Andrenia BURGIS, Christopher Burgos, Leticia Smith, Samuel Duncan, Alonzo Hudgins, Rashid Smith, Doren Pink, Anthony Joseph, Israel Dejesus, on their own behalf and on behalf of others similarly situated, Plaintiffs-Appellants, v. NEW YORK CITY DEPARTMENT OF SANITATION, John J. Doherty, Defendants-Appellees, John Does, 1-10, Defendants, 798 F.3d 63

JED S. RAKOFF, Senior District Judge: Plaintiffs are individuals employed by the New York City Department of Sanitation (“DSNY”), who allege that the defendants, the City of New York and John J. Doherty, Commissioner of the DSNY, discriminated against them and others similarly situated on the basis of their race and/or national origin in the DSNY’s promotional practices. Plaintiffs bring this putative class action suit under the Fourteenth Amendment Equal Protection Clause, under 42 U.S.C. § 1981, under Title VII (42 U.S.C. § 2000e), and under the New York State and City human rights laws.

The district court dismissed all of plaintiffs’ claims. We affirm. BACKGROUND The second amended complaint (“SAC”) makes the following factual allegations. After the DSNY promotes individuals in its ranks of sanitation workers to supervisors, it may then promote those supervisors to the position of general superintendent, ranging from general superintendent level 1 (the lowest level) to general superintendent level 4 (the highest level).

Prior to 1979, DSNY based promotion to supervisor and to general superintendent levels 1-3 on written examinations, but after 1979 the examinations were eliminated for promotions to levels 2 and 3. Promotions to levels 2, 3, and 4 are now done on the basis of-recommendations. Plaintiffs allege that by using recommendations in this way, the DSNY has created a supervisory workforce that is not representative of the racial and/or national origin composition of the sanitation worker workforce.

The SAC includes the following chart, which summarizes the racial makeup of sanitation workers, supervisors, and general superintendents in Fiscal Year 2011: Title % White % Black % Hispanic Sanitation Worker 56 23.5 18 Supervisor 81 11 10 General Superintendent Level 1 81 13 9 General Superintendent Level 2 & 3 91 4 3 General Superintendent Level 4 80 10 10 Plaintiffs further allege that there are qualified candidates for supervisor in the same percentages racially as the percentages for these races that are present in the sanitation worker workforce as a whole.

Finally, they allege that even with respect to the positions requiring a written test (i e., supervisor and general superintendent level 1), the selection process invites some subjective evaluation of candidates. As to the individual plaintiffs, the SAC alleges as follows: Plaintiff Andrenia Burgis, a Black female, started as a sanitation worker in 1998; she was promoted to supervisor in 2003, and, in 2007, she was promoted to general superintendent level 1. By 2009, Ms. Burgis had obtained the prerequisites for promotion to level 2, but instead she was replaced in her command at level 1 by a White male.

He was subsequently promoted to level 2. She did not obtain another command until 2012, and a number of less qualified Wdiite individuals were and have since been promoted. Plaintiff Christopher Burgos, a Hispanic male, started at DSNY as a sanitation worker in 2000; he was promoted to supervisor in 2006. He took the exam for promotion to level 1 and is currently on the list for promotion. Plaintiff Leticia Smith, a Hispanic female, started at DSNY as a sanitation worker in 1995; she was promoted to supervisor in 2001, and she was promoted to general superintendent level 1 in 2007.

She has never been recommended for level 2, though she has the requisite experience and Wfiiite females and males with less experience have been promoted. Plaintiff Samuel Duncan, a Hispanic/Black male, started at DSNY as a sanitation worker in 2001; he was promoted to supervisor in 2008. He was demoted during a probationary period, while White supervisors in the same position were allowed to finish probation and remain as supervisors.

Plaintiff Alonzo Hudgins, a Black male, started at DSNY as a sanitation worker in 1995; he was promoted to supervisor in 2001, and was subsequently promoted to general superintendent level 1. He has never been recommended to level 2, even though Wfiiite males in his borough have been recommended and promoted. Plaintiff Rashid Smith, a Black male, started at DSNY as a sanitation worker in 1995; he was promoted to supervisor in 2000, and he was subsequently promoted to general superintendent level 1.

He has never been recommended for level 2, even though less qualified Wfiiite males in his borough have been recommended and promoted. Plaintiff Doren Pink, a Black male, started at DSNY as a sanitation worker in 1999; he was promoted to supervisor in 2004. He took the test for promotion to general superintendent level 1, but has not been promoted, though many of his Wdiite counterparts who took the test at the same time have been promoted despite being less qualified. Plaintiff Anthony Joseph, a Black male, started at DSNY as a sanitation worker in 1989; he was subsequently promoted to supervisor.

Despite having passed the exam, he has not been promoted to general superintendent level 1, while White employees with lower scores have been promoted. Plaintiff Israel DeJesus, a Hispanic male, started at DSNY as a sanitation worker in 1995; he was promoted to supervisor in 2005, and he was promoted to general superintendent level 1 in 2008. He has never been recommended for level 2, even though less qualified White employees have been recommended and promoted.

As for individual defendant Doherty, plaintiffs allege that he “at all times, was aware that his supervisory workforce was skewed and did not reflect either the racial or national origin makeup of DSNY’s non-supervisory workforce, or the general population of the City of New York. Doherty, reflecting his approval of this disparate classification, has taken no action to change it, and has, therefore condoned the discrimination.” On March 31, 2014, the district court granted defendants’ motion to dismiss.

With respect to the § 1981 and Equal Protection claims, the district court held that plaintiffs had failed to sufficiently allege discriminatory intent. As to the City and defendant Doherty in his official capacity, the district court further found that plaintiffs had not sufficiently pleaded that the alleged discrimination was the result of an official policy, custom, or practice.

As to defendant Doherty in his individual capacity, the district court concluded that plaintiffs had not adequately alleged that he was personally involved in the alleged discriminatory promotion practices and only made conelusory allegations that he knew the racial composition of the workforce. Finally, the district court dismissed plaintiffs’ Title VII disparate impact claim for failure to exhaust administrative remedies.

DISCUSSION We review da novo a grant of a motion to dismiss under Rule 12(b)(6), accepting as true all well-pleaded allegations in the complaint and drawing all reasonable inferences in plaintiffs’ favor. Anschutz Corp. v. Merrill Lynch & Co., 690 F.3d 98, 107 (2d Cir.2012). However, the complaint “must contain sufficient factual matter ... to state a claim to relief that is plausible on its face.” Ashcroft v. Iqbal, 556 U.S. 662, 678, 129 S.Ct. 1937, 173 L.Ed.2d 868 (2009) (internal quotation marks omitted).

I To state a discrimination claim under the Fourteenth Amendment Equal Protection Clause and/or § 1981, plaintiffs must sufficiently allege that defendants acted with discriminatory intent. Gen. Bldg. Contractors Ass’n v. Pennsylvania, 458 U.S. 375, 391, 102 S.Ct. 3141, 73 L.Ed.2d 835 (1982). Here, plaintiffs fail to allege in other than conelusory fashion any specific instances of discrimination with respect to any individual plaintiff or others similarly situated.

Indeed, each plaintiff has at some point been promoted from a sanitation worker to supervisor, and five of the nine were subsequently promoted to general superintendent level 1. While the SAC generally alleges with respect to seven plaintiffs that they have been passed over for subsequent promotions while White individuals, who were allegedly less qualified, were promoted, the SAC fails to provide meaningful specifics of the alleged difference in qualifications, let alone discriminatory intent. With respect to the other two plaintiffs, plaintiffs Burgos and Duncan, the SAC alleges even less.

It simply alleges that plaintiff Burgos was promoted to supervisor in 2006, has taken the exam for general superintendent level 1, and is currently on the list for promotion. Plaintiff Duncan, it alleges, was promoted to supervisor but subsequently was demoted during a probationary period while White individuals in an allegedly similar situation were not demoted. These bare allegations do not present circumstances that “give rise to an inference of unlawful discrimination.” Texas Dep’t of Cmty. Affairs v. Burdine, 450 U.S. 248, 253, 101 S.Ct. 1089, 67 L.Ed.2d 207 (1981).

Without any specificity as to the qualifications considered for each position and without any reference to specific statements or individual circumstances that suggest discriminatory treatment, plaintiffs’ allegations do not support a finding that defendants acted with a discriminatory purpose. For all that one can tell from the SAC (plaintiffs’ “third bite at the apple”), it is equally possible that plaintiffs have not been promoted' for valid, nondiscriminatory reasons.

See Iqbal, 556 U.S. at 678, 129 S.Ct. 1937 (“Where a complaint pleads facts that are merely consistent with a defendant’s liability, it stops short of the line between possibility and plausibility of entitlement to relief.” (internal quotation marks omitted)). Plaintiffs argue, however, that statistics alone may be sufficient to warrant a plausible inference of discriminatory intent if they show a pattern or practice that cannot be explained except on the basis of intentional discrimination.

This is an issue of first impression for this Circuit in the context of a putative class action alleging employment discrimination under § 1981 and/or the Equal Protection Clause. We now hold that, as some of our Title VII cases have hinted, in certain circumstances (described below), statistics alone may be sufficient. Cf. United States v.

City of New York, 717 F.3d 72, 84 (2d Cir.2013) (noting that while, in a Title VII disparate treatment pattern or practice case, “instances of discrimination against particular employees are relevant to show a policy of intentional discrimination, they are not required; a statistical showing of disparate impact might suffice”); see also Hudson v. Int’l Bus. Machines Corp., 620 F.2d 351, 355 (2d Cir.1980) (stating, in a case preceding Iqbal, that “in class actions alleging racially discriminatory employment practices, statistical evidence may establish a prima facie case of discriminatory effect”).

However, to show discriminatory intent in a § 1981 or Equal Protection case based on statistics alone, the statistics must not only be statistically significant in the mathematical sense, but they must also be of a level that makes other plausible non-discriminatory explanations very unlikely. Cf. Hazelwood Sch. Dist. v. United States, 433 U.S. 299, 307-08, 97 S.Ct. 2736, 53 L.Ed.2d 768 (1977) (holding that to “constitute prima facie proof of a pattern or practice of discrimination,” a plaintiff must show “gross statistical disparities”); Ottaviani v.

State Univ. of New York at New Paltz, 875 F.2d 365, 371 (2d Cir.1989) (“The threshold question in [Title VII pattern-or-practice] disparate treatment cases, then, is: ‘At what point is the disparity in selection rates sufficiently large, or the probability that chance was the cause sufficiently low, for the numbers alone to establish a legitimate inference of discrimination?’ ” (quoting Palmer v. Shultz, 815 F.2d 84, 92 (D.C.Cir.1987) (alterations omitted))). In the instant case, plaintiffs have failed to allege statistics that meet the standards articulated above.

Among other shortcomings, the statistics provided by plaintiffs show only the raw percentages of White, Black, and Hispanic individuals at each employment level, without providing any detail as to the number of individuals at each level, the qualifications of individuals in the applicant pool and of those hired for each position, or the number of openings at each level. See Lomotey v.

CT Dep’t of Transp., 355 Fed.Appx. 478, 481 (2d Cir.2009) (summary order) (“[Plaintiffs] evidence that only Caucasians were selected for [job] placements amounts to nothing more than raw numbers which, without further information on key considerations such as the racial composition of the qualified labor pool, cannot support an inference of discrimination.”). Furthermore, the fact that each of the plaintiffs has been promoted at some point to the position of supervisor undermines their allegations of discrimination in the promotion of sanitation workers to supervisors.

Five of the nine plaintiffs have also been promoted from supervisor to general superintendent level 1, again calling into doubt their allegations of discrimination at that phase, especially given that the racial composition between individuals in the supervisor and general superintendent level 1 positions is essentially the same. While there is an increase in racial and national origin disparities between general superintendent level 1 and levels' 2 and 3, the racial composition returns at level 4 to essentially the same percentage as at level l.

Accordingly, the district court’s decision granting defendants’ motion to dismiss the Equal Protection and § 1981 claims is affirmed. II Turning 'to plaintiffs’ Title VII claim, here discriminatory intent need not be alleged in a case based on so-called “disparate impact.” But we need not decide whether plaintiffs’ statistics show any disparate impact, because we agree with the district court that the Title VII claim must be dismissed for failure to exhaust administrative remedies.

To bring a claim under Title VII, a plaintiff must first have filed a complaint with the Equal Employment Opportunity Commission (“EEOC”) or a state equivalent — here, the New York State Division of Human Rights (“SDHR”). See 42 U.S.C. § 2000e-5. Courts may only hear claims “reasonably related” to allegations set forth in the administrative complaint. Kirkland v. Buffalo Bd. of Educ., 622 F.2d 1066, 1068 (2d Cir.1980). Here, plaintiffs Burgis and Bur-gos submitted complaints of racial discrimination to the SDHR and received letters to sue from the EEOC.

However, plaintiff Burgos’s administrative complaint does not allege discrimination in defendants’ promotion practices. And while plaintiff Burgis’s complaint alleges that she was discriminated against on the basis of race when defendants removed her as general superintendent level 1 for approximately three years before giving her another assignment, and by failing to promote her to level 2 when the White male who initially replaced her was promoted, her complaints of individualized disparate treatment are not reasonably related to a Title VII disparate impact claim alleging class-wide discrimination. Woodman v.

WWOR-TV, Inc., 293 F.Supp.2d 381, 390 (S.D.N.Y. 2003), aff'd, 411 F.3d 69 (2d Cir.2005). Accordingly, plaintiffs failed to exhaust administrative remedies prior to filing the Title VII disparate impact claim in this action, and the district court order dismissing plaintiffs’ Title VII claim is affirmed. CONCLUSION For the reasons set forth above, the judgment of the district court is AFFIRMED. .

The complaint names the DSNY as a defendant in the lawsuit; however, as the New York City Charter requires all actions to be brought against the City, as opposed to any agency of the City, the district court correctly dismissed the DSNY as a party and construed the claims as though they were asserted against the City of New York. .On October 4, 2013, plaintiffs moved to substitute a "revised” version of the first amended complaint. The district court allowed the filing but correctly referred to the revised first amended complaint as the second amended complaint. .

We have held that statistics alone are not sufficient to allege discriminatory intent in § 1981 or Equal Protection cases brought by an individual only on her own behalf, Martin v. Citibank, N.A., 762 F.2d 212, 218 (2d Cir.1985), or in cases brought against a state official in her individual capacity, Reynolds v. Barrett, 685 F.3d 193, 204-05 (2d Cir.2012).

In these situations, we have held, broad statistical evidence, without evidence of specific instances of discrimination, is not relevant to whether any specific individual acted with discriminatory intent or whether any specific individual was the target of discrimination. . In Hazelwood, the Supreme Court held that measuring the significance of statistical disparities "involves calculation of the 'standard deviation’ as a measure of predicted fluctuations from the expected value of a sample.” 433 U.S. at 308 n. 14, 97 S.Ct. 2736.

As a general rule, "if the difference between the expected value and the observed number is greater than two or three standard deviations, then the hypothesis that [the alleged discriminatory action was taken] without regard to race would be suspect.” Id. (internal quotation marks omitted); see also Ramona L. Paetzold & Steve L. Willborn, The Statistics of Discrimination § 4:11 (2014). . Plaintiffs alternatively argue that the "rule of exclusion” test established in Castaneda v. Partida, 430 U.S. 482, 97 S.Ct. 1272, 51 L.Ed.2d 498 (1977), applies here.

Under this test, in asserting a claim of discrimination in the composition of grand juries, a plaintiff must 1) show that the group is a "recognizable, distinct class, singled out for different treatment under the laws,” and 2) prove the degree of underrepresentation “by comparing the proportion of the group in the total population to the proportion called to serve ..., over a significant period of time.” Id. at 494, 97 S.Ct. 1272.

This rule is hardly applicable here, where individuals are not promoted at random from the general population, but are considered for promotion on the basis of their individual qualifications.

"When special qualifications are required to fill particular jobs, comparisons to the general population (rather than to the smaller group of individuals who possess the necessary qualifications) may have little probative value.” Hazelwood, 433 U.S. at 308 n. 13, 97 S.Ct. 2736. .Plaintiffs' § 1981 claim and Equal Protection claim (asserted under § 1983) against the City and defendant Doherty in his official capacity also fail, because they have not alleged the discriminatory practice was caused by a municipal policy, custom or practice.

"[W]hen the defendant sued for discrimination under § 1981 or § 1983 is a municipality — or an individual sued in his official capacity — the plaintiff is required to show that the challenged acts were performed pursuant to a municipal policy or custom.” Patterson v. Cty. of Oneida, 375 F.3d 206, 226 (2d Cir.2004) (internal citations omitted). Here, plaintiffs again rely on statistical evidence to infer a municipal policy of discrimination.

For the same reasons discussed above, the statistics are not sufficient to give rise to an inference of discriminatory intent, and, as the district court found, plaintiffs do not even describe the discriminatory policy, custom, or practice on which their complaint is based. Their conclusory allegations that defendant Doherty knew about the discriminatory practice, without any allegation as to how he knew or should have known of it is not sufficient to establish a custom or policy.

Furthermore, when compared to the fact that the statistical allegations are insufficient to show discriminatory intent on the part of the municipality, they are likewise insufficient to show liability by Doherty in his individual capacity. See Reynolds, 685 F.3d at 204-05. Accordingly, the district court's dismissal of the case against him individually is affirmed. .

In an appropriate case, the facts pleaded in an administrative charge may well encompass a disparate impact claim even if the plaintiff has only ''labeled” her claim as "disparate treatment.” To be reasonably related, a claim must be one that reasonably would be expected to fall within the scope of the conduct complained of in the administrative charge. See Ximines v. George Wingate High Sch., 516 F.3d 156, 158-59 (2d Cir.2008) (per curiam).

Here, however, the conduct about which Burgis and Burgos complained to the SDHR would not reasonably have led that agency to investigate — beyond the facts of their individual cases — the effects on a protected class of DSNY's promotional practices.