Picture a rental applicant with a steady job and years of on-time rent payments. She applies for an apartment and is turned down within minutes. The landlord’s notice names a tenant screening company and informs her she can request a copy of her report. The report shows a credit score just below the landlord’s cutoff, and nothing else. Her years of on time rent never appear, because her landlords never reported them to a credit bureau. Nothing in the report is wrong. It simply leaves out the information that best predicts whether she will pay rent.
Now picture a second applicant. Her report lists an eviction case from six years ago that was dismissed. She never lost a judgment and never owed money. Nonetheless, the screening software counted the filing anyway because it does not distinguish an eviction filing from an eviction judgment.[1]
Federal consumer reporting law treats these two applicants very differently. The second may have a right to dispute the denial of her application. The first almost certainly has nothing to dispute. That gap shows how the law regulates the companies that increasingly decide who can rent a home. The Fair Credit Reporting Act (“FCRA”) asks whether a tenant-screening report is accurate.[2] It does not ask whether the score built from that report predicts anything.[3]
Tenant screening has become a large, concentrated industry. A small number of firms, including TransUnion’s SmartMove, CoreLogic, RealPage, and AppFolio, combine credit histories, eviction records, and rental histories into a single report.[4] The product usually ends in an acceptance, denial, or conditional recommendation, or a numerical risk score that claims to predict whether a tenant will pay rent, damage the unit, or breach the lease.[5] Applicants bear the cost of these products as part of their application fee.[6] Landlords favor them because they cut a review that once took days of checking records and references to a few minutes.[7]
Credit history is at the center of these products. The National Consumer Law Center reports that as many as 90% of landlords use it in screening, and that many automatically reject applicants with low credit scores.[8]
The law treats screening companies as “consumer reporting agencies,” which brings them under the FCRA.[9] They must follow “reasonable procedures to assure maximum possible accuracy,” give consumers access to their files, and investigate disputes.[10] A landlord who rejects an applicant based on a report must send an adverse action notice.[11] The notice must name the screening company and explain the applicant’s right to receive and dispute the report.[12] On paper, this process looks like a complete system of accountability, but the enforcement record reveals its continuous flaws.
The two best-known federal actions against tenant-screening companies are examples of accuracy enforcement. In 2018, RealPage agreed to pay $3 million to settle Federal Trade Commission charges.[13] The agency alleged that RealPage’s procedures for matching records to applicants were so weak that other people’s records showed up on applicants’ reports.[14] Two years later, AppFolio settled for about $4.3 million.[15] The government alleged that AppFolio did not reasonably check whether the identifiers on the court records it reported actually matched the applicant.[16]
These were serious failures, but consider what the government was policing. Each case involved systems that pulled the wrong person, the wrong record, or a record that should have been sealed. Each case asked whether the underlying data was correct and properly sourced.[17] Neither asked whether the scoring model built on that data was a sound way to decide who should get an apartment.[18] As commentators have put it, enforcement in this area has reached only the most egregious abuses and avoided the industry’s deeper structural problems.[19] This limited reach reflects the FCRA itself, which gives regulators little room to address anything beyond the most flagrant violations.
The FCRA was built for traditional credit reporting, and its main concern is accuracy.[20] Even with its narrow focus on accuracy, the act sets a modest bar. Courts have read the “maximum possible accuracy” requirement to mean reasonable quality-control procedures, not error-free reports.[21] Given the well-known problems with court records and name-matching, a fair number of mistakes can pass through a system that still counts as “reasonable.”[22]
The bigger problem is what the statute leaves out entirely. The FCRA does not regulate whether the screening criteria themselves are fair or sensible.[23] It covers only the accuracy of the underlying data.[24] An applicant rejected because of a mismatched record has a right to dispute. An applicant rejected because the algorithm gave too much weight to an accurate data point has no FCRA remedy.[25] The Consumer Financial Protection Bureau has the same limit. Its authority covers the FCRA’s accuracy and procedural requirements, not whether screening algorithms are substantively sound.[26]
This would matter less if the criteria were known to work. For credit scores, the evidence is weak. NCLC reports that “there are no quantitative or scientific studies showing that credit information accurately predicts a successful tenancy.”[27] Credit scores have some predictive value for rent payment. But they are imperfect stand-ins for reliability as a tenant, especially for low-income applicants whose finances may be affected by factors unrelated to their ability to pay rent.[28]
Applicants with thin or nonexistent credit files face a further problem. A score built on little or no data says little about how they will behave as tenants.[29] Eviction records raise similar concerns. They often reflect a tenant’s past housing conditions, landlord harassment, or a sudden financial shock rather than any choice the tenant made.[30]
The lack of transparency makes this worse. Screening companies generally treat their algorithms as trade secrets.[31] As a result, neither applicants nor regulators can easily see how inputs are weighted, what cutoffs are used, or whether the model has ever been validated.[32] The adverse action notice usually names the vendor and gives no real explanation of what drove the denial.[33] Applicants cannot challenge reasoning they cannot see.
Reforming the FCRA does not mean abandoning its accuracy framework. Landlords have a real interest in judging whether an applicant will pay rent and comply with their lease.[34] The task for regulators is to ensure that screening products marketed to landlords serve that interest.[35] Four reforms would better align tenant screening products with landlords’ interest in assessing an applicant.
A consumer reporting agency that sells a risk score or an accept/deny recommendation, not just raw records, should have to show that its inputs reasonably predict the outcomes the score claims to measure. This applies a basic model-risk principle to an industry that currently has none.[36] Inputs that can’t be validated, such as a credit-score cutoff with no demonstrated link to rent payment, would have to be dropped or given less weight.
Adverse action notices should list the specific factors that led to a denial, not just the name of the company that produced the report.[37] An applicant who learns that a single collection account on her credit file was decisive can contest or explain it; one who is only told that “SmartMove said no” cannot.
Screening models should be allowed, and encouraged, to use data that bears more directly on whether someone will pay rent. Examples include rental payment history reported by landlords or rent-reporting services, utility payments, and bank account transaction data.[38] These sources have their own limits, but they may give more accurate signals of an applicant’s ability to pay rent than a general-purpose credit score.[39] For the applicant in the opening example, they would put her strongest evidence into the file.
New York City’s Local Law 144 already requires employers that use automated hiring tools to obtain annual bias audits and to notify candidates.[40] A similar requirement for tenant-screening products, with published audits testing accuracy and predictive validity, would shift oversight from one-off, after-the-fact enforcement to regular review.[41] Early experience with Local Law 144 shows that audit mandates need clear standards and real enforcement as industry compliance has been opaque.[42] Thus, the FCRA should incorporate carefully designed audit requirements that enable regulators to collect the information necessary for effective oversight.
The RealPage and AppFolio settlements demonstrate that the FCRA can address inaccurate consumer reports, including reports that attribute another person’s records to an applicant. But tenant-screening decisions increasingly depend on how scoring models weigh accurate data, especially the applicant’s credit score. Until consumer reporting law asks whether a tenant-screening score actually predicts an applicant’s ability to pay, accurate reports will continue to produce arbitrary denials.
[1] See Nat’l Consumer L. Ctr., Digital Denials: How Abuse, Bias, and Lack of Transparency in Tenant Screening Harm Renters 38–39 (Sept. 2023), https://www.nclc.org/wp-content/uploads/2023/09/202309_Report_Digital-Denials.pdf; see also Consumer Fin. Prot. Bureau, Tenant Background Checks Market 2 (Nov. 2022), https://files.consumerfinance.gov/f/documents/cfpb_tenant-background-checks-market_report_2022-11.pdf.
[2] See Consumer Fin. Prot. Bureau, Consumer Snapshot: Tenant Background Checks, 11 (Nov. 2022), https://files.consumerfinance.gov/f/documents/cfpb_consumer-snapshot-tenant-background-check_2022-11.pdf.
[3] See id. at 4.
[4] See Amanda Abrams, Tenant Screening: A Billion-Dollar Industry with Little Oversight, Shelterforce (June 11, 2025), https://shelterforce.org/2025/06/11/tenant-screening-a-billion-dollar-industry-with-little-oversight-whats-being-done-to-protect-renters/; see also Cyrus Farivar, Tenant Screening Software Faces National Reckoning, NBC News (Mar. 11, 2021), https://www.nbcnews.com/tech/tech-news/tenant-screening-software-faces-national-reckoning-n1260975.
[5] See Consumer Fin. Prot. Bureau, Consumer Snapshot, supra note 2, at 9.
[6] See id. at 8.
[7] See Mathew H. Leiwant, Locked Out: How Algorithmic Tenant Screening Exacerbates the Eviction Crisis in the United States, 6 Geo. L. Tech. Rev. 1, 5 (2022).
[8] See Nat’l Consumer L. Ctr., Past Imperfect: How Credit Scores and Other Analytics “Bake In” and Perpetuate Past Discrimination 3 (updated Feb. 2024), https://www.nclc.org/wp-content/uploads/2016/05/20240227_Issue-Brief_Past-Imperfect.pdf.
[9] See Fair Credit Reporting Act, 15 U.S.C. § 1681.
[10] See id. § 1681e.
[11] See id. § 1681m(a).
[12] See id. § 1681e.
[13] See Press Release, Fed. Trade Comm’n, Texas Company Will Pay $3 Million to Settle FTC Charges That It Failed to Meet Accuracy Requirements for Its Tenant Screening Reports (Oct. 16, 2018), https://www.ftc.gov/news-events/news/press-releases/2018/10/texas-company-will-pay-3-million-settle-ftc-charges-it-failed-meet-accuracy-requirements-its-tenant.
[14] See id.
[15] See Lesley Fair, FTC’s AppFolio Case: The Fair Credit Reporting Act Does More Than Just Abide, Fed. Trade Comm’n: Bus. Blog (Dec. 8, 2020), https://www.ftc.gov/business-guidance/blog/2020/12/ftcs-appfolio-case-fair-credit-reporting-act-does-more-just-abide.
[16] See id.
[17] See FTC v. RealPage, Inc., No. 3:18-cv-02737-N (N.D. Tex. Oct. 19, 2018); see also FTC v. AppFolio, Inc., No. 1:20-cv-03563 (D.D.C. Dec. 8, 2020).
[18] See id.
[19] See Cyrus Farivar, Tenant Screening Software Faces National Reckoning, NBC News (Mar. 11, 2021), https://www.nbcnews.com/tech/tech-news/tenant-screening-software-faces-national-reckoning-n1260975.
[20] See id.
[21] See Cahlin v. Gen. Motors Acceptance Corp., 936 F.2d 1151, 1156 (11th Cir. 1991) (holding that FCRA does not impose strict liability for inaccurate reports).
[22] See Leiwant, Locked Out, supra note 7, at 10-11.
[23] See Fair Credit Reporting Act, 15 U.S.C. § 1681.
[24] See Leiwant, Locked Out, supra note 7, at 14-15.
[25] See id. at 16-17.
[26] See Letter from Letitia James, N.Y. Att’y Gen., et al., to Rohit Chopra, Dir., Consumer Fin. Prot. Bureau, & April Tabor, Sec’y, Fed. Trade Comm’n 2, 6 (May 30, 2023), https://oag.ca.gov/system/files/attachments/press-docs/Multi%20State%20State%20Attorneys%20General%20Comment%20Letter%20to%20the%20FTC%20CFPB%20on%20Tenant%20Screening%20%281%29.pdf.
[27] Nat’l Consumer L. Ctr., Past Imperfect: How Credit Scores and Other Analytics “Bake In” and Perpetuate Past Discrimination 3 (updated Feb. 2024), https://www.nclc.org/wp-content/uploads/2016/05/20240227_Issue-Brief_Past-Imperfect.pdf.
[28] See Wonyoung So, Which Information Matters? Measuring Landlord Assessment of Tenant Screening Reports, 33 Hous. Pol’y Debate 1484, 1488 (Aug. 30, 2022), https://doi.org/10.1080/10511482.2022.2113815.
[29] See id.
[30] See id.
[31] See Leiwant, Locked Out, supra note 7, at 7.
[32]See Eva Rosen, Philip M.E. Garboden & Jennifer E. Cossyleon, Racial Discrimination in Housing: How Landlords Use Algorithms and Home Visits to Screen Tenants, 86 Am. Socio. Rev. 787 (2021).
[33] See Consumer Fin. Prot. Bureau, Consumer Snapshot, supra note 2, at 9–10.
[34] See Leiwant, Locked Out, supra note 7, at 6.
[35] See id. at 20.
[36] See Wonyoung So, Which Information Matters? Measuring Landlord Assessment of Tenant Screening Reports, 33 Hous. Pol’y Debate 1484, 1502-03 (Aug. 30, 2022), https://doi.org/10.1080/10511482.2022.2113815.
[37] See Consumer Fin. Prot. Bureau, Consumer Snapshot, supra note 2, at 9–10.
[38] See Jung Hyun Choi et al., Reducing the Black-White Homeownership Gap Through Underwriting Innovations: The Potential Impact of Alternative Data in Mortgage Underwriting 1–3 (Urban Inst. Oct. 2022), https://www.urban.org/sites/default/files/2022-10/Reducing%20the%20Black-White%20Homeownership%20Gap%20through%20Underwriting%20Innovations.pdf.
[39] See id. at 17.
[40] See N.Y.C. Local Law No. 144 (2021) (enforcement began July 5, 2023).
[41] See id.
[42] See Lucas Wright et al., Null Compliance: NYC Local Law 144 and the Challenges of Algorithm Accountability, in Proc. of the 2024 ACM Conf. on Fairness, Accountability, and Transparency 1701, 1708–09 (2024).
