Miller v. RP On-Site, LLC ( 2020 )

1 2 3 4 5 6 7 8 UNITED STATES DISTRICT COURT 9 NORTHERN DISTRICT OF CALIFORNIA 10 SAN JOSE DIVISION 11 12 BRIAN MILLER, Case No. 19-CV-02114-LHK 13 Plaintiff, ORDER DENYING MOTION FOR CLASS CERTIFICATION 14 v. Re: Dkt. No. 54 15 RP ON-SITE, LLC, 16 On-Site. 17 18 Plaintiff Brian Miller brings this putative class action against Defendant RP On-Site, LLC 19 (“On-Site”) under the Fair Credit Reporting Act (“FCRA”). Before the Court is Plaintiff’s motion 20 for class certification. ECF No. 53-4 (“Mot.”). Having considered the submissions of the parties, 21 the relevant law, and the record in this case, the Court DENIES Plaintiff’s class certification 22 motion. 23 I. BACKGROUND 24 A. Factual Background on On-Site 25 On-Site is a consumer reporting company that specializes in consumer reports for the real 26 estate industry. Mot. at 3 (citing deposition testimony). On-Site’s customers are property managers 27 who ask On-Site to prepare reports on tenant applicants. Id. (same). As relevant here, some reports 1 contain a tenant applicant’s criminal record. Sohal Decl. ¶ 5, ECF No. 55-7 (declaration of the 2 director who manages On-Site’s tenant screening platforms). Preparing a report on a tenant 3 applicant’s criminal record takes roughly four steps. 4 First, On-Site obtains publicly available criminal records from thousands of jurisdictions. 5 Id. ¶ 8. To obtain these records, On-Site generally relies on “bulk” data vendors. Where vendor 6 data is incomplete, however, On-Site obtains records from other sources. Id. ¶ 15. 7 Second, when On-Site receives a request for a consumer report on a tenant applicant, On- 8 Site searches for data to complete that report. Id. ¶ 16. If a report requests information that On-Site 9 does not have on hand, it sometimes makes a specific records request to a specific jurisdiction at 10 the time of the request. Id. ¶ 18. 11 Third, On-Site checks each consumer report to ensure that the public records matched to 12 the applicant appear complete. For instance, On-Site checks whether a conviction linked to an 13 applicant has a corresponding severity (i.e., felony or misdemeanor). Id. ¶ 20. Convictions that 14 lack a severity designation are routed to a manual review queue. Id. 15 Fourth, On-Site provides the consumer report to the tenant applicant and the property 16 manager at the same time. Mot. at 3 (citing deposition testimony). The tenant applicant can then 17 contact On-Site to dispute the report’s accuracy. Disputes are routed to a team of investigators. 18 Sohal Decl. ¶ 22. A low percentage of reports are disputed. Id. ¶ 23 (sealed percentage). An even 19 lower percentage of reports are found to have errors. Id. 20 In this process, On-Site may receive two types of criminal records particularly relevant 21 here. One type is criminal records about expunged, set aside, or dismissed convictions. Receiving 22 such records is unusual because, typically, they are publicly unavailable. Id. ¶ 25. Still, sometimes 23 On-Site receives such records because state law makes them available, or a jurisdiction has failed 24 to update its own records. Id. ¶ 26. Where a jurisdiction labels a record as set aside or dismissed, 25 On-Site excludes that record from consumer reports. Id. ¶ 29. 26 The other type of particularly relevant criminal record is records from departments of 27 correction (“DOCs”). On-Site receives public records from dozens of DOCs. Id. ¶ 31 (sealed 1 number). Most convictions that DOCs provide to On-Site are the conviction(s) that put the tenant 2 applicant in prison. Id. ¶ 32. However, On-Site cannot automatically filter DOC records for 3 convictions versus disciplinary infractions committed while in prison. Id. ¶ 35. Rather, On-Site 4 checks whether DOC records, like other records, are complete. Id. ¶ 36. If a DOC conviction or 5 disciplinary infraction lacks a corresponding severity (i.e., felony or misdemeanor), On-Site puts 6 that record into its manual review process. 7 B. Factual Background on Named Plaintiff 8 Between 1996 and 2005, Plaintiff committed three violent felonies. Specifically, in 9 October 1996, Plaintiff pled guilty to felony robbery in Arizona Superior Court in the County of 10 Maricopa (“Maricopa Superior Court”). See On-Site’s Opp’n to Class Cert. at 6, ECF No. 55-6 11 (“Opp’n”) (citing deposition testimony). In April 2000, Plaintiff pled guilty to felony aggravated 12 assault in the same court. In 2005, Plaintiff again pled guilty to felony aggravated assault in the 13 same court. These convictions are still publicly available on the Arizona courts website. Id. at 7 14 (same). 15 While in prison, Plaintiff was not a model prisoner. He committed at least 15 incidents of 16 threatening someone with harm; four incidents of assault; two incidents of tampering; and one 17 incident each of disorderly conduct, possessing a weapon, conspiracy, and stalking. Id. at 6 (citing 18 deposition testimony). 19 As a result of his crimes, Plaintiff lost certain civil liberties. Among these liberties was 20 Plaintiff’s right to own a firearm. After being released from prison, Plaintiff thrice asked the 21 Maricopa Superior Court to restore his right to own a firearm. Id. Each time, Plaintiff was denied. 22 Thus, Plaintiff admitted at his deposition that his convictions still subject him to legal disabilities. 23 Id. Certain other disabilities were lifted, however, after Plaintiff successfully completed Arizona 24 law’s requirements for restoring a felon’s civil rights. Mot. at 5 (citing A.R.S. § 13-907 and 25 arguing that the convictions were “set aside”). 26 On November 5, 2018, Plaintiff applied for an apartment at the complex “Indigo Lofts.” 27 Opp’n at 7. For that application, On-Site provided Indigo Lofts with a consumer report on Plaintiff 1 (the “Report”). Id. The Report listed Plaintiff’s 1996 and 1999 felony convictions, as well as 26 2 additional “felonies” reported by the Arizona Department of Correction. Id. About 25 of those 3 “felonies” were records of the disciplinary incidents that Plaintiff committed in prison. Id. 4 Even so, the Report recommended that Indigo Lofts approve Plaintiff’s application based 5 on Indigo Loft’s guideline of considering crimes committed in the past decade. Id. at 8 (citing 6 Report and deposition testimony). In turn, Indigo Lofts approved Plaintiff’s application. Id. 7 Ultimately, though, Plaintiff decided not to live at Indigo Lofts. In his view, the Indigo Lofts 8 manager vetting his application was rude. The manager “had her attitude. I’m not going to live 9 somewhere like that.” Id. at 138:15–17. 10 C. Procedural History 11 On April 19, 2019, Plaintiff brought this putative class action against On-Site. ECF No. 1 12 (“Compl.”). Plaintiff alleged that On-Site violated the FCRA by (1) willfully not “follow[ing] 13 reasonable procedures to assure maximum possible accuracy” of On-Site’s reports, Compl. ¶ 53 14 (quoting 15 U.S.C. § 1681e(b)); and (2) willfully providing adverse information, other than 15 records of convictions, that antedated reports by more than seven years, Compl. ¶ 62 (citing 15 16 U.S.C. § 1681c(a)(2) & (5)). On June 13, 2019, On-Site filed an answer to the Complaint. ECF 17 No. 18. 18 On September 4, 2020, Plaintiff filed the instant motion for class certification. ECF No. 53 19 (sealed version). Plaintiff moved to certify three classes: 20 [“Disciplinary Infraction Class”] All natural persons with an address in the United States and its Territories about whom, from April 19, 2017 continuing through the 21 resolution of this action, Defendants [sic] prepared a consumer report which 22 included information regarding one or more administrative action undertaken by a department of corrections which Defendants [sic] reported as a “felony conviction.” 23 [“Antedated Report Class”] All natural persons with an address in the United States 24 and its Territories who were subjects of tenant screening reports created by 25 Defendants [sic] from April 19, 2017 and continuing through the resolution of this action, that contained adverse information other than convictions (including 26 criminal cases that had been expunged, sealed, set aside, or dismissed) that 27 preceded the report by more than seven years. [“Expungement Class”—abandoned on reply] All natural persons with an address 1 in the United States and its Territories about whom, from April 19, 2017 and 2 continuing through the resolution of this action, Defendants [sic] prepared a consumer report which included information regarding one or more criminal case 3 which had been expunged, sealed, or otherwise removed from public dissemination 4 at the time the report was prepared. 5 Mot. at 1–2. On-Site filed its opposition to class certification on October 5, 2020. ECF No. 55-6 6 (sealed version). On October 19, 2020, Plaintiff filed his reply. ECF No. 58. In his reply, Plaintiff 7 abandoned certification of the Expungement Class because Plaintiff found many of On-Site’s 8 arguments “well-taken.” Id. at 9 n.1. 9 II. LEGAL STANDARD 10 Class actions are governed by Rule 23 of the Federal Rules of Civil Procedure. Rule 23 11 does not set forth a mere pleading standard. To obtain class certification, Plaintiffs bear the burden 12 of showing that they have met each of the four requirements of Rule 23(a) and at least one 13 subsection of Rule 23(b). Zinser v. Accufix Research Inst., Inc., 253 F.3d 1180, 1186, amended by 14 273 F.3d 1266 (9th Cir. 2001). “A party seeking class certification must affirmatively demonstrate 15 . . . compliance with the Rule[.]” Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 350 (2011). 16 Rule 23(a) provides that a district court may certify a class only if: “(1) the class is so 17 numerous that joinder of all members is impracticable; (2) there are questions of law or fact 18 common to the class; (3) the claims or defenses of the representative parties are typical of the 19 claims or defenses of the class; and (4) the representative parties will fairly and adequately protect 20 the interests of the class.” Fed. R. Civ. P. 23(a). That is, the class must satisfy the requirements of 21 numerosity, commonality, typicality, and adequacy of representation to maintain a class action. 22 Mazza v. Am. Honda Motor Co., Inc., 666 F.3d 581, 588 (9th Cir. 2012). 23 If all four prerequisites of Rule 23(a) are satisfied, the Court must also find that Plaintiffs 24 “satisfy through evidentiary proof” at least one of the three subsections of Rule 23(b). Comcast 25 Corp. v. Behrend, 569 U.S. 27, 33 (2013). Rule 23(b) sets forth three general types of class 26 actions. See Fed. R. Civ. P. 23(b)(1)–(b)(3). As relevant here, Plaintiffs seek certification under 27 Rule 23(b)(3). A class may be certified under Rule 23(b)(3) if a court finds that “questions of law 1 or fact common to class members predominate over any questions affecting only individual 2 members, and that a class action is superior to other available methods for fairly and efficiently 3 adjudicating the controversy.” Fed. R. Civ. P. 23(b)(3). 4 “[A] court’s class-certification analysis must be ‘rigorous’ and may ‘entail some overlap 5 with the merits of the plaintiff’s underlying claim[.]’” Amgen Inc. v. Conn. Ret. Plans & Trust 6 Funds, 568 U.S. 455, 465–66 (2013) (quoting Dukes, 564 U.S. at 351); see also Mazza, 666 F.3d 7 at 588 (“‘Before certifying a class, the trial court must conduct a ‘rigorous analysis’ to determine 8 whether the party seeking certification has met the prerequisites of Rule 23.’” (quoting Zinser, 253 9 F.3d at 1186)). This “rigorous” analysis applies to both Rule 23(a) and Rule 23(b). Comcast, 569 10 U.S. at 34 (discussing how Congress included “addition[al] . . . procedural safeguards for (b)(3) 11 class members beyond those provided for (b)(1) or (b)(2) class members (e.g., an opportunity to 12 opt out)” and how a court has a “duty to take a ‘close look’ at whether common questions 13 predominate over individual ones”). 14 Nevertheless, “Rule 23 grants courts no license to engage in free-ranging merits inquiries 15 at the certification stage.” Amgen, 568 U.S. at 466. “Merits questions may be considered to the 16 extent—but only to the extent—that they are relevant to determining whether the Rule 23 17 prerequisites for class certification are satisfied.” Id. If a court concludes that the moving party has 18 met its burden of proof, then the court has broad discretion to certify the class. Zinser, 253 F.3d at 19 1186. 20 III. DISCUSSION 21 On-Site argues that Plaintiff’s class certification motion fails on five independent grounds: 22 (1) lack of evidence on numerosity; (2) atypicality; (3) inadequacy of Plaintiff as class 23 representative and Plaintiff’s counsel as class counsel; (4) lack of predominance and commonality; 24 and (5) unmanageability of class adjudication. The Court addresses numerosity and Plaintiff’s 25 inadequacy as class representative below. Because both grounds independently require denial of 26 class certification, the Court need not reach On-Site’s other grounds for denying class certification. 27 A. Plaintiff fails to prove numerosity by a preponderance of the evidence. 1 On-Site first argues that Plaintiff fails to prove that any of the purported classes satisfy 2 Rule 23(a)(1)’s numerosity requirement. The Court agrees. 3 Plaintiff bears the “burden of affirmatively demonstrating that the class meets the 4 requirements of Federal Rule of Civil Procedure 23.” Mazza, 666 F.3d at 588. Rule 23(a)(1) 5 requires that “a class must be ‘so numerous that joinder of all members is impracticable.’” 6 Sarmiento v. Sealy, Inc., No. 18-CV-01990-JST, 2020 WL 4458915, at *4 (N.D. Cal. May 27, 7 2020) (quoting Fed. R. Civ. P. 23(a)(1)). The numerosity requirement “requires examination of the 8 specific facts of each case.” Rannis v. Recchia, 380 F. App’x 646, 651 (9th Cir. 2010) (quoting 9 Gen. Tel. Co. of the Nw., Inc. v. EEOC, 446 U.S. 318, 330 (1980)). Given plaintiff’s burden of 10 showing numerosity and the Court’s “require[d] examination of the specific facts of each case,” 11 plaintiff’s “failure to present evidence to show numerosity precludes class certification.” Uschold 12 v. Carriage Servs., Inc., No. 17-CV-04424-JSW, 2020 WL 1466172, at *11 (N.D. Cal. Mar. 6, 13 2020), appeal dismissed, No. 20-15523, 2020 WL 3470090 (9th Cir. Apr. 24, 2020). In short, 14 “mere conclusory allegations as to the estimated class size are insufficient for plaintiff’s burden of 15 proof.” Hernandez v. City of San Jose, No. 16-CV-03957-LHK, 2019 WL 4450930, at *18 (N.D. 16 Cal. Sept. 17, 2019) (quoting Sueoka v. United States, 101 F. App’x 649, 653 (9th Cir. 2004)). 17 Here, Plaintiff entirely “fail[s] to present evidence” to show numerosity. Uschold, 2020 18 WL 1466172, at *11. Plaintiff instead makes “mere conclusory allegations as to the estimated 19 class size.” Hernandez, 2019 WL 4450930, at *18 (quoting Sueoka, 101 F. App’x at 653). 20 Confirming this conclusion is a simple comparison of Plaintiff’s relatively narrow class definitions 21 with his conclusory “evidence” of numerosity. 22 Plaintiff’s two putative classes are the “Disciplinary Infraction Class” and “Antedated 23 Report Class.” Both classes comprise persons who have been subjects of On-Site consumer 24 reports containing particular errors during the class period April 19, 2017 through the resolution of 25 this action. 26 The “Disciplinary Infraction Class” includes persons about whom On-Site erroneously 27 1 reported, during the class period April 19, 2017 through the resolution of this action, a department 2 of corrections “administrative action” as a “felony conviction.” Specifically, Plaintiff defines the 3 “Disciplinary Infraction Class” as follows: 4 All natural persons with an address in the United States and its Territories about whom, from April 19, 2017 continuing through the resolution of this action, 5 Defendants [sic] prepared a consumer report which included information regarding 6 one or more administrative action undertaken by a department of corrections which Defendants [sic] reported as a “felony conviction.” 7 Mot. at 2. Plaintiff asserts that the consumer reports at issue in the Disciplinary Infraction Class 8 willfully violate FCRA’s “reasonable procedures” requirement. 15 U.S.C. § 1681e(b). Under that 9 requirement, On-Site must “follow reasonable procedures to assure maximum possible accuracy of 10 the information concerning the individual about whom [a] report relates.” Compl. ¶ 53 (quoting 15 11 U.S.C. § 1681e(b)); see id. ¶ 55 (discussing “administrative actions by departments of 12 corrections”). Here, a report is inaccurate if it contains “administrative action[s] undertaken by a 13 department of corrections” that are not in fact “felony convictions.” Accurately reporting felony 14 convictions does not violate FCRA. See, e.g., Shaw v. Experian Info. Sols., Inc., 891 F.3d 749, 755 15 (9th Cir. 2018) (holding that a plaintiff must first show that defendant prepared an inaccurate 16 report). 17 The “Antedated Report Class” comprises persons about whom, during the class period of 18 April 19, 2017 through the resolution of this action, On-Site erroneously reported “adverse 19 information” (other than convictions) more than seven years older than On-Site’s report. 20 Specifically, Plaintiff defines the “Antedated Report Class” as follows: 21 All natural persons with an address in the United States and its Territories who 22 were subjects of tenant screening reports created by Defendants [sic] from April 19, 23 2017 and continuing through the resolution of this action, that contained adverse information other than convictions (including criminal cases that had been 24 expunged, sealed, set aside, or dismissed) that preceded the report by more than 25 seven years. 26 Id. Plantiff asserts that the consumer reports at issue in the Antedated Report Class willfully 27 violate FCRA’s requirement that consumer reports must exclude “adverse item[s] of information” 1 more than seven years old, unless those adverse items are “records of convictions of crimes.” 15 2 U.S.C. § 1681c(a)(5). 3 For both putative classes, Plaintiff’s only “evidence” for numerosity is two law review 4 articles. Plaintiff cites one article for the Disciplinary Infraction Class, and the other article for the 5 Antedated Report Class. Yet neither article analyzes the instant case or the parties. Not only that, 6 neither article quantifies consumer reports—let alone On-Site consumer reports that were 7 erroneous in specific ways during the class period across the United States. Below, the Court 8 discusses (1) the Disciplinary Infraction Class; (2) the Antedated Report Class; and then (3) case 9 law that underscores how Plaintiff has failed to prove numerosity. 10 1. Plaintiff fails to prove numerosity of the Disciplinary Infraction Class. 11 To prove the Disciplinary Infraction Class’s numerosity, Plaintiff relies solely on Richard 12 Tewksbury et al., Disciplinary Infractions Behind Bars, 39 Crim. Just. Rev. 201, 209 (2014) 13 (“Tewksbury article”). Plaintiff claims this article shows that “at least 52.5% of all prisoners incur 14 at least one disciplinary infraction during their prison stays.” Mot. at 7. 15 However, prisoners who “incur at least one disciplinary infraction” are not the Disciplinary 16 Infraction Class. The Disciplinary Infraction Class instead comprises persons about whom On-Site 17 erroneously reported, during the class period of April 19, 2017 through the resolution of this 18 action, a prison’s “administrative action” as a “felony conviction.” Thus, Plaintiff makes a leap of 19 logic from the Tewksbury article to numerosity. Plaintiff assumes that On-Site erroneously 20 reported administrative actions as felony convictions 52.5% of the time—the same percentage as 21 prisoners who “incur at least one disciplinary infraction.” Mot. at 7. Plaintiff then multiplies 22 52.5% by 107,183—the number of individuals for whom, between April 19, 2017 and June 19, 23 2020, On-Site reported a felony conviction—and assumes that the result proves numerosity. 24 Plaintiff assumes far too much. At least three baseless inferences underlie Plaintiff’s theory 25 of numerosity. First, Plaintiff assumes that the Tewksbury article concluded that “at least 52.5% of 26 all prisoners incur at least one disciplinary infraction during their prison stays.” Mot. at 7. Not so. 27 In fact, the Tewksbury article analyzed only state prisoners and only prisoners in the year 2011— 1 about six years before the start of the class period on April 19, 2017. See Tewksbury et al., supra, 2 at 208. 3 Second, even if the Tewksbury article actually analyzed the disciplinary infractions of “all 4 prisoners,” Mot. at 7, “all prisoners” is many steps removed from the Disciplinary Infraction 5 Class. For one, the Disciplinary Infraction Class only includes persons screened by consumer 6 reports. Mot. at 1–2. For another, not all consumer reports count: only On-Site’s consumer reports 7 do. 8 Furthermore, the Disciplinary Infraction Class is a subset of persons screened by On-Site 9 consumer reports. The Disciplinary Infraction Class includes only those whose On-Site consumer 10 report issued since April 19, 2017 through the resolution of this action and erroneously “included 11 information regarding one or more administrative action undertaken by a department of 12 corrections which [On-Site] reported as a ‘felony conviction.’” 13 The Tewksbury article, by contrast, analyzed disciplinary infractions received by state 14 inmates in 2011. The article did not mention—let alone analyze—an array of factors key to 15 assessing the Disciplinary Infraction Class’s numerosity. For instance, the Tewksbury article did 16 not mention (1) the time period from April 19, 2017 through the resolution of this action; 17 (2) consumer reports; (3) former inmates subject to the “tenant screening” consumer reports 18 described in the Disciplinary Infraction Class; (4) On-Site or any other consumer reporting 19 agency; or (5) whether On-Site erroneously reported disciplinary infractions as “felony 20 convictions” between April 19, 2017 and the resolution of this action. The Tewksbury article’s 21 2011 analysis of state prisoners is thus entirely unmoored from the Disciplinary Infraction Class. 22 Lastly, the Court would have to assume that the Tewksbury article comprises valid 23 “evidence” or “common sense assumptions” that support numerosity. Uschold, 2020 WL 1466172, 24 at *11 (first quote); Sarmiento, 2020 WL 4458915, at *5 (second quote). Yet Plaintiff cannot 25 vouch for the statistics nor replicate their calculation. Nor can Plaintiff make the article’s authors 26 available to On-Site for cross-examination because the authors are not disclosed experts in this 27 case. Thus, the Court would have to take judicial notice of the statistics over On-Site’s vigorous 1 objections. Doing that would be unwarranted. See, e.g., Mitchell v. Tillett, 715 F. App’x 741, 742 2 n.1 (9th Cir. 2018) (refusing to judicially notice a law review article because it “is a source whose 3 accuracy may be debated and questioned”); Crocker v. Glanz, 752 F. App’x 564, 568 (10th Cir. 4 2018) (also refusing to judicially notice a law review article cited by plaintiff, even at the motion 5 to dismiss stage). All told, Plaintiff fails to prove that the Disciplinary Infraction Class is “so 6 numerous that joinder of all members is impracticable.” Fed. R. Civ. P. 23(a)(1). 7 2. Plaintiff fails to prove numerosity of the Antedated Report Class. 8 To prove the Antedated Report Class’s numerosity, Plaintiff relies solely on J.J. Prescott & 9 Sonja B. Starr, Expungement of Criminal Convictions: An Empirical Study, 133 Harv. L. Rev. 10 2460, 2490 (2020) (“Prescott & Starr article”). Plaintiff claims this article shows that 11 “approximately 6.5% of all persons with felony convictions get their convictions expunged, set 12 aside, or dismissed.” Mot. at 7. 13 The same logical fallacies that defeated the Disciplinary Infraction Class also defeat the 14 Antedated Report Class. “[A]ll persons with felony convictions [who] get their convictions 15 expunged, set aside, or dismissed” are not the Antedated Report Class. Mot. at 7. Rather, the 16 Antedated Report Class comprises persons who meet five conditions: (1) “persons with an address 17 in the United States and its Territories” (2) “who were subjects of tenant screening reports created 18 by [On-Site] (3) “that contained adverse information other than convictions (including criminal 19 cases that had been expunged, sealed, set aside, or dismissed),” where (4) that adverse information 20 “preceded the report by more than seven years” and (5) the report was created during the class 21 period of April 19, 2017 through the resolution of this action. Mot. at 2. 22 Thus, connecting the Prescott & Starr article with the Antedated Report Class requires 23 multiple leaps of logic. Plaintiff assumes that On-Site erroneously reports adverse non-conviction 24 information that is more than seven years old—such as convictions expunged more than seven 25 years ago—6.5% of the time. Id. Then, like he did with the Disciplinary Infraction Class, Plaintiff 26 multiplies this general percentage by 107,183—the number of individuals for whom, between 27 April 19, 2017 and June 19, 2020, On-Site reported a felony conviction. In Plaintiff’s view, the 1 product of 6.5% times 107,183 proves numerosity. 2 Again, Plaintiff fails. Three baseless inferences underlie Plaintiff’s theory of numerosity. 3 First, Plaintiff assumes that the Prescott & Starr article concluded that “approximately 6.5% of all 4 persons with felony convictions get their convictions expunged, set aside, or dismissed.” Mot. at 7. 5 Plaintiff misrepresents the article. The Prescott & Starr article analyzed only felons in Michigan 6 with expunged (or expungement-eligible) convictions dated before at least March 2014—more 7 than three years before the start of the class period on April 19, 2017. See Prescott & Starr, supra, 8 at 2483–84. Suffice it to say, the conviction expungement rate in Michigan before at least March 9 2014 might not mirror the conviction expungement rate “in the United States and its Territories” 10 from April 19, 2017 through the resolution of this action. Mot. at 2. 11 Second, even if the Prescott & Starr article actually analyzed “all persons with felony 12 convictions [who] get their convictions expunged, set aside, or dismissed,” Mot. at 7, that 13 population is many steps removed from the Antedated Report Class. The Antedated Report Class, 14 like the Disciplinary Infraction Class, only includes persons screened by consumer reports. Mot. at 15 1–2. Moreover, not all consumer reports count: only On-Site’s consumer reports do. 16 Furthermore, not all On-Site consumer reports trigger membership in the Antedated Report 17 Class. Three more conditions apply: (1) the On-Site consumer report must have been issued during 18 the class period April 19, 2017 through the resolution of this action; (2) the On-Site consumer 19 report issued during that class period must erroneously contain “adverse information other than 20 convictions,” such as expunged criminal cases; and (3) the erroneously reported adverse 21 information must antedate the On-Site consumer report by more than seven years. The Prescott & 22 Starr article fails to quantify any number of consumer reports, let alone the number of On-Site 23 consumer reports that satisfy all these conditions of the Antedated Report Class. 24 On-Site, for its part, presents unrebutted evidence that the Antedated Report Class is small. 25 On-Site generally does not report expunged, sealed, set aside, or dismissed criminal cases. See 26 Sohal Decl. ¶¶ 24–30 (describing On-Site’s procedures). When On-Site does report these criminal 27 cases, it is sometimes because state law makes the cases reportable. See, e.g., G.D. v. Kenny, 15 1 A.3d 300, 313 (N.J. 2011) (“As can be seen, the expungement statute does not obliterate the record 2 of a conviction.”). Other times, On-Site erroneously reports convictions that were “expunged after 3 On-Site reported [them].” Opp’n at 5 (citing Sohal Decl. ¶ 26). Yet these expunged convictions 4 would postdate On-Site’s report—not precede it by more than seven years. Thus, not all reporting 5 errors trigger membership in the Antedated Report Class. Overall, the Antedated Report Class is a 6 long chain of inferences away from Prescott & Starr’s analysis of felony expungement rates in 7 Michigan. 8 Lastly, Plaintiff would have the Court assume that the Prescott & Starr article, like the 9 Tewskbury article, comprises valid “evidence” or “common sense assumptions” that support 10 numerosity. Uschold, 2020 WL 1466172, at *11 (first quote); Sarmiento, 2020 WL 4458915, at *5 11 (second quote). Again, the Court cannot make such a bold assumption. See, e.g., Mitchell, 715 F. 12 App’x at 742 n.1 (refusing to judicially notice law review article); Crocker v. Glanz, 752 F. App’x 13 at 568 (same). Rather, the Court must “rigorous[ly] analy[ze]” whether Plaintiff has met his 14 burden of proving numerosity. Mazza, 666 F.3d at 588 (quoting Zinser, 253 F.3d at 1186). Plaintiff 15 has utterly failed to do so here. 16 3. Courts find numerosity lacking where, as here, plaintiff relies on generalized statistics unlinked to class size. 17 Other courts have also rejected tenuous chains of inferences just like Plaintiff’s. 18 Specifically, courts have denied class certification in FCRA cases where plaintiff relies on an 19 aggregate number of consumer reports to prove numerosity. Two FCRA cases are especially 20 instructive. 21 In Jones v. Sterling Infosystems, Inc. 317 F.R.D. 404, 407 (S.D.N.Y. 2016), plaintiff sought 22 to certify “a Rule 23(b)(3) class of employees or applicants for employment anywhere in the 23 United States who were the subject of a report sold by [d]efendant [‘Sterling’] for employment 24 purposes that contained at least one criminal conviction in New York.” To prove numerosity, 25 plaintiff “relied on the fact that almost 5,000 background reports were sold by Sterling for 26 employment purposes that contained information obtained through [the New York Office of Court 27 1 Administration] that related to a person with a then-current address in New York State” during the 2 class period. Id. at 413. The Jones Court rejected plaintiff’s supposed evidence. Id. The Jones 3 Court reasoned that plaintiff had merely provided “the universe of persons on whom Sterling 4 reported information; it is not the universe of persons as to whom Sterling reported information 5 that was not complete and up to date.” Id. The Jones plaintiff—like Plaintiff here—“provided no 6 information regarding the size of the smaller group of actual class members.” Id. Thus, the Jones 7 Court denied class certification. 8 In Wilson v. Corelogic SafeRent, LLC, No. 14-CV-2477-JPO, 2017 WL 4357568 (S.D.N.Y. 9 Sept. 29, 2017), plaintiff also sought to certify a Rule 23(b)(3) class. That class comprised people 10 (1) who had disputed a consumer report for containing information which had been “expunged . . . 11 or otherwise removed from the public record[] prior to the date of the report”; and for whom 12 (2) “[d]efendant’s comments in connection with the dispute included the root terms ‘unable to 13 verify’; ‘remov!’; ‘delet!’; ‘elim!’; or, ‘eras!’ or any iteration of those root terms.” Id. at *5. To 14 prove numerosity, the Wilson plaintiff presented statistics that—unlike the statistics here—were 15 relatively specific to defendant’s consumer reports. Analyzing defendant’s business records, the 16 Wilson plaintiff presented 3,993 consumer disputes, of which 685 included the “root terms” in the 17 putative class definition. Id. at *6. Even so, the Wilson court held that “[p]laintiff has failed to 18 satisfy his burden to prove numerosity.” Id. at *7 (quoting Jones, 317 F.R.D. at 413). The Wilson 19 Court reasoned that plaintiff had failed to show numerosity of inaccurate reports. Id. Here too, 20 Plaintiff’s general statistic about the number of On-Site’s consumer reports cannot satisfy 21 Plaintiff’s burden to prove numerosity. 22 Non-FCRA cases also confirm that general statistics about defendant’s conduct cannot 23 prove numerosity. In an ERISA class action about termination benefits, for instance, a plaintiff 24 cannot prove numerosity by pointing to the “31,000 employees covered by the plan [who also] lost 25 their jobs.” Siles v. ILGWU Nat. Ret. Fund, 783 F.2d 923, 930 (9th Cir. 1986) (affirming denial of 26 class certification). Similarly, in a consumer class action alleging that defendant misled 27 subscribers, plaintiff cannot merely rely on defendant having 649,576 subscribers to prove that 1 numerous subscribers were misled. See Diacakis v. Comcast Corp., No. 11-CV-3002-SBA, 2013 2 WL 1878921, at *5 (N.D. Cal. May 3, 2013) (denying certification of Rule 23(b)(2) and (b)(3) 3 class). 4 Lastly, if Plaintiff’s failure to prove numerosity were not enough, On-Site presents 5 evidence against numerosity. On-Site identifies procedures it uses to check the completeness and 6 accuracy of its reports. See Sohal Decl. ¶¶ 6–22, 27–37 (describing procedures). These procedures 7 result in a tiny fraction of On-Site’s reports being disputed. See Sohal Decl. ¶ 23 (sealed 8 percentages). An even smaller fraction of reports require correction after On-Site investigates a 9 dispute. Id. 10 In sum, Plaintiff has failed to prove numerosity under Rule 23(a)(1). 11 B. Plaintiff is an inadequate class representative. 12 Plaintiff’s class certification motion fails for yet another reason: Plaintiff has failed to 13 prove that he is an adequate representative of the putative classes. To the contrary, Plaintiff is an 14 inadequate representative. A class may be certified only if “the representative parties will fairly 15 and adequately protect the interests of the class.” Fed. R. Civ. P. 23(a)(4). “Basic consideration of 16 fairness require that a court undertake a stringent and continuing examination of the adequacy of 17 representation by the named class representatives at all stages of the litigation where 18 absent members will be bound by the court’s judgment.” Susman v. Lincoln Am. Corp., 561 F.2d 19 86, 89–90 (7th Cir. 1977) (quoting National Association of Regional Medical Programs v. 20 Mathews, 551 F.2d 340, 344–45 (D.C. Cir. 1976)). “To determine legal adequacy, we resolve two 21 questions: ‘(1) do the named plaintiffs and their counsel have any conflicts of interest with other 22 class members and (2) will the named plaintiffs and their counsel prosecute the action vigorously 23 on behalf of the class?’” In re Hyundai & Kia Fuel Econ. Litig., 926 F.3d 539, 566 (9th Cir. 2019) 24 (en banc) (quoting Hanlon v. Chrysler Corp., 150 F.3d 1011, 1020 (9th Cir. 1998)). 25 Plaintiff cannot “prosecute the action vigorously on behalf of the class” because he has 26 repeatedly made false statements—including several under penalty of perjury—about key facts 27 underlying his FCRA claims. Id. As this Court and others have held, a putative class representative 1 is inadequate where, as here, “the representative’s credibility is questioned on issues directly 2 relevant to the litigation or there are confirmed examples of dishonesty.” In re Arris Cable Modem 3 Consumer Litig., 327 F.R.D. 334, 356 (N.D. Cal. 2018) (quoting Harris v. Vector Mktg. Corp., 753 4 F. Supp. 2d 996, 1015 (N.D. Cal. 2010)), leave to appeal denied sub nom. Reyna v. Arris Int’l 5 PLC, No. 18-80099, 2018 WL 6167340 (9th Cir. Nov. 8, 2018). The reason for this rule is that 6 “there is a danger that absent class members will suffer if their representative is preoccupied with 7 defenses unique to [Plaintiff].’” Weidenhamer v. Expedia, Inc., 2015 WL 7157282, at *19 (W.D. 8 Wash. Nov. 13, 2015) (quoting Hanon v. Dataproducts Corp., 976 F.2d 497, 508 (9th Cir. 1992)); 9 see, e.g., Savino v. Computer Credit, Inc., 164 F.3d 81, 87 (2d Cir. 1998) (affirming denial of class 10 certification because “the fact that [plaintiff] offered differing accounts about the letters that form 11 the very basis for his lawsuit surely would create serious concerns as to his credibility at any 12 trial.”). Here, in court filings and sworn statements under penalty of perjury, Plaintiff has 13 misrepresented not only the felony convictions underlying the allegedly inaccurate consumer 14 report that On-Site produced, but also the alleged injury that consumer report caused. The Court 15 analyzes Plaintiff’s convictions and alleged injury in turn. 16 Plaintiff misrepresented his convictions in his complaint and class certification motion. In 17 his complaint, Plaintiff alleged that his convictions had been “set aside and expunged from his 18 record” so that “[u]nder the laws of the State of Arizona, all rights he had before his troubles with 19 the law have been restored to him, and he carries forward no disabilities from his prior troubles 20 with the law.” Compl. ¶ 14. Based on this allegation, Plaintiff claimed that “Defendants [sic] 21 violated [15 U.S.C. §] 1681e(b) in its consumer reports issued to its customers by reporting 22 convictions which had been expunged, sealed, or otherwise removed from public dissemination.” 23 Id. ¶ 54. In particular, On-Site allegedly violated FCRA by reporting Plaintiff’s “set aside” 24 convictions to Indigo Lofts, an apartment complex where Plaintiff had applied to live. Id. ¶ 17. 25 Plaintiff then doubled down on this argument in his class certification motion. There, 26 Plaintiff argued that “under A.R.S. § 13-907, all of his convictions had previously been set aside as 27 a matter of Arizona law” and so should not have been reported to Indigo Lofts. Mot. at 5. 1 Yet Plaintiff now concedes, as he must, that much of what he represented about his 2 convictions is false. Specifically: 3 • Plaintiff represented that “all rights he had before his troubles with the law have been restored to him.” Compl. ¶ 14. In reality, that was “not a true statement,” Plaintiff 4 confirmed at his deposition. Dep. Tr. 81:14–82:2, ECF No. 55-2. 5 • Plaintiff represented that On-Site had reported convictions “removed from public 6 dissemination.” Compl. ¶ 54. In truth, Plaintiff’s convictions are still “publicly available on the Arizona courts website.” Dep. Tr. 45:4–6. 7 • Plaintiff represented that his convictions had been “expunged from his record,” Compl. 8 ¶ 14, and that “under A.R.S. § 13-907, all of his convictions had previously been set 9 aside.” Mot. at 5. However, Arizona law is clear that “section 13-907 does not expunge or remove the fact of conviction in Arizona.” Russell v. Royal Maccabees Life Ins. Co., 974 10 P.2d 443, 446 (Ariz. Ct. App. 1998) (emphasis in original). “A person whose felony 11 conviction has been set aside under A.R.S. § 13-907 must still disclose it when asked if the person has ever been convicted of a felony.” Parsons v. Arizona Dep’t of Health Servs., 395 12 P.3d 709, 712 (Ariz. Ct. App. 2017). 13 The above falsehoods are not even the most egregious, however. Plaintiff’s worst 14 falsehoods relate to the alleged injury that On-Site’s consumer report inflicted on Plaintiff. 15 Plaintiff repeatedly averred under penalty of perjury or represented that On-Site’s consumer report 16 resulted in Indigo Lofts denying his tenancy application. In truth, Indigo Lofts approved Plaintiff’s 17 tenancy application. Specifically: 18 • In response to Interrogatory No. 3, Plaintiff averred under penalty of perjury that “[a]fter 19 the consumer report was conducted, the apartment manager and/or property manager at Indigo notified claimants that the consumer report had contained numerous criminal 20 offenses including federal convictions for sexual harassment[.] As a result, Plaintiff was 21 informed that his application would be denied.” Dep. Tr. 143:2–12 (quoting Interrogatory No. 3). Yet at Plaintiff’s deposition, Plaintiff admitted “[i]t is false to say Plaintiff was 22 informed that his application would be denied.” Id. at 144:11–13. 23 • Similarly, in his declaration supporting the class certification motion, Plaintiff averred 24 under penalty of perjury that “because of the incident with my wife and the manager at Indigo Lofts, I knew that my application to rent would be denied.” Miller Decl. ¶ 24, ECF 25 No. 54-2. Again, Plaintiff admitted “that’s also false, correct, because [his] application 26 wasn’t denied by Indigo Lofts[.]” Dep. Tr. 138:13–14. Rather, Indigo Loft’s manager was simply rude to him. As Plaintiff testified, “it’s the way she had -- she had her attitude. I’m 27 not going to live somewhere like that. . . . That’s why we [Plaintiff and his wife] didn’t take it.” Id. at 138:15–17, 143:24. 1 • In Plaintiff’s class certification motion, Plaintiff represented that “[a]s a result of the items 2 contained in the Report [issued by On-Site], Brian Miller’s application for housing was 3 turned down.” Mot. at 5. Plaintiff admitted this representation was “false.” Id. at 171:8. 4 Given the many confirmed falsehoods detailed above, “[Plaintiff]’s credibility is 5 questioned on issues directly relevant to the litigation [and] there are confirmed examples of 6 dishonesty.” In re Arris Cable Modem Consumer Litig., 327 F.R.D. at 356 (N.D. Cal. 2018) 7 (quoting Harris, 753 F. Supp. 2d at 1015). Plaintiff is thus an inadequate class representative. 8 Accordingly, the Court cannot certify any of Plaintiff’s putative classes. 9 Indeed, courts have denied class certification for less. Three cases are illustrative. In 10 Kaplan v. Pomerantz, plaintiff gave false deposition testimony that possibly was “of marginal 11 relevance to th[e] lawsuit.” 132 F.R.D. 504, 510 (N.D. Ill. 1990) (Rovner, Circuit J.). Still, then- 12 District Judge Illana Rovner decertified the class. Judge Rovner reasoned that plaintiff had 13 “evince[d] a willingness to give intentionally false and misleading testimony in an effort to further 14 his interests in this litigation. Under these circumstances, allowing him to prosecute this case as a 15 class action would not be fair to the Court, to the defendants, or to the other individuals whose 16 interests plaintiff purports to represent.” Id. Here, allowing Plaintiff “to prosecute this case as a 17 class action” would be even more unfair because Plaintiff’s falsehoods are more than “of marginal 18 relevance to this lawsuit.” Id. Rather, Plaintiff’s false statements—including several under penalty 19 of perjury—exaggerated the inaccuracy of On-Site’s consumer reports and the harm those 20 consumer reports inflict. 21 Analogously, in Kline v. Wolf, 702 F.2d 400 (2d Cir. 1983), it was not even clear whether 22 both named plaintiffs had lied or whether their lies were relevant to all their class claims. One 23 named plaintiff “testified that he had relied upon a report which he now admits did not exist at the 24 time he allegedly relied upon it.” Id. at 403. The other named plaintiff had “a shadow over” her 25 testimony because a third-party disputed the testimony’s veracity. Id. Given plaintiffs’ dubious 26 testimony, the district court “made a preliminary determination that their credibility was 27 vulnerable to attack” and denied class certification. Id. 1 The Kline plaintiffs appealed, arguing that (1) “there was insufficient evidence to find that 2 the plaintiffs’ credibility was seriously in doubt,” id. at 402; and (2) their questionable testimony 3 “should not preclude their representation of the class since this testimony was irrelevant to their 4 alternate claim.” Id. at 403. The Second Circuit disagreed. In response to plaintiffs’ first argument, 5 the Second Circuit reasoned that even if the false “testimony was the product of an innocent 6 mistake, it subjects [plaintiff’s] credibility to serious question.” Id. In response to plaintiffs’ 7 second argument, the Second Circuit reasoned that “[s]ince plaintiffs’ testimony on an issue 8 critical to one of their two causes of action was subject to sharp attack, the district court 9 reasonably concluded that their credibility in general was sufficiently in doubt to justify denying 10 them a fiduciary role as class representatives with respect to both claims.” Id. Thus, the Second 11 Circuit affirmed the district court’s denial of class certification. 12 Here, it is harder to assume that Plaintiff’s false statements were “the product of an 13 innocent mistake.” Id. Plaintiff repeated the falsehoods unequivocally, repeatedly, in writing, and 14 under penalty of perjury. Plaintiff did so in the Complaint, a sworn interrogatory response, his 15 class certification motion, and a sworn declaration supporting his class certification motion. 16 Moreover, Plaintiff’s falsehoods are not “irrelevant to [an] alternate claim” for which Plaintiff 17 seeks class certification. Id. Plaintiff seeks class certification under two FCRA provisions: 15 18 U.S.C. §§ 1681e(b) & 1681c(a)(5). See Compl. ¶¶ 53, 63. Under either provision, it is relevant 19 whether a consumer report accurately reports convictions. See id. §§ 1681e(b) (requiring 20 “reasonable procedures to assure maximum possible accuracy”) & 1681c(a)(5) (exempting from 21 liability reports on “records of convictions”). Thus, it is relevant that Plaintiff misrepresented 22 several of his convictions as “expunged,” “removed from public dissemination,” and thus exempt 23 from On-Site’s reporting. 24 The last illustrative case is also a FCRA case. In Searcy v. eFunds Corp., defendants 25 submitted a declaration that “cast significant doubt on the veracity of [plaintiff]’s sworn 26 statement.” No. 08-CV-985, 2010 WL 1337684, at *1 (N.D. Ill. Mar. 31, 2010). The Searcy Court 27 thus decertified the two FCRA classes it had certified only a couple months before. Id. at *6. The 1 Searcy Court decertified both classes even “[]though [plaintiff] did not offer the problematic 2 || declaration in support of her adequacy to represent [one] class.” /d. at *5. In the court’s view, 3 “(plamtiff]’s credibility issues likewise affect her ability to adequately represent the interests of 4 || that class.” /d. Here, by contrast, Plaintiff’s credibility issues affect his ability to adequately 5 || represent the interests of a// the FCRA classes. 6 With all this authority against him, Plaintiff relies extensively on one district court case. 7 Plaintiff claims that in McIntyre v. On-Site, No. 18-CV-3934, 2020 WL 5017612, at *8 (E.D. Pa. 8 || Aug. 25, 2020), the district court “certified a class against On-Site’s inaccuracies related to 9 || reporting of evictions on the very same consumer reports that are at issue here.” Mot. at 10. 10 || Plaintiff overstates McIntyre. Far from addressing “the very same consumer reports that are at 11 issue here,” Mc/ntyre analyzed reports on eviction records, not convictions. See McIntyre, 2020 12 || WL 5017612, at *1. Regardless, Plaintiff’s reliance on Mc/ntyre is misplaced. In McIntyre, On- 13 Site failed to contest numerosity and “there [was] no indication” that plaintiff was an inadequate S 14 || representative. /d. at *10, 13. Here, by contrast, On-Site has vigorously contested numerosity, and 3 15 Plaintiff is a patently inadequate class representative. A 16 All told, the Court denies class certification for two independent reasons. First, Plaintiff 2 17 has failed to prove that any putative class is “so numerous that joinder of all members is Z, 18 impracticable.” Fed. R. Civ. P. 23(a)(1). Second, Plaintiff is an inadequate class representative. 19 || IV. CONCLUSION 20 For the foregoing reasons, the Court DENIES Plaintiff’s motion for class certification. 21 IT IS SO ORDERED. 22 23 Dated: November 25, 2020 24 J uy K ‘ beh. LUCY. KOH 25 United States District Judge 26 27 28 20 Case No. 19-CV-O2114-LHK