessy, The Limits of Unbundled Legal Assistance: A Randomized Study in a Massachusetts District Court and Prospects for the Future, 126 Harv L Rev 901, 931 (2013); Jessica K. Steinberg, In Pursuit of Justice? Case Outcomes and the Delivery of Unbundled Legal Services, 18 Georgetown J Poverty L & Pol 453, 494 (2011); Carroll Seron, Martin Frankel, and Gregg Van Ryzin, The Impact of Legal Counsel on Outcomes for Poor Tenants in New York City’s Housing Court: Results of a Randomized Experiment, 35 L & Soc Rev 419, 428 (2001).
  • 32. Franzese, Gorin, and Guzik, 69 Rutgers L Rev at 34 (cited in note 19).
  • 33. Rabin, 69 Cornell L Rev at 521 (cited in note 18) (noting the deep involvement of courts and legislatures in implementing the warranty); Franzese, Gorin, and Guzik, 69 Rutgers L Rev at 12 (cited in note 19) (noting the hopefulness of social justice and tenants’ rights advocates that the warranty would increase the habitability of residential dwellings); Mosier and Soble, 7 U Mich J L Ref at 13 (cited in note 27) (noting the need for reform recognized by commentators, courts, legislatures, and model landlord-tenant codes).
  • 34. See text accompanying notes 75–78.
  • 35. For a more detailed discussion of caveat lessee and early landlord-tenant law, see generally Thomas M. Quinn and Earl Phillips, The Law of Landlord-Tenant: A Critical Evaluation of the Past with Guidelines for the Future, 38 Fordham L Rev 225 (1969).
  • 36. See Franzese, Gorin, and Guzik, 69 Rutgers L Rev at 10–11 (cited in note 19).
  • 37. See Richard H. Chused, Saunders (a.k.a. Javins) v. First National Realty Corporation, 11 Georgetown J Poverty L & Pol 191, 197 (2004).
  • 38. Id. The early common law rules even held tenants liable for rent after the premises had been destroyed by fire or other natural disasters. See id at 197 n 18. Many state legislatures changed these rules by statute in the nineteenth century. Id.
  • 39. Id at 198 (“The basic lease—the exchange of possession for rent—was both substantively and procedurally independent from other contractual terms.”). When leases contained other covenants, those covenants were construed to be independent of each other, and thus a landlord’s violation of one covenant did not relieve a tenant of his or her obligations under another covenant. See id.
  • 40. See Mosier and Soble, 7 U Mich J L Ref at 12 (cited in note 27). Marilyn Mosier and Richard Soble also observe that in an agrarian context the dwellings conveyed were simple, and thus repairs were relatively inexpensive. Id.
  • 41. See Javins v First National Realty Corp, 428 F2d 1071, 1074 (DC Cir 1970) (citation omitted):

    [I]n the case of the modern apartment dweller, the value of the lease is that it gives him a place to live. The city dweller who seeks to lease an apartment on the third floor of a tenement has little interest in the land 30 or 40 feet below, or even in the bare right to possession within the four walls of his apartment. When American city dwellers, both rich and poor, seek “shelter” today, they seek a well known package of goods and services—a package which includes not merely walls and ceilings, but also adequate heat, light and ventilation, serviceable plumbing facilities, secure windows and doors, proper sanitation, and proper maintenance.

    Courts also recognized that landlord-tenant law had failed to keep pace with developments in contract law, where judicial interpretation has “sought to protect the legitimate expectations of the buyer and ha[s] steadily widened the seller’s responsibility for the quality of goods and services through implied warranties of fitness and merchantability.” Id at 1075.

  • 42. See id at 1078–79 (noting that “[l]ow and middle income tenants, even if they were interested in making repairs, would be unable to obtain any financing for major repairs since they have no long-term interest in the property”). Discriminatory federal housing policies severely restricted the housing options available to minority populations while at the same time facilitating white flight out of cities. The result was that minority tenants were forced into a limited supply of urban tenements, and cities became drained of their tax bases as property values plummeted. See Super, 99 Cal L Rev at 402 (cited in note 19).
  • 43. Quinn and Phillips, 38 Fordham L Rev at 225 (cited in note 35), observing that tenants lived in

    the most wretched living conditions, littered and unlit hallways, stairways with steps and banisters missing, walls and ceilings with holes, exposed wiring, broken windows, leaking pipes, stoves and refrigerators that do not work or work only now and then. And always the cockroaches, the rats, and the dread of the winter cold and uncertain heat.

    Substandard conditions can cause serious physical and emotional harm. See Super, 99 Cal L Rev at 452 (cited in note 19) (“Chipping and peeling paint at home is the dominant cause of childhood lead poisoning, which can profoundly and permanently stunt children’s intellectual and emotional development. Asthma is the leading cause of urban school absences, and roach, rodent, and mold infestation are leading causes of asthma.”) (citations omitted).

  • 44. See Mosier and Soble, 7 U Mich J L Ref at 12–13 (cited in note 27).
  • 45. Id.
  • 46. See generally Trends in Landlord-Tenant Law Including Model Code, 6 Real Prop Prob & Trust J 550 (1971). See also Quinn and Phillips, 38 Fordham L Rev at 239–49 (cited in note 35). Additionally, the warranty of habitability intended to harmonize the decline of caveat emptor in contract law with housing law. See Super, 99 Cal L Rev at 400 (cited in note 19).
  • 47. Several courts noted that the establishment of housing codes reflected the legislative reversal of the doctrine of caveat lessee. See, for example, Pines v Perssion, 111 NW2d 409, 412–13 (Wis 1961):

    [T]he legislature has made a policy judgment—that it is socially (and politically) desirable to impose these duties [of repair] on a property owner—which has rendered the old common law rule obsolete. To follow the old rule of no implied warranty of habitability in leases would, in our opinion, be inconsistent with the current legislative policy concerning housing standards.

    See also Green v Superior Court, 517 P2d 1168, 1175 (Cal 1974) (“These comprehensive housing codes affirm that, under contemporary conditions, public policy compels landlords to bear the primary responsibility for maintaining safe, clean and habitable housing.”). The development of the doctrine of constructive eviction further contributed to the erosion of caveat lessee. See Mosier and Soble, 7 U Mich J L Ref at 12 (cited in note 27). Under this doctrine, the tenant is entitled to terminate the lease by vacating the property if the premises are in such disrepair that they are unfit for human use. Upon vacating the premises, the tenant’s rental obligation ends. However, commentators at the time noted that while commercial lessees were in a position to take advantage of this development, the law was largely meaningless for residential tenants, for whom no better housing options were available if they opted to terminate their current lease. Id.

  • 48. For an excellent discussion of current limitations in the enforcement of housing codes, see generally Kathryn A. Sabbeth, (Under)enforcement of Poor Tenants’ Rights, 27 Georgetown J Poverty L & Pol 99 (2019).
  • 49. Quinn and Phillips, 38 Fordham L Rev at 240 (cited in note 35).
  • 50. See Super, 99 Cal L Rev at 414 (cited in note 19).
  • 51. Quinn and Phillips, 38 Fordham L Rev at 240–41 (cited in note 35) (noting that “[s]ending landlords to prison is not very popular” and that the moral effect of criminal liability remains small: “What about the opprobrium of a conviction? That carries about the same sting as a traffic ticket.”).
  • 52. See id at 240.
  • 53. Id at 241.
  • 54. See Super, 99 Cal L Rev at 402 (cited in note 19).
  • 55. See Javins, 428 F2d at 1079–80 (noting that “the findings by various studies of the social impact of bad housing has led to the realization that poor housing is detrimental to the whole society, not merely to the unlucky ones who must suffer the daily indignity of living in a slum”). It was also widely understood that other available mechanisms for holding landlords accountable for property maintenance were insufficient. See Quinn and Phillips, 38 Fordham L Rev at 243 (cited in note 35). The doctrine of constructive eviction allowed tenants to break their leases where landlords so badly neglected the premises that they became unlivable. Id at 242. However, tenants could only exercise this defense if they actually abandoned the building, essentially defeating the whole purpose of raising it. Id. Some jurisdictions also had rent-withholding laws, which allowed tenants to deposit their rent into escrow in court rather than pay the landlord when they experienced uninhabitable housing conditions, but commentators noted that tenants lacked bargaining power to invoke this law once their lease neared expiration. Id at 242–43. Moreover, the typical “urban ghetto tenant,” who lived in buildings in the worst condition, had tenancy rights only as a “tenant by sufferance.” Id at 243. This meant that the tenants most in need of the protection of the law lacked sufficient leverage to use rent-withholding on its own effectively. Id. In New York, § 755 of the New York Real Property Actions & Proceedings Law (RPAPL) also allowed a tenant to withhold rent for lack of services, but this section only applied (and continues to apply today) when “a serious violation against the landlord [has been] recorded by a government bureau.” Id at 245. The statute, therefore, does not help tenants with a collection of smaller issues in an apartment. Besides withholding of rent, § 302A of New York’s Multiple Dwelling Law allows for rental abatement if the landlord fails to supply services, but a tenant can only invoke § 302A after suffering from the issue for six months. Id at 247. Furthermore, a landlord can prevail at court in a § 302A action simply by repairing the major violation and allowing the smaller issues to continue—thus exposing the tenant to a defeat in court and “$100 in court costs plus the rent.” Id at 247 (citations omitted). Upon critiquing these available mechanisms, Professors Thomas Quinn and Earl Phillips proposed treating “the rent as a package containing payment components,” which would give landlords an absolute rent floor for possession and allow tenants to withhold rent above that floor for service failures. Id at 253.
  • 56. See Quinn and Phillips, 38 Fordham L Rev at 225 (cited in note 35):

    [T]he law in this area is a scandal. More often than not unjust in its preference for the cause of the landlord, it can only be described as outrageous when applied to the poor urban tenant in the multi-family dwelling. . . . Surely the law in a civilized urban society cannot tolerate such conditions. But it does! Let that be said frankly and without hedging.

    See also Super, 99 Cal L Rev at 402 (cited in note 19) (“Deteriorating housing conditions have serious negative effects on surrounding communities: they depress property values and hence property tax revenues, contribute to the spread of insect and rodent infestation, give cities a negative image with visitors, and are correlated with crime.”).

  • 57. See Mark D. Naison, The Rent Strikes in New York, in Stephen Burghardt, ed, Tenants and the Urban Housing Crisis 19, 19 (New Press 1972); Thea K. Flaum and Elizabeth C. Salzman, The Tenants’ Rights Movement 3–4, 16–18 (Urban Research Corporation 1969).
  • 58. See Mosier and Soble, 7 U Mich J L Ref at 13–14 (cited in note 27); Note, Tenant Unions: Collective Bargaining and the Low-Income Tenant, 77 Yale L J 1368, 1392 (1968). Advocates also sought to prohibit discrimination, impose rent control, limit evictions, and expand subsidies to support affordable housing development. Note, 77 Yale L J at 1370–73.
  • 59. See Super, 99 Cal L Rev at 398–99 (cited in note 19).
  • 60. Id at 401. This reciprocity was a sharp departure from long-standing common law rules that lease terms were substantively and procedurally independent from one another. See Chused, 11 Georgetown J Poverty L & Pol at 198 (cited in note 37). Under this regime, the landlord’s failure to comply with one obligation could not be used to defend a claim that the tenant breached a different obligation (such as the payment of rent). Id.
  • 61. See Super, 99 Cal L Rev at 401 (cited in note 19).
  • 62. See Franzese, Gorin, and Guzik, 69 Rutgers L Rev at 12 (cited in note 19).
  • 63. See Super, 99 Cal L Rev at 403 (cited in note 19). Professor David Super further notes that advocacy to establish the implied warranty of habitability was also grounded in “a desire to redistribute power, wealth, and income into the hands of low-income people.” Id at 402 (citations omitted).
  • 64. See id at 403 (noting further that this threat “would be much more likely to motivate landlords to make concessions to their tenants in the form of needed repairs”).
  • 65. Javins, 428 F2d at 1072.
  • 66. 428 F2d 1071 (DC Cir 1970).
  • 67. Id at 1072–73. The Court reasoned that the outdated principle that a lease conveys only a possessory interest in land “may have been reasonable in a rural, agrarian society,” but was no longer sensible “in the case of the modern apartment dweller.” Id at 1074.
  • 68. The Javins litigation arose out of a rent strike waged by poor tenants living in deplorable conditions in a low-income, minority neighborhood of Washington, DC. See Chused, 11 Georgetown J Poverty L & Pol at 206–10 (cited in note 37). The tenants had no heat for six weeks in winter and were facing a host of other conditions that the landlord was refusing to address. After a series of protests and sit-ins at government offices, none of which compelled the landlord to make repairs, twenty-nine tenants collectively organized and sent a letter to the landlord declaring that they were withholding rent until the conditions were repaired. The landlord began suing tenants for possession and won, which caused other tenants to surrender their withheld rent. Six tenants, however, continued to strike, and their eviction cases eventually became those that were taken up on appeal in Javins. For a detailed description of the events that led to the Javins litigation, see id at 194–210.
  • 69. Javins, 428 F2d at 1073. Specifically, the tenants “alleged numerous violations of the Housing Regulations as an equitable defense or [a] claim by way of recoupment or set-off in an amount equal to the rent claim, as provided in the rules of the Court of General Sessions.” Id (quotation marks omitted). The tenants claimed

    [t]hat there are approximately 1500 violations of the Housing Regulations of the District of Columbia in the building . . . where defendant resides[,] some affecting the premises of this Defendant directly, others indirectly, and all tending to establish a course of conduct of violation of the Housing Regulations to the damage of Defendants.

  • 70. See Mosier and Soble, 7 U Mich J L Ref at 10 (cited in note 27).
  • 71. Javins, 428 F2d at 1082.
  • 72. In Javins, the court held specifically that the lower court must determine “(1) whether the alleged violations existed during the period for which past due rent is claimed, and (2) what portion, if any or all, of the tenant’s obligation to pay rent was suspended by the landlord’s breach.” Id at 1082–83 (citations omitted).
  • 73. Id.
  • 74. Id.
  • 75. See Roger A. Cunningham, The New Implied and Statutory Warranties of Habitability in Residential Leases: From Contract to Status, 16 Urban L Ann 3, 6–8 (1979); Green, 517 P2d at 1181; Lund v MacArthur, 462 P2d 482, 483 (Hawaii 1969); Lemle v Breeden, 462 P2d 470, 475 (Hawaii 1969).
  • 76. Super observes that the simultaneous progression of the implied warranty of habitability through courts and legislatures was unusual as compared to other law reform initiatives. See Super, 99 Cal L Rev at 398–99 (cited in note 19).
  • 77. See Cunningham, 16 Urban L Ann at 7 nn 8–11 (cited in note 75).
  • 78. Id at 7–8 (listing statutes and cases).
  • 79. Notice is deemed to be constructive when the landlord knew or should have known about the conditions based on interactions with the property. For example, landlords are often held to have constructive notice of a condition when the condition existed at the time they purchased the property or because the condition exists in plain view and the landlord has entered the premises. See, for example, Whitney v Valentin, 963 NYS2d 109, 110 (NY App 2013).
  • 80. In jurisdictions with more progressive forms of the law, tenants also are not required to deposit withheld rent into court nor to demonstrate “good faith” withholding—any tenant who has experienced conditions of disrepair during the course of their tenancy can assert breach of the implied warranty of habitability either affirmatively in a suit against their landlord or defensively in an eviction action for nonpayment of rent. See, for example, NY Real Prop Law § 235-b. The warranty of habitability is also deemed nonwaivable. Katheryn M. Dutenhaver, Non-Waiver of the Implied Warranty of Habitability in Residential Leases, 10 Loyola U Chi L J 41, 55 (1978). In at least one jurisdiction, tenants may also assert the claim as a defense to no fault evictions. See Mass Gen Laws Ann ch 239, § 8A. In many jurisdictions, courts and legislatures adopted corollary laws prohibiting landlords from evicting tenants in retaliation for invoking their rights under the warranty of habitability. See Super, 99 Cal L Rev at 411 n 118 (cited in note 19).
  • 81. Retaliatory Eviction of Tenant for Reporting Landlord’s Violation of Law, 23 ALR5th 140 § 2[a] (1994); Mosier and Soble, 7 U Mich J L Ref at 13 (cited in note 27). The warranty of habitability is also generally considered nonwaivable, such that any effort to contract around it in the lease is void as against public policy. See Dutenhaver, 10 Loyola U Chi L J at 55 (cited in note 80).
  • 82. These reasons included a desire to harmonize landlord-tenant law with broader principles of contract and consumer protection law, recognition that the doctrine of caveat lessee was ill-fitted with the realities of modern urban living, and a questioning of the common law assumption that the land was the most important feature of a leasehold. See Javins, 428 F2d at 1077–78.
  • 83. See Super, 99 Cal L Rev at 402 (cited in note 19) (noting that some courts and legislatures “saw the implied warranty and its enforceability in actions for nonpayment of rent as a means of compelling landlords to maintain their buildings up to minimum standards of repair”).
  • 84. Javins, 428 F2d at 1079.
  • 85. Id (citations omitted).
  • 86. Green, 517 P2d at 1174. See also Karen Tokarz and Zachary Schmook, Law School Clinic and Community Legal Services Providers Collaborate to Advance the Remedy of Implied Warranty of Habitability in Missouri, 53 Wash U J L & Pol 169, 187 (2017) (observing that the implied warranty of habitability “developed, in part, as a response to a chronic and prolonged housing shortage, particularly for low-income households”).
  • 87. See Boston Housing Authority v Hemingway, 293 NE2d 831, 839–40 (Mass 1973).
  • 88. Although many advocates hoped that the implied warranty of habitability would be held constitutionally required, the US Supreme Court rejected this argument. See Lindsey v Normet, 405 US 56, 64 (1972). The Court held that federal constitutional principles of due process and equal protection do not require that a tenant be allowed to raise conditions issues as a defense to a nonpayment of rent eviction. Id at 68–69.
  • 89. See Super, 99 Cal L Rev at 425, 425 n 172 (cited in note 19). Super finds that most states have good faith requirements. See id.
  • 90. Some commentators defend these laws on the grounds that tenants should not be allowed to raise the warranty of habitability as a “legal afterthought.” See, for example, Samuel Jan Brakel, URLTA in Operation: The Oregon Experience, 5 Am Bar Found Rsrch J 565, 569 (1980).
  • 91. See Franzese, Gorin, and Guzik, 69 Rutgers L Rev at 13–14 (cited in note 19). Some jurisdictions have mandatory rent escrow requirements, in which all tenants who wish to withhold rent must deposit their rent with the court. Id. Other jurisdictions hold hearings in which judges make individualized determinations of whether rent escrow will be required based on the circumstances of the case. Id. Proponents of landlords’ protective orders (LPOs) have justified them as necessary to prevent tenants from using the implied warranty of habitability in bad faith to shirk valid rental obligations. Id at 13. Many scholars, however, criticize LPOs as creating artificial barriers to access the warranty. See, for example, id at 17–18, noting that rent escrow requirements

    put[ ] aggrieved tenants into the untenable position of having to decide whether to relocate (a task that is both disruptive and costly), or remain on site, submit to judicial proceedings, and be forced to deposit into escrow the full rent due no matter the premises’ defective condition, a task that is both onerous and counter-productive to the goal of improving stocks of decent rental housing.

    See also Cotton, 19 CUNY L Rev at 71–73 (cited in note 28).

  • 92. There are also some jurisdictions in which rent escrow orders are available only upon motion by the landlord and at the discretion of the judge. See Alaska Stat Ann § 34.03.190(a)(3); Ariz Rev Stat Ann § 33-1365(A); Hinson v Delis, 102 Cal Rptr 661, 666 (Cal App 1972); Javins, 428 F2d at 1083 n 67; Rotheimer v Arana, 892 NE2d 1183, 1194–95 (Ill App 2008); Iowa Code § 562A.24(1); Kan Stat Ann § 58-2561(a); Ky Rev Stat Ann § 383.645(1); Mass Gen Laws Ann ch 239, § 8A; Mont Code Ann § 70-24-421(1); Neb Rev Stat § 76-1428(1); Or Rev Stat § 90.370(1)(b); Pugh v Holmes, 405 A2d 897, 907 (Pa 1979); RI Gen Laws § 34-18-32(a); Teller v McCoy, 253 SE2d 114, 129–30 (W Va 1978).
  • 93. At least one appellate court, the Maryland Court of Special Appeals, has found that rent escrow requirements that apply to rental arrears (as opposed to applying only to ongoing rent that comes due after a case has been commenced) violate due process. See Lucky Ned Pepper’s Ltd v Columbia Park and Recreation Association, 494 A2d 947, 953 (Md Spec App 1985). In Lucky Ned, the court considered a state law that required the deposit of all arrears allegedly due as a condition of obtaining a jury trial. Id at 950. The court held that the law erroneously presupposed that the rent withheld was in fact owed, and therefore improperly interfered with the tenant’s right to a jury trial. Id at 951.
  • 94. Super, 99 Cal L Rev at 435 (cited in note 19).
  • 95. See id at 433 (noting that tenants may be forced to spend their rent money to mitigate the damages caused by the landlord’s failure to repair); Franzese, Gorin, and Guzik, 69 Rutgers L Rev at 36 (cited in note 19) (noting that tenants use withheld rent “to make the essential repairs themselves in view of landlord intransigence”).
  • 96. An unabated bedbug infestation, for example, will require tenants to buy new bedding and furniture.
  • 97. See Super, 99 Cal L Rev at 426 (cited in note 19).
  • 98. See id; Myrah v Campbell, 163 P3d 679, 683 (Utah App 2007).
  • 99. See, for example, Dugan v Milledge, 494 A2d 1203, 1206 (Conn 1985).
  • 100. See Super, 99 Cal L Rev at 426 (cited in note 19).
  • 101. Id at 424.
  • 102. Id at 428.
  • 103. Id.
  • 104. Super, 99 Cal L Rev at 425 (cited in note 19).
  • 105. Id. See also Brakel, 5 Am Bar Found Rsrch J at 578 (cited in note 90); 280 Broad, LLC v Adams, 2006 WL 2790909, *7 (Conn Super).
  • 106. Whether the warranty of habitability actually aids low-income tenants has also long been the subject of academic debate. See Bruce Ackerman, Regulating Slum Housing Markets on Behalf of the Poor: Of Housing Codes, Housing Subsidies and Income Redistribution Policy, 80 Yale L J 1093, 1179–81 (1971); Werner Z. Hirsch, Joel G. Hirsch, and Stephen Margolis, Regression Analysis of the Effects of Habitability Laws upon Rent: An Empirical Observation on the Ackerman-Komesar Debate, 63 Cal L Rev 1098, 1129–36 (1975); Duncan Kennedy, The Effect of the Warranty of Habitability on Low Income Housing: “Milking” and Class Violence, 15 Fla St U L Rev 485, 496 (1987); Neil K. Komesar, Return to Slumville: A Critique of the Ackerman Analysis of Housing Code Enforcement and the Poor, 82 Yale L J 1175, 1192 (1973); Rabin, 69 Cornell L Rev at 580 (cited in note 18). The “mainstream” view believes that the increased costs imposed by code requirements and the warranty of habitability are passed from landlords to tenants, thereby hurting tenants (low-income tenants especially) in the long run. See Kennedy, 15 Fla St U L Rev at 497 (cited in note 106); Rabin, 69 Cornell L Rev at 558–59 (cited in note 18). However, the overall impact of habitability regulations on housing costs varies wildly from study to study. See David Listokin and David B. Hattis, Building Codes and Housing, 8 Cityscape 21, 21 (2005) (finding that studies on the subject have claimed that building code regulations increase housing costs anywhere between 1 and 200 percent). Furthermore, some scholars—notably Professor Bruce Ackerman discussing his hypothetical town of “Slumville”—have argued that code enforcement and the warranty of habitability will help tenants without increasing their rents. See Ackerman, 80 Yale L J at 1177–86 (cited in note 106); Kennedy, 15 Fla St U L Rev at 499 (cited in note 106).
  • 107. See Mosier and Soble, 7 U Mich J L Ref at 33 (cited in note 27).
  • 108. See id.
  • 109. See id. The study found that the full rent claim was excused in 0.7 percent of contested nonpayment cases or 0.1 percent of all nonpayment cases, and it was partially excused in 11.9 percent of contested nonpayment cases or 2 percent of all nonpayment cases. However, these figures include cases in which the landlord received less than the full amount of rent claimed for reasons other than a rent abatement in satisfaction of the tenant’s implied warranty of habitability claim, including when the rent claimed had been miscalculated and when the tenant had made all or partial payment. See id.
  • 110. Fusco, Collins, and Birnbaum, 17 Urban L Ann at 109 (cited in note 27). One additional study conducted during the same time period produced similar findings. See Ben H. Logan III and John J. Sabl, Note, The Great Green Hope: The Implied Warranty of Habitability in Practice, 28 Stan L Rev 729, 744 (1976) (“During the period examined, the implied warranty of habitability was pled as an affirmative defense in 56 cases constituting 4 percent of all unlawful detainer actions and representing 27 percent of all contested unlawful detainer actions filed in that court for the 5-month period in question.”).
  • 111. It is unclear whether this sample is a statistically significant representative sample. See Bezdek, 20 Hofstra L Rev at 547 n 52 (cited in note 27). The study also involved court record review and exit interviews with litigants. Id at 547–48 n 53–54.
  • 112. See id at 554. Rent was ordered into escrow in 4.3 percent of all cases.
  • 113. See Franzese, Gorin, and Guzik, 69 Rutgers L Rev at 20 (cited in note 19).
  • 114. Id at 21.
  • 115. Id.
  • 116. Id at 22. This conclusion is based on the “far greater statistical likelihood that significant housing code violations exist on leased premises in Essex County.” Id. The authors do not state specifically what the statistical likelihood is that substandard conditions exist in the premises. See id. They cite only to HUD data on the prevalence of substandard housing conditions nationwide. Id at 5 n 11. One year later, in 2017, Professors Karen Tokarz and Zachary Schmook published the results of a study that looked broadly at outcomes in eviction cases in St. Louis, Missouri. See Tokarz and Schmook, 53 Wash U J L & Pol at 176–78 (cited in note 86). While the study did not look specifically at the frequency with which the warranty of habitability was raised, it found that only 0.03 percent of cases resulted in judgments for the tenant whereas 77.5 percent of cases resulted in judgments for the landlord (with the remainder of cases resulting in dismissal). Id at 176. Based on these findings, the authors concluded that the warranty of habitability is rarely used. Id at 186–87.
  • 117. See Cotton, 19 CUNY L Rev at 62–63 (cited in note 28).
  • 118. Id at 63.
  • 119. Id at 72. Specifically, tenants received abatements in 42 percent of cases in which they had established the elements required for this relief. Id.
  • 120. Id at 62–64 (cited in note 28).
  • 121. See Steinberg, 42 L & Soc Inquiry at 1072 (cited in note 22), noting that 

    [e]ven when a study demonstrates that one class of litigants—tenants, for example—routinely achieves unfavorable outcomes, it can be difficult to ascertain whether the poor outcomes are the result of unmeritorious claims, or are due to more problematic factors, such as lack of representation or structural unfairness within the adjudicatory process.

  • 122. The only exception is Franzese, Gorin, and Guzik’s research on the use of the warranty of habitability as a tool to compel landlords to make needed repairs. See Franzese, Gorin, and Guzik, 69 Rutgers L Rev at 24–25, 30–31 (cited in note 19).
  • 123. Mosier and Soble, 7 U Mich J L Ref at 45 (cited in note 27).
  • 124. Id at 35. Anthony Fusco, Nancy Collins, and Julian Birnbaum’s study also found that tenants who were represented by counsel achieved significantly better outcomes than unrepresented tenants. See Fusco, Collins, and Birnbaum, 17 Urban L Ann at 115 (cited in note 27).
  • 125. In addition, Professor Jessica Steinberg’s study of the impact of unbundled legal aid found that tenants who were provided with unbundled legal services were significantly more likely to raise cognizable defenses as compared with unassisted tenants. See Steinberg, 18 Georgetown J Poverty L & Pol at 494 (cited in note 31). The study did not isolate breach of warranty claims specifically. See id. Steinberg also found that full representation had a significant impact on the likelihood of the tenant receiving payments from the landlord at the conclusion of the case, while unbundled legal assistance had no positive effect on the tenant’s likelihood of receiving a payment from the landlord. See id at 486. The study did not determine whether the payment reflected a rent abatement based on the landlord’s violation of warranty of habitability or alternatively was based on some other monetary claim. Id at 486–88.
  • 126. Seron, Frankel, and Van Ryzin, 35 L & Soc Rev at 426 (cited in note 31).
  • 127. All cases included in the study population had been determined as cases in which the tenant was likely to benefit from legal support. See id at 423–24. This assessment was based on the presence of defenses and claims (beyond only the warranty of habitability), as well as nonlegal characteristics of the tenant and case. Id.
  • 128. Greiner, Pattanayak, and Hennessy, 126 Harv L Rev at 931 (cited in note 31).
  • 129. Under Massachusetts law, there are numerous counterclaims available to tenants in nonpayment of rent eviction cases which carry monetary damages. See, for example, Mass Gen Laws Ann ch 93A, §§ 2, 9; Mass Gen Laws Ann ch 186, § 14 (lessor’s obligation to furnish utilities); Mass Gen Laws Ann ch 186, § 18 (lessor’s obligation to correct unsafe conditions upon notice); Mass Gen Laws Ann ch 186, § 15B (prohibition against lessor on entering premises during lease term).
  • 130. See Greiner, Pattanayak, and Hennessy, 126 Harv L Rev at 931 (cited in note 31).
  • 131. See, for example, Mosier and Soble, 7 U Mich J L Ref at 62 (cited in note 27) (“Another reason for the insignificant effect of the legislation on Detroit tenants is that while the legislation augments a tenant’s possible defenses, it does not provide for representation of those tenants in court.”); Fusco, Collins, and Birnbaum, 17 Urban L Ann at 114–16 (cited in note 27) (emphasizing the importance of representation in determining tenant outcomes); Franzese, Gorin, and Guzik, 69 Rutgers L Rev at 31 (cited in note 19) (proposing increased access to counsel as a solution to improve the effectiveness of the warranty of habitability); Cotton, 19 CUNY L Rev at 83–84 (cited in note 28) (citing lack of access to counsel as a barrier to effective assertion of the warranty of habitability).
  • 132. See, for example, Boston Bar Association Task Force on the Civil Right to Counsel, The Importance of Representation in Eviction Cases and Homelessness Prevention *3 (Mar 2012), archived at https://perma.cc/PWL2-SWAW (determining that only 6 to 10 percent of tenants in Massachusetts are represented); Russell Engler, Connecting Self-Representation to Civil Gideon: What Existing Data Reveal About When Counsel Is Most Needed, 37 Fordham Urban L J 37, 47 n 44 (2010) (citing representation rates in multiple jurisdictions); Maya Dukmasova, New Data Reveals Impact of Being Lawyerless in Chicago Eviction Court (The Chicago Reader, Sept 14, 2017), archived at https://perma.cc/M83U-V5XE (stating that only 12 percent of tenants in Cook County are represented); Charles Allen, Kenyan R. McDuffie, and Mary M. Cheh, Low-Income Tenants in D.C. May Soon Get Legal Help (Wash Post, May 18, 2017), online at https://www.washingtonpost.com/blogs/all-opinions-are-local/wp/2017/05/18/low-income-tenants-in-d-c-may-soon-get-legal-help (visited November 20, 2019) (Perma archive unavailable) (stating that fewer than 10 percent of tenants in DC are represented). In New York City, where a right to counsel law was recently enacted, the percentage of tenants represented has risen from 1 percent before 2014 to 30 percent in in the final quarter of 2018. NYC Human Resources Administration, Office of Civil Justice, Universal Access to Legal Services: A Report on Year One of Implementation in New York City *4 (2018), archived at https://perma.cc/43MV-H2DL.
  • 133. See Franzese, Gorin, and Guzik, 69 Rutgers L Rev at 13 (cited in note 19). But see Cotton, 19 CUNY L Rev at 84 (cited in note 28) (noting that “[t]he lack of counsel means that the parties are particularly dependent on the court to ensure that the rule of law is applied”); id at 86–87 (arguing that advocates hoping to improve utilization of the implied warranty of habitability should not focus their efforts on access to counsel because the data suggest that all efforts thus far have faltered, and moreover the provision of additional lawyers would impose considerable resource demands on the courts); Bezdek, 20 Hofstra L Rev at 538 n 16 (cited in note 27) (arguing against solutions involving access to counsel because it is “parentalistic [sic] and it lets us off the hook for our parts in the charade of legal entitlement and rights vindication”).
  • 134. See Bezdek, 20 Hofstra L Rev at 538 (cited in note 27); Super, 99 Cal L Rev at 406–07 (cited in note 19); Cotton, 19 CUNY L Rev at 66 (cited in note 28) (arguing that the legalese on pleadings acts as a barrier to unrepresented tenants asserting the warranty).
  • 135. See Super, 99 Cal L Rev at 407 (cited in note 19) (drawing attention to the “little-appreciated substantive doctrines” that emerged after the law’s original enactment and arguing that they have operated as major barriers to the warranty’s effectiveness); Franzese, Gorin, and Guzik, 69 Rutgers L Rev at 20–22 (cited in note 19) (arguing that New Jersey’s rent escrow requirement is one of the primary reasons for their findings regarding the low frequency with which the warranty is raised). On paper, the rent escrow requirement in New Jersey gives trial courts the discretion to order rent be paid into escrow during the pendency of the eviction case. Franzese, Gorin, and Guzik found that in practice, however, judges treat escrow hearings with little individualized attention, and as a matter of course order rent be deposited with the court, regardless of the conditions of the premises. Id at 19–20, 37. The authors acknowledge that they do not know whether their findings regarding the presence of the rent escrow requirement and the low usage rates are correlative or causative. Id at 20. See also Tokarz and Schmook, 53 Wash U J L & Pol at 178 (cited in note 86).
  • 136. See Super, 99 Cal L Rev at 425–29 (cited in note 19).
  • 137. Id at 423–26.
  • 138. Id at 423.
  • 139. Id.
  • 140. Super, 99 Cal L Rev at 432 (cited in note 19).
  • 141. See Franzese, Gorin, and Guzik, 69 Rutgers L Rev at 36 (cited in note 19).
  • 142. Since the warranty’s initial enactment, scholars have emphasized that entrenched power differentials between landlords and tenants, along with court cultures that privilege landlords and stigmatize tenant litigants, act as significant barriers to the law’s effectiveness. See Bedzek, 20 Hofstra L Rev at 571–72, 568 (cited in note 27) (observing that in Baltimore, “the formal allocation of responsibilities between landlord and tenant is effectively overwritten by the ‘tenant as deadbeat’ subtext which is reiterated by the court on behalf of the class of landlord litigants,” and arguing that “[i]n a jurisdiction with a functioning warranty of habitability, the subtext in tenant-claiming cases would be: it is the landlord who has done wrong by failing to fulfill societally recognized obligations”); Cotton, 19 CUNY L Rev at 85 (cited in note 28) (“It may also be the case that any uncertainty about the law that results in an environment of limited appellate guidance will be resolved against the less powerful party in the litigation, which in this situation is the tenant.”); Super, 99 Cal L Rev at 451 (cited in note 19) (“[E]ither abandoning or destabilizing courthouse culture could have resulted in much broader application of the implied warranty.”); Mosier and Soble, 7 U Mich J L Ref at 63 (cited in note 27):

    The disparities in help given to landlords and tenants and the treatment of late landlords and tenants are an indication of the perhaps inevitable bias of the court toward the landlord. Most of the judges and court personnel have a middle-class background, and they have become familiar with many landlords and attorneys appearing regularly in the court. The court had years of experience as a vehicle for rent collection and eviction where no defenses could be raised.

    Scholars have also highlighted the constraints that judges face in enforcing the laws. Judges have large numbers of cases on their dockets and lack access to important fact-finding tools and resources. See Cotton, 19 CUNY L Rev at 85–86 (cited in note 28).

  • 143. See Super, 99 Cal L Rev at 408 (cited in note 19).
  • 144. See Franzese, Gorin, and Guzik, 69 Rutgers L Rev at 22 (cited in note 19).
  • 145. Id.
  • 146. Id at 36, 27 n 106 (noting that New York City has a Housing Code violation database that is publicly available online and that the Housing Court provides a computer on each judge’s bench). An even more robust technology solution was urged by Professor Mary Marsh Zulack nearly a decade prior. See Mary Marsh Zulack, If You Prompt Them, They Will Rule: The Warranty of Habitability Meets New Court Information Systems, 40 John Marshall L Rev 425, 449–53 (2007). Specifically, Zulack proposed a computerized system that would “prompt judges through repair-related information gathering, retrieval, and adjudication steps.” Id at 425. Zulack predicted that such a system would “lead[ ] efficiently to outcomes that link the application of the warranty of habitability doctrine to real world improvements in rental premises.” Id.
  • 147. New York City is also the nation’s largest rental market and one notorious for substandard housing conditions. See note 21; Grace Ashford, Leaks, Mold and Rats: Why New York City Goes Easy on Its Worst Landlords (NY Times, Dec 26, 2018), archived at https://perma.cc/WVS7-6URQ.
  • 148. The objective here is not to determine whether the outcome was “just,” but whether tenants who appeared to have meritorious claims received the benefits the law affords for those claims. See Paula Hannaford-Agor and Nicole Mott, Research on Self-Represented Litigation: Preliminary Results and Methodological Considerations, 24 Just Sys J 163, 178 (2003) (noting that “whether the litigant received a just or appropriate outcome” is “one of the most difficult questions for which to formulate accurate and reliable measures for empirical analysis”).
  • 149. The data in this study showed that proper assertion of the warranty of habitability as a claim in the tenant’s answer was largely insignificant as a factor predicting whether the claim was used successfully. Approximately half of the tenants who received rent abatements never actually asserted the claim. This finding is consistent with what one would expect given liberal pleading amendment rules. These rules have the effect of making actual amendments unnecessary in proceedings that usually resolve in relatively expeditious out-of-court settlements, such as eviction proceedings, where it is understood that the party could receive the amendment if leave was sought, and thus to avoid unnecessary litigation the parties treat the pleadings as if they were amended without actually going through the judicial procedures to do so.
  • 150. See Part II.C.
  • 151. See Franzese, Gorin, and Guzik, 69 Rutgers L Rev at 20–22 (cited in note 19) (citing to New Jersey’s rent escrow requirement as one of the primary reasons for their findings regarding the infrequency with which the warranty is raised); Super, 99 Cal L Rev at 432 (cited in note 19) (concluding that rent escrow laws “likely are a significant contributor to the low rate of relief granted [for breach of the warranty of habitability] to low-income tenants”); id at 441 (arguing that good faith requirements may make tenants incapable of pursuing warranty of habitability claims).
  • 152. For a detailed description of the meaning and significance of a possessory judgment, see note 23.
  • 153. See NY Real Prop Law § 235-b.
  • 154. See Chapman v Silber, 760 NE2d 329, 334 (NY 2001) (stating that notice is adequate if the landlord “reserves the right to enter in order to inspect or to make [ ] repairs”).
  • 155. See Franzese, Gorin, and Guzik, 69 Rutgers L Rev at 19, 22, 36, 38 (cited in note 19).
  • 156. See NY Real Prop Law § 235-b. This legislation followed a New York Appellate Division case, Tonetti v Penati, 367 NYS2d 804 (NY App 1975), which laid the initial groundwork for a warranty of habitability in New York. In Tonetti, a tenant argued that he should be entitled to the return of his security deposit—even though he left an apartment many months before the expiration of his lease—due to the overpowering stench of dog urine. Id at 805. The Appellate Division agreed. The Tonetti court held, “It is evident that the rationale behind the common-law rule, which likened a lease to the sale of a chattel and therefore applied the ancient doctrine of caveat emptor, has no rational basis in a modern, urban society.” Id at 807. Senate Bill 3331B, which passed and later became codified as New York’s Real Property Law § 235-b, represented a direct response to the case. See Kaplan v Coulston, 381 NYS2d 634, 635 (NY City Civ 1976).
  • 157. NY Real Prop Law § 235-b(1). This provision has been interpreted to impose repair obligations on landlords where premises are not “fit for their intended purposes,” and tenants have been “subjected to conditions which are dangerous to their life, health, and safety.” See K.E.V. Realty Co, Inc v Kelly, NY L J 26, 27 (NY City Civ May 31, 1996).
  • 158. See NY Real Prop Law § 235-b(2), (3)(a). Section 235-b(3)(b) provides that if the failure to repair is caused due to a labor strike, and the landlord has made a good-faith effort to cure the conditions, then the tenant cannot recover damages. Section 235-b(3)(c) is designed to avoid double recovery for tenants in already-protected housing. Specifically, this section limits the recovery of tenants in housing subject to rent stabilization, rent control, the “emergency tenant protection act of nineteen seventy-four [1974],” or “the city rent and rehabilitation law.” NY Real Prop Law § 235-b(3)(c). The section states that if a tenant living in one of these types of housing receives a rent reduction from the New York State Division of Housing and Community Renewal (DHCR), then the amount a tenant recovers due to a landlord’s breach of the warranty of habitability must be reduced by the amount of this rent reduction. NY Real Prop Law § 235-b(3)(c).
  • 159. Kaplan, 381 NYS2d at 635.
  • 160. Ocean Rock Associates v Cruz, 411 NYS2d 663, 663 (NY App 1978). The tenant must allow the landlord to enter the premises to make repairs; a tenant’s refusal to allow access provides a defense for landlords to damages for breach of the warranty of habitability. Fifty-Seven Associates, LP v Feinman, 2011 WL 749255, *1 (NY Sup). However, a landlord cannot merely assert a good faith defense by attempting (and failing) to cure: because the warranty of habitability reflects a contractual obligation, courts interpret the breach strictly. Joseph v Varna Trust, NY L J 32 (NY City Civ Feb 13, 2003).
  • 161. Tenants in New York City generally do not bring affirmative warranty of habitability claims where they face conditions of disrepair; they instead bring Housing Part (HP) actions. See Dennis E. Milton, Comment, The New York City Housing Part: New Remedy for an Old Dilemma, 3 Fordham Urban L J 267, 270 (1975). A designated section of the Housing Court adjudicates HP actions. Any time a landlord violates or appears to have violated New York City’s Housing Maintenance Code or the New York City Civil Court Act, a tenant can initiate an HP action. Id.
  • 162. See NYC Human Resources Administration, NYC Office of Civil Justice 2017 Annual Report and Strategic Plan *19, archived at https://perma.cc/CYR4-A3MD. In 2016, the year this study was conducted, there were 202,300 nonpayment cases filed. The number of nonpayment cases filed has steadily decreased since 2013. Eviction cases brought for reasons other than nonpayment of rent, such as termination of the tenancy or violation of the lease, are considered “holdover[s].” In 2016, there were 31,584 holdover cases filed. A total of 22,089 eviction cases resulted in actual eviction that year, but the percentage breakdown between holdovers and nonpayment cases is unknown. See id at *19–20.
  • 163. In this study, less than 1 percent of nonpayment of rent evictions went to trial. For a discussion of the widespread practice across jurisdictions of resolving eviction cases through “hallway negotiations,” see Engler, 37 Fordham Urban L J at 47 (cited in note 132).
  • 164. The data in this study showed that 22 percent of all nonpayment cases in which the tenant appeared were resolved through a settlement agreement in which the landlord agreed to discontinue the case (presumably because all the arrears had been paid or otherwise accounted for). One percent of cases resulted in settlement agreements in which the tenant agreed to move out, 0.5 percent resulted in dismissal (presumably because of a procedural or other type of defect), and 8 percent resulted in a default judgment. Cases that resulted in a discontinuance, move out agreement, or default judgment were excluded from the analysis unless otherwise indicated.
  • 165. The size of a rent abatement is measured by the diminution of value of the premises, in other words, by calculating the difference between the value of the premises in good repair and the value of the premises in their defective condition. This difference is then multiplied by the length of time for which the defective conditions existed, from the time of notice to the landlord to the time of repair. Rent abatements may also be awarded at an abatement hearing held by a judge prior to the full trial. Because very few cases go to trial, few abatement hearings are held. All abatements awarded after a hearing were included in the data coding, analysis, and results. The amount of arrears claimed by the landlord may also be reduced for other reasons such as improper rental overcharges, the attribution of arrears to a public housing authority responsible for making Section 8 payments, or for other monetary claims asserted by the tenant.
  • 166. The rent abatement and its purpose (to satisfy the tenant’s warranty of habitability claims) are almost always expressly stated because landlords want to ensure that tenants cannot seek to recover on the claims again in a subsequent settlement agreement in the same case, or in a separate court proceeding. To check for the possibility that settlement agreements included “hidden” abatements, I also coded for two settlement outcomes that could be equivalent to a rent abatement: a promise to pay rental arrears in an amount of $7,000, $9,000, or $11,000, and agreements that prospectively set the rent. In 2016, $7,000 $9,000, and $11,000 were the maximum arrears amounts that City voucher programs (respectively) would pay when granting a tenant a new voucher. A tenant who is facing eviction for nonpayment of rent and cannot afford the rent going forward may apply for and receive the voucher if the amount of rent he or she owes is no greater than the particular amount ($7,000, $9,000, or $11,000, depending on the program). When the City grants the voucher, it will also pay off those arrears. The possibility thus exists that instead of granting a rent abatement explicitly, a landlord may satisfy the tenant’s warranty claims by agreeing to reduce the total arrears to either $7,000, $9,000, or $11,000 so that the tenant can qualify for the voucher. Similarly, a landlord may satisfy the tenant’s claims by agreeing to a future rent amount that is lower than the legal rent (known as a “preferential rent”) or the rent he or she would otherwise charge (if unregulated), rather than expressly granting an abatement. However, the coding revealed that both of these outcomes were extremely rare—they existed in the cases of less than 1 percent of tenants with meritorious warranty claims. Because the outcomes were so rare and it remains ambiguous whether they even truly reflect a rent abatement awarded for the tenant’s warranty of habitability claims—landlords could have other reasons for wanting their tenants accepted into the voucher program or for setting a prospective rent in the settlement agreement—I did not count these cases as having rent abatements for the purpose of the data analysis.
  • 167. It is generally understood that these latter two outcomes—amount of time to pay and whether a judgment issues—operate in an inverse relationship in negotiations. Thus, the landlord will agree to a stipulation without a judgment and a shorter period of time to pay the arrears, or a stipulation with a judgment and a longer period of time to pay.
  • 168. Where a tenant fails to pay by the payment deadline and the stipulation includes a judgment, the tenant will file a post-judgment “Order to Show Cause” seeking a stay in the execution of the eviction. See Orders to Show Cause (New York State Unified Court System), archived at https://perma.cc/E63G-MSDR. Orders to Show Cause are liberally granted, and thus landlords tend to agree to a settlement allowing for a new deadline for the payment of the arrears. Where the original settlement stipulation does not include a judgment, the landlord will file a motion for issuance of the judgment and the execution upon the tenant’s failure to pay by the payment deadline. Such a motion will also typically resolve in a subsequent settlement stipulation, this time including a judgment, with a new payment deadline. These subsequent settlement stipulations are allocated in the same manner as initial settlement stipulations, and thus will include provisions requiring the performance of repairs with the same regularity.
  • 169. There are two general standards for the granting of orders to show cause in New York City Housing Court. First, if the order to show cause will grant merely a stay of execution for an eviction, there is wide judicial discretion in determining whether or not to grant the order—the court will grant the order if that is determined to be “just.” See NY CPLR § 2201. See also Joseph v Cheeseboro, 248 NYS2d 969, 971 (NY City Civ 1964) (stating that the standard for granting such orders is “the court’s own sense of discretion, prudence, and justice”), revd on other grounds, 251 NYS2d 975 (NY Sup 1964). However, if the order to show cause will lead to vacatur of the judgment for eviction, a different standard prevails. In such cases (which generally result from a default judgment against the tenant), the party bringing the order to show cause must show that the default was “excusable default.” See NY CPLR § 5015(a)(1). A showing of excusable default has two components that the tenant must show: “a reasonable excuse for defaulting and a meritorious defense to the proceeding.” East 168th Street Associates v Castillo, 79 NYS3d 485, 489 (NY City Civ 2018).
  • 170. The pro se answer form is a checkbox form that tenants complete orally at the Housing Court clerk’s window. The form asks tenants whether “[t]here are or were conditions in the apartment and/or the building/house which the Petitioner did not repair and/or services which the Petitioner did not provide.” Civil Court of the City of New York, Answer in Writing and Verification (Form CIV-LT-91b) (May 2013), archived at https://perma.cc/L72Z-ERBH. This plain language wording is distinct from the legalese often used in pro se pleading forms in other jurisdictions. See Cotton, 19 CUNY L Rev at 66 (cited in note 28) (noting that the pro se pleading form in other jurisdictions asks tenants to “state whether they want relief based on violation of the ‘warranty of habitability’ and the ‘covenant of quiet enjoyment,’ terms which have no meaning to these tenants or even most lay people”). The pro se answer form used in New York City Housing Court does not provide space for tenants to specify which repairs are needed. Thus, as described in the text accompanying note 191, cases are never identified as having a meritorious warranty of habitability claim based solely on the assertion of needed repairs in the Answer.
  • 171. An allocution is a judge’s review of the stipulation with an unrepresented party to ensure that the party enters into the stipulation freely and voluntarily and understands the terms to which he or she is agreeing. Because questions about repairs are part of judges’ standardized allocutions, many landlord attorneys will ask tenants if repairs are needed and will include repair obligations in the stipulation voluntarily.
  • 172. See, for example, Judicial Request/Order for Housing Inspection, Beaumont Management Group, LLC v Jackson, LT-021832-16/BX, *5 (NY City Civ 2016).
  • 173. 2016 was the most recent year for which complete case data was available during the time period this study was conducted (May–October 2018). Many cases filed in 2017, particularly those filed in the latter half of the year, were still ongoing in 2018.
  • 174. A tenant appears by filing an Answer at the Housing Court clerk’s office. Cases in which the tenant defaulted were excluded because a default judgment generally precludes the tenant from asserting claims and defenses. Even where a tenant is successful in removing a default judgment at a later stage in the case, the tenant typically negotiates at a weakened bargaining position and thus does not have the same leverage to invoke the warranty of habitability. See Franzese, Gorin, and Guzik, 69 Rutgers L Rev at 21 (cited in note 19) (“The entry of a default judgment against a tenant who does not (or cannot) appear in court limits that tenant’s range of options and all but closes the window of opportunity for consideration of viable defenses and alternatives to dispossession.”). Inclusion of cases with default judgments in the study would have muddied the data, causing the findings to reflect both the structural barriers to usage andthe lack of availability of the claim due to the default. Since the goal of the study was to assess the structural barriers to usage, defaulted cases were excluded.
  • 175. The NYU Furman Center was provided this database by the Office of Court Administration pursuant to a data use agreement that restricts usage to certain research purposes.
  • 176. See note 162.
  • 177. This dataset also included other information; however, the only data used for this study were the index number, case type, and appearance of the tenant. This data was used only to determine the size of the total study population and to identify a random representative sample of cases.
  • 178. A total of 202,300 nonpayment of rent eviction petitions were filed in 2016. See note 162. Thus, the tenant defaulted in over half of all the nonpayment proceedings.
  • 179. A stratified sample is one that is proportional to certain differentiating criteria. Thus here, the number of cases from each borough in the sample was proportional to the number of cases from that borough in the total dataset. The sample was a 0.5 percent stratified sample.
  • 180. The margin of error states the amount of random sampling error in a study’s results. The confidence interval is a type of interval estimate that might contain the true value of an unknown population parameter. The associated confidence level quantifies the level of confidence that the parameter lies in the interval. The response distribution is the probability distribution of the response (target) variable. Fifty percent is the most conservative choice for the response distribution, yielding the largest sample size.
  • 181. “Hazardous” Housing Code violations are classified as “Class B” level violations and “immediately hazardous” violations are classified as “Class C” level violations. See note 24.
  • 182. The Office of Court Administration dataset included the unit-level address for each case filed. For each Housing Code violation, the HPD dataset included the unit-level address, the dates the violation was open and closed, and the violation classification level (A, B, or C). The HPD data did not include information for violations at properties owned by the New York City Housing Authority (NYCHA)—in other words, public housing—and thus the matched dataset used for this study was not inclusive of nor can it be taken to reflect outcomes involving NYCHA units.
  • 183. Representation status was coded based on whether the tenant was represented when he or she entered into the first settlement agreement in the case.
  • 184. These outcomes were only recorded for the first settlement agreement because this agreement reflects what is generally the only substantive negotiation in the case. A subsequent agreement (other than a discontinuance) will only occur if a tenant has defaulted on the first agreement, and thus a tenant in that posture is in a weakened negotiating position. A tenant in that posture will also typically have waived defenses and claims in the first agreement, particularly if judgment has entered.
  • 185. The pro se Answer form provides an option for tenants to assert that repairs are needed in their apartments. See note 170. The form does not prompt tenants to specify which repairs are needed. Settlement agreements, by contrast, nearly always specify the repairs to be performed where they require repairs.
  • 186. Repairs of all conditions issues that qualify as rent impairing pursuant to NY Multiple Dwelling Law § 302-a were included as “substantial repairs.” All conditions that have been found to constitute a violation of the warranty of habitability were also included. These include, inter alia: lack of heat and/or hot water, see Parker 72nd Associates v Isaacs, 436 NYS2d 542, 544 (NY City Civ 1980); flooding, see Spatz v Axelrod Management Co, Inc, 630 NYS2d 461, 463–64 (NY City Civ 1995); fumes and smoke, see Goldman v O’Brien, NY L J 28 (NY Sup Aug 14, 2000); leaking gas, see Goodman v Ramirez, 420 NYS2d 185, 188 (NY City Civ 1979); lead paint, see Chase v Pistolese, 739 NYS2d 250, 252–53 (NY City Ct 2002); bedbugs, see Jefferson House Associates, LLC v Boyle, 2005 WL 465171, *3 (NY Just Ct); mold, see 360 West 51st Street v Cornell, NY L J 28 (NY Civ Ct Sept 6, 2005), affd, 831 NYS2d 634, 635 (NY Sup 2007); broken appliances (for example, refrigerator or stove), see Rosewohl Enterprises, LLC v Schiffer, 2006 WL 1981750, *1 (NY Sup); cockroaches, see 501 New York LLC v Anekwe, 2006 WL 3859077, *1 (NY Sup); secondhand smoke, see Poyck v Bryant, 820 NYS2d 774, 777 (NY City Civ 2006); mice and/or rats, see Northwood Village, Inc v Curet, NY L J 34 (NY Dist May 6, 1998); noise and/or dust, see Mantica R Corp NV v Malone, 436 NYS2d 797, 800 (NY City Civ 1981); failure to install kitchen facilities, see Varna Trust, NY L J at 32; and broken locks, see Jangla Realty Co v Gravagna, 447 NYS2d 338, 341 (NY City Civ 1981).
  • 187. The tenant had the ability to pursue claims and defenses in these cases because the tenant filed an Answer. A tenant who does not file an Answer defaults and, in most instances, will receive a default judgment. Although it is possible to defend a case after receiving a default judgment, a tenant in this posture will not have the same opportunity to pursue claims and defenses as a tenant who appears. See note 168.
  • 188. Some cases did not fall into either classification because it was ambiguous whether the tenant had a meritorious warranty of habitability claim. These cases were excluded from the analysis.
  • 189. The available information, however, did not provide insight into whether the tenant had suffered conditions of disrepair sufficient to constitute a violation of the warranty of habitability at an earlier time in his or her tenancy. Thus, there may have been some cases included in the comparison group that were cases in which the tenant had the ability to pursue a warranty of habitability claim.
  • 190. All three conditions were required to be met for a case to be assigned to the comparison group. Cases in which needed repairs were asserted in the Answer but in which substantial repairs were not included in the settlement agreement were not included in either group because it was ambiguous whether the tenant had a meritorious warranty of habitability claim. These cases were excluded from the analysis.
  • 191. It is unknown to what extent the tenants’ assertions may have been untruthful—tenants could have, for example, invoked the claim without basis because they believed it would bolster their defense. To assess for this possibility, I compared the frequency with which tenants asserted needed repairs in their Answer with the frequency with which tenants claimed a service defect, which was offered as another checkbox option on the standardized form. A service defect is in some ways a stronger defense to an eviction case than a warranty of habitability claim—where a tenant has not been properly served, the court has no jurisdiction and the case must be dismissed. Yet only 10 percent of tenants claimed this defense. This finding suggests that tenants were not simply checking every box that could be beneficial to their case, and thus supports the truthfulness of tenants’ assertions of needed repairs. Moreover, research in other jurisdictions has found that tenants’ allegations of conditions of disrepair are generally valid. In a longitudinal study of seventy-three landlord-tenant cases in a housing court in Washington, DC, Professor Steinberg found that “98 percent of [landlords] later subject to housing inspections were deemed responsible for at least one housing code violation.” Steinberg, 42 L & Soc Inquiry at 1079 (cited in note 22). The primary purpose of this housing court, known as the Housing Conditions Court, is to address substandard housing. Id at 1064–66.
  • 192. The average length of time between the Answer and the settlement agreement was twenty-one days.
  • 193. This figure is likely relatively low in part because many cases do not involve multiple settlement agreements.
  • 194. Cases were only included in the “meritorious claim” group where the conditions requiring repairs, as stated in the settlement stipulation, were sufficient to constitute a warranty of habitability violation. Thus, where a settlement stipulation required a landlord to repair only a minor condition that did not affect habitability, the case was not included in the “likely meritorious warranty claim” group.
  • 195. The first group includes all cases in which it was likely that the tenant had a meritorious warranty of habitability claim. Virtually all cases (over 99 percent of cases in the “all nonpayment of rent eviction cases” dataset) result in a settlement stipulation, and the inclusion of substantial repairs in the stipulation likely indicates that the tenant had a meritorious claim. However, there is a possibility that the tenant was lying by saying repairs were needed, or that perhaps the tenant had not notified that repairs were needed prior to the settlement discussion. Thus, this definition could be overly inclusive by encompassing cases in which the tenant did not have a meritorious claim. The second group includes cases in which there was a near certainty that the tenant had a meritorious claim. If the tenant stated that repairs were needed in his or her Answer and the landlord agreed to make substantial repairs in not one but two or more settlement stipulations, we know that the landlord had notice of the conditions and failed to make repairs. Moreover, the tenant’s persistence in asserting the conditions and the need for repairs suggests a low probability of falsification. However, the use of this definition is likely to exclude cases in which the tenant has a meritorious claim. Many cases resolve with only one settlement stipulation, and it is possible that some tenants are not asked or do not know to mention that repairs are needed when they file their Answer.
  • 196. The goal of using these criteria was to identify cases in which the landlord appeared to have shirked his or her obligations to repair in the first agreement. Where the landlord had shirked such obligations, there is a strong likelihood that the tenant had a meritorious warranty of habitability claim because the landlord was on notice and failed to make the necessary repairs. It is unknown in these cases, however, if the failure to repair was the result of the tenant’s refusal to provide access.
  • 197. Conditions that qualify as Class C violations include, inter alia, rodents and inadequate supply of heat or hot water. Conditions that qualify as Class B violations include, inter alia, inoperable smoke detectors, mold, and vermin issues. Ninety-five percent of the cases included in the violation dataset had at least one open Class C violation.
  • 198. In order for a violation to be closed (often referred to as “certified”), there must be a determination that the violation has been corrected. Prior to the deadline for correcting the violation (twenty-four hours for a Class C violation, thirty days for a Class B violation, and ninety days for a Class A violation), a landlord may self-certify the violation as corrected by mail or through an online system. Once the deadline for correction of the violation has passed, a landlord must submit a dismissal request to the Code enforcement agency (the Department of Housing Preservation and Development, or “HPD”). Upon the filing of a dismissal request, an inspection will be conducted and the housing inspector will deem the violation corrected when so warranted. Certain violations require the submission of documentation along with the request for dismissal. Where a violation has been open for longer than twelve months and no new violations have been issued during that time period, the landlord can apply for a voluntary reissuance of the violation and may then self-certify the violation as corrected by the newly established deadline for correction. See New York City Department of Housing Preservation & Development, Violation Removal—Overdue Violations *4–10 (Mar 2017), archived at https://perma.cc/8UME-89DQ. It is possible that in some cases included in this dataset, the violation had been corrected but the landlord had not yet undertaken the appropriate procedures to close the violation. It is also possible that there were some cases with uncorrected Class B and Class C violations at the time of case filing that were not included in the dataset because the landlord had falsely certified the violations as corrected. See generally Ashford, Leaks, Mold and Rats (cited in note 147) (reporting instances of false correction certifications by landlords).
  • 199. See generally New Settlement Apartments’ Community Action for Safe Apartments (CASA) and the Community Development Project (CDP) at the Urban Justice Center,Tipping the Scales: A Report of Tenant Experiences in Bronx Housing Court (Mar 2013), archived at https://perma.cc/MB6A-BN7M. Oral or written notice to the landlord is sufficient to satisfy the notice requirement of a warranty of habitability claim. See note 154.
  • 200. Repairs were asserted in the Answer in 71 percent of violation cases, and substantial repairs were included in the settlement agreement in 68 percent of violation cases. In 19 percent of violation cases, substantial repairs were included in multiple settlement agreements, and in 16 percent of violation cases, substantial repairs were included in multiple settlement agreements and repairs were asserted in the Answer.
  • 201. For an explanation of how I coded the awarding of a rent abatement, see the Appendix.
  • 202. See Franzese, Gorin, and Guzik, 69 Rutgers L Rev at 12 (cited in note 19); notes 44–53 and accompanying text.
  • 203. For a detailed description of the meaning and significance of a possessory judgment, see note 23.
  • 204. Tenants were slightly less likely to receive possessory judgments in cases in which there were open code violations, but this finding was not statistically significant.
  • 205. The length of repayment period is compared separately for cases with and without possessory judgments because these two outcomes are typically negotiated in an inverse relationship with each other—tenants who wish to avoid a judgment can typically do so in exchange for a shorter repayment period, whereas tenants who prefer a longer repayment period can typically achieve this by agreement to a possessory judgment. See note 167.
  • 206. In a significant share of cases (22 percent), the settlement agreement was an agreement to discontinue the case (a “discontinuance”) rather than a repayment agreement. A discontinuance generally results where the tenant has paid the entirety of the rent owed. The likelihood of a tenant receiving a discontinuance did not appear to be affected by the presence of a warranty of habitability claim. In fact, the likelihood of receiving a discontinuance was lower among tenants who appeared more likely to have meritorious warranty of habitability claims as compared with tenants who did not. The discontinuance rate among tenants with repairs asserted in their Answer was 20 percent, compared with 25 percent among tenants without repairs asserted in the Answer. The discontinuance rate in all violation cases was 19 percent. Among tenants with substantial repairs asserted in their settlement agreement, the discontinuance rate was 13 percent as compared with 35 percent among tenants with no repairs included in their settlement agreement. The latter disparity—and the low discontinuance rate when repairs were included in the settlement in particular—may exist because judges do not consistently perform allocutions of the settlement agreement where the agreement is a discontinuance. Thus, many tenants who needed substantial repairs may not have had the opportunity to include those repairs in their settlement. Nevertheless, the comparison among cases with and without repairs asserted in the Answer and violation cases indicates that tenants with likely warranty of habitability claims did not appear to be using their claims to achieve discontinuances.
  • 207. The p-value, or probability value of asymptotic significance, indicates the level of statistical significance of the outcome. P-values less than or equal to 0.05 indicate statistical significance, whereas p-values greater than 0.05 indicate that the outcome is not statistically significant.
  • 208. It is also possible that the difference in the rate of possessory judgments is attributable to differences in preferences between tenants with code violations and those with no conditions of disrepair. To the extent tenants with code violations are genuinely withholding rent and have saved the money, they may be more likely to prefer an outcome comprised of a shorter repayment period and no possessory judgment rather than one comprised of a longer repayment period and the award of a possessory judgment.
  • 209. Settlement agreements in cases that were converted to holdovers were excluded from this analysis.
  • 210. In theory, a case could also have multiple settlement agreements because the landlord failed to make the ordered repairs and the tenant brought the case back to court on that basis. However, virtually none of the cases included in either sample involved multiple settlement agreements for this reason. It is also unlikely that there is a selection bias such that cases that have multiple settlement agreements are more likely to be those in which the landlord failed to make repairs. Obligations in settlement agreements for a tenant to repay arrears and for a landlord to make repairs are not construed as mutually dependent, and thus a landlord’s failure to make repairs is not grounds for the tenant to fail to comply with his or her repayment obligations. Thus, a tenant electing to not make her arrears payment because of the landlord’s failure to comply with its repair obligations would essentially be subjecting herself to the possibility of immediate eviction without the benefit of the warranty of habitability as a defense.
  • 211. If the initial settlement agreement includes a possessory judgment for the landlord, the landlord’s next step toward eviction will be to issue a warrant of eviction. The tenant will then have to file an order to show cause to bring the case back to court. If the initial settlement agreement does not include a possessory judgment for the landlord, the landlord’s next step toward eviction will be to file a motion in court seeking a judgment and issuance of a warrant of eviction. This motion will bring the case back to court.
  • 212. In some cases, either or both of the parties will choose to go before the judge for a hearing rather than enter into a new settlement agreement. Such cases were excluded from the analysis described in this Section.
  • 213. It is possible that tenants are lying and saying that repairs are still needed after the repairs have already been completed. However, there does not appear to be an incentive for tenants to make such a misrepresentation. Tenants are not excused from their repayment obligations, nor do they receive any other direct benefits, as a result of the landlord’s failure to comply with the repair order.
  • 214. See Steinberg, 42 L & Soc Inquiry at 1083–84 (cited in note 22) (highlighting the “housing inspector’s central role in prompting landlords to repair housing code violations,” in part through their roles as fact finders, in the context of the Housing Conditions Court in Washington, DC).
  • 215. Although the form judges use to solicit a Housing Code inspection of a unit is termed a “Judicial Request/Order for Housing Inspection,” in practice an inspection is always scheduled once a judge completes the form. Thus, I describe this authority as the authority to “order” a Code inspection, although technically speaking the authority is to “order or request” a Code inspection. See note 172.
  • 216. If a landlord does not make a repair as ordered in an eviction case settlement agreement, the tenant must bring the failure to repair to the court’s attention for a judge to enforce the order. There are many barriers to the effectiveness of this enforcement mechanism: tenants may not be aware of or know how to bring the failure to the court’s attention; tenants may fear retaliation if they choose to bring the failure to the court’s attention; and tenants may not want to risk keeping their eviction case open (by bringing the failure to the court’s attention) if they continue to owe the landlord rent.
  • 217. See Franzese, Gorin, and Guzik, 69 Rutgers L Rev at 27 (cited in note 19).
  • 218. It is not possible to tell from the data the extent to which repairs are not performed because the tenant does not provide access on the agreed upon dates.
  • 219. Records of Housing Code violations are accessible through a centralized public online database.
  • 220. This outcome was not measured in the “all nonpayment of rent cases” dataset because very few of those cases had open code violations in the unit, and thus it would have been difficult to interpret the meaning of the rate there.
  • 221. These findings should be interpreted with some caution, as this study did not involve the randomized assignment of representation. The cases compared have equally strong evidence of warranty of habitability claims; however, it is possible that there are other factors that led counsel to accept some cases and not others. For example, counsel may have selected cases based on the presence of other claims and defenses, or because of the willingness of the tenant to participate in the case. However, to the extent these factors affected the selection of cases for representation, the results are likely biased upward so as to overestimate the impact of legal representation.
  • 222. The length of repayment periods and the rate of possessory judgments were not compared because the sample size among represented tenants was too small to obtain results with statistical significance.
  • 223. It is unknown to what extent the substantial repairs needed in the represented versus unrepresented cases were equivalent. Thus, it is possible that counsel were selecting for cases with more serious needed repairs, or for cases where more evidence existed documenting the severity of the repairs and notice to the landlord.
  • 224. Specifically, substantial repairs were included in the settlement agreement in 51 percent of unrepresented cases and 53 percent of represented cases.
  • 225. Where repairs were asserted in the Answer, the abatement rate was 22 percent for represented tenants compared with 1 percent for unrepresented tenants.
  • 226. See Franzese, Gorin, and Guzik, 69 Rutgers L Rev at 34 (cited in note 19); Super, 99 Cal L Rev at 434–36 (cited in note 19); Tokarz and Schmook, 53 Wash U J L & Pol at 178 (cited in note 86).
  • 227. See Super, 99 Cal L Rev at 458–60 (cited in note 19); Franzese, Gorin, and Guzik, 69 Rutgers L Rev at 31–34 (cited in note 19).
  • 228. See Part IV.D.
  • 229. See Vicki Been, et al, Implementing New York City’s Universal Access to Counsel Program: Lessons for Other Jurisdictions *2 (NYU Furman Center, Dec 11, 2018), archived at https://perma.cc/L686-6L62 (Furman Center Report).
  • 230. Id at *2 & n 2.
  • 231. See Laura Waxmann, Tenant Advocacy Groups Set to Receive Funding Under ‘Right to Counsel’ Program (San Francisco Examiner, Nov 28, 2018), archived at https://perma.cc/HDG6-BKH5; Jared Brey, Tenants’ Right to Counsel on the Move, Next Stop Newark, (Next City, Jan 10, 2019) archived at https://perma.cc/PWD7-9S9H. Other jurisdictions have also introduced or piloted legislation to create similar policies, including Boston, Philadelphia, Los Angeles, and Washington, DC. See also Furman Center Report at *2 (cited in note 229); City of Boston, Mayor Walsh Announces 2019 Housing Security, Economic Mobility Legislative Agenda (Jan 7, 2019), archived at https://perma.cc/8CU9-6Z2T.
  • 232. See Furman Center Report at *3–6 (cited in note 229).
  • 233. See Franzese, Gorin, and Guzik, 69 Rutgers L Rev at 13 (cited in note 19); Cotton, 19 CUNY L Rev at 84 (cited in note 28) (noting that “[t]he lack of counsel means that the parties are particularly dependent on the court to ensure that the rule of law is applied”); id at 86–87 (arguing that advocates hoping to improve utilization of the implied warranty of habitability should not focus their efforts on access to counsel because the data suggest that all efforts thus far have faltered, and moreover the provision of additional lawyers would impose considerable resource demands on the courts). But see Bezdek, 20 Hofstra L Rev at 538 n 16 (cited in note 27) (arguing against a solution involving access to counsel because it is “parentalistic [sic] and it lets us off the hook for our parts in the charade of legal entitlement and rights vindication”).
  • 234. See note 233.
  • 235. See Franzese, Gorin, and Guzik, 69 Rutgers L Rev at 37 (cited in note 19).
  • 236. Id at 22.
  • 237. These findings suggest that a more tightly structured system for integrating eviction case adjudication with code enforcement records, like that proposed by Professor Zulack, may be needed to ensure that judges in fact take advantage of the availability of code enforcement records. See Zulack, 40 John Marshall L Rev at 449–53 (cited in note 146).
  • 238. See Myron Moskovitz, The Implied Warranty of Habitability: A New Doctrine Raising New Issues, 62 Cal L Rev 1444, 1504 (1974). Moskovitz further hailed the new law as providing greater bargaining leverage to tenants in settlement negotiations with their landlords in eviction cases. Id.
  • 239. Id at 1503.
  • 240. All cases included in this study were from 2016. See Part III.
  • 241. NY Real Prop Law § 747-a (1997), repealed by L2019, ch 36, pt M § 18 (effective June 14, 2019).
  • 242. No scholarly publication of which I am aware has referenced cure period restrictions as a factor in the effectiveness of the warranty of habitability.
  • 243. See Cal Civ Pro Code § 1174.2(a) (five-day cure period which judge has no authority to extend); 25 Del Code Ann § 5716 (ten-day cure period if “good faith dispute” caused the nonpayment); Mass Gen Laws Ann ch 239, § 8A (seven-day nondiscretionary cure period); Mich Comp Laws Ann § 600.5744(4) (nondiscretionary ten-day cure period); NM Stat Ann § 47-8-33.E(2) (three-day nondiscretionary cure period so long as tenant complies with requirements of state’s rent withholding statute); 12 Okla Stat Ann § 1148.10B.B (three-day nondiscretionary period conditional upon tenant’s compliance with certain notice requirements); Wash Rev Code Ann § 59.18.410 (five-day nondiscretionary cure period).
  • 244. See note 243.
  • 245. See Ala Code Ann § 35-9A-461(e); Ala Rule Civ P 62(a), 62(dc); Alaska Stat Ann §§ 34.03.190, 34.03.220(b); Ariz Rev Stat § 12-1178.C; Colo Rev Stat Ann § 13-40-122; Ga Code Ann § 44-7-53; Hawaii Rev Stat §§ 666-11, 666-14; Idaho Code § 6-316; 735 ILCS 5/9-209; Ind Code Ann § 32-31-1-6; Iowa Code § 562A.27; Kan Stat Ann § 58-2561; Ky Rev Ann Stat § 383.240; La Civ Code Ann § 2704; Md Real Prop Code Ann § 8-401(e)(1); Minn Stat §§ 504B.291, 504B.285; Miss Code Ann §§ 11-25-23, 89-7-41; Mo Rev Stat § 535.020; Mont Code Ann § 70-24-427; Neb Rev Stat § 76-1446; Nev Rev Stat Ann § 40.414; NH Rev Stat Ann § 540.14; NJ Stat Ann § 2A:42-92; NC Gen Stat § 42-26; ND Cent Code § 47-32-01; Ohio Rev Code Ann § 1923.13; Or Rev Stat §§ 105.145, 105.161; 68 Pa Cons Stat Ann § 250.503; RI Gen Laws § 34-18-51; SD Cod Laws § 21-16-1 et seq; Tenn Code Ann §§ 29-18-126, 66-28-501; Tex Rule Civ P 509, 510; Tex Prop Code §§ 92.056, 92.0561, 92.056; Utah Code Ann §§ 57-22-5, 57-22-6, 78B-6-808, 78B-6-811; Va Code Ann § 55.1-1251; W Va Code §§ 37-6-8, 37-6-19; Wis Stat § 799.44; Wyo Stat Ann §§ 1-21-1008, 1-21-1206(d).
  • 246. The average length of the repayment period among all nonpayment of rent cases (in the “all nonpayment of rent cases” dataset) was 38.6 days. See Part IV.B.1. There was no statistically significant difference between the length of the repayment period in “likely meritorious claim” cases and “likely no meritorious claim” cases.
  • 247. Settlements are negotiated “in the shadow of the law.” See generally Don L. Coursey and Linda R. Stanley, Pretrial Bargaining Behavior Within the Shadow of the Law: Theory and Experimental Evidence, 8 Intl Rev L & Econ 161 (1988); Robert Cooter, Stephen Marks, and Robert Mnookin, Bargaining in the Shadow of the Law: A Testable Model of Strategic Behavior, 11 J Legal Stud 225 (1982). According to this theory, parties justify their own settlement positions and evaluate the strength of the opposing side’s positions based on an analysis of likely outcomes and willingness to go to trial. Cooter, Marks, and Mnookin, 11 J Legal Stud at 228–29 (cited in note 247). Where one side is aware that the other side is unable to support their position at trial, or is unlikely to incur the risk involved with taking the case to trial, that side is unlikely to cede to the other side’s demands. Id at 245.
  • 248. See, for example, Bezdek, 20 Hofstra L Rev at 569 (cited in note 27) (qualitatively describing nonpayment of rent proceedings and calling them “scene[s] . . . of debt collection”).
  • 249. An eviction complaint is referred to as a “petition” under New York law. Landlords are required to state the rent-regulation status of the unit in the petition. See NY Real Prop Law § 741.