FRANK MAAS, United States Magistrate Judge.
Adam T. Klein, Justin Mitchell Swartz, Lewis M. Steel, Melissa E. Pierre–Louis, Ossai Miazad, Samuel Rand Miller, Amber Christine Trzinski, Rachel Megan Bien, Sally Jasmine Abrahamson, Outten & Golden, LLP, Darius Charney, Lansner & Kubitschek, Jackson Chin, Puerto Rican Legal Defense and Education Fund, Inc., Judith Melissa Whiting, The Legal Action Center, Paul Francis Keefe, Community Service Society, Richard F. Bellman, Anti–Discrimination Center of Metro New York, Inc., New York, NY, Audrey Jordan Wiggins, Jane Dolkart, Ray P. McClain, Sarah Crawford, Lawyers' Committee for Civil Rights Under Law, Michael T. Kirkpatrick, Public Citizen Litigation Group, Washington, DC, Robert T. Coulter, Indian Legal Resource Center, Helena, MT, Sharon Dietrich, Community Legal Services, Inc., Philadelphia, PA, for Plaintiffs.
Louis Anthony Pellegrino, U.S. Attorney Office, Natalie Nancy Kuehler, Tara Marie La Morte, Daniel Post Filor, Tomoko Onozawa, New York, NY, for Defendant.
MEMORANDUM DECISION AND ORDERFRANK MAAS, United States Magistrate Judge.
In this putative class action, Plaintiffs, individually and on behalf of others similarly situated, allege that the process by which the United States Census Bureau (“Census Bureau”) screens applicants for temporary jobs for the Decennial Census is racially discriminatory and therefore violates Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e et seq. (“Title VII”). Specifically, the Plaintiffs challenge (a) the Census Bureau's policy requiring all applicants with criminal records to provide “official court documentation” of their prior arrests and convictions within thirty days after their receipt of a demand letter, and (b) the criteria the Census Bureau uses to determine whether an applicant who complies with such a demand is suitable for employment. The Plaintiffs contend that the Census Bureau's screening practices are neither job-related nor consistent with business necessity, and disproportionately preclude African–Americans and Latinos from obtaining employment with the Census Bureau because these groups have higher arrest and conviction rates than Caucasians.
On July 8, 2013, the Plaintiffs filed a motion to certify this action as a class action under Rule 23 of the Federal Rules of Civil Procedure. (ECF No. 166). On December 16, 2013, before the parties finished briefing the class certification motion, the Census Bureau filed a motion to dismiss for lack of subject matter jurisdiction. (ECF No. 225). As grounds for that motion, the Census Bureau alleges that each of the named plaintiffs lacks constitutional standing to pursue relief under Title VII. Both motions are now fully submitted.
The parties have filed certain exhibits under seal pursuant to a stipulated protective order. (ECF No. 91).
I. Background
A. Census Bureau
The Census Bureau is an agency of the United States Department of Commerce. (Who We Are, United States Census Bureau, http://www.census.gov/aboutus/who. html (last visited July 1, 2014)). Every ten years, as Article I, Section 2 of the United States Constitution requires, the Census Bureau conducts a Population and Housing Census, commonly known as the Decennial Census. (About What We Do, United States Census Bureau, http://www. census.gov/aboutus (last visited July 1, 2014); What is the Census?, United States Census 2010, http://www.census.gov/2010census/ about (last visited July 1, 2014)). The Government uses the data collected during each Decennial Census to apportion seats in the United States House of Representatives, to inform redistricting decisions, and to allocate substantial amounts of federal funding. (ECF No. 204 (Decl. of Ass't U.S. Att'y Tara La Morte, dated Oct. 28, 2013), Exs. 1 at USA38988, 2 at USA 7667).
B. Census Bureau Hiring Process
1. Initial Application StageThe 2010 Decennial Census required the Census Bureau to fill over 1.3 million temporary positions nationwide between October 2008 and September 2010. (ECF No. 205 (Decl. of Viola Lewis Willis, dated Oct. 28, 2013 (“Willis Decl. I”)), ¶ 2). Recruiting and hiring for these positions was conducted on a regional basis by “Local Census Offices” (“LCOs”). (Id. ¶ 7). Although each LCO had its own hiring needs, applicants across the country all were subject to the same initial application procedures and had to meet the same basic qualifications. At the outset, each applicant had to complete an application form to confirm that the applicant was over eighteen years of age, had a Social Security number, and had registered for the selective service if he was a male born after 1959. (ECF Nos. 169–170 (Decl. of Ossai Miazad, Esq., dated June 28, 2013 (“Miazad Decl. I”)), Ex. 71 at USA32563–67). The application form further inquired as to the applicant's citizenship status, hours of availability, means of transportation, former military service, and foreign language skills, although none of these criteria was dispositive at the initial application stage. (Id. ).
After submitting initial applications, applicants were required to take a written test, which the Census Bureau used to ascertain whether the applicant could follow written instructions, perform simple mathematical computations, and complete other job-related tasks. (ECF No. 227 (Decl. of Viola Lewis Willis, dated Dec. 16, 2013 (“Willis Decl. II”)), Ex. A at USA1758). To qualify for a spot in the applicant pool, candidates generally had to score 70 or higher on the exam, inclusive of any “preference points” awarded to veterans. (Miazad Decl. I, Ex. 56 at USA 45163; Willis Decl. I, ¶ 10). Applicants who scored below 70 were permitted to take the test again in an effort to improve their scores. (Willis Decl. I, ¶ 4). The highest score that an applicant achieved would supersede all prior test scores on file. (Id. ).
2. Criminal Background Screening and AdjudicationApplicants who submitted an application and completed the written exam were required to undergo a criminal background check. (ECF No. 206 (Decl. of Sandra Patterson, dated Oct. 28, 2013 (“Patterson Decl.”)), ¶ 3). At the outset of the process, the Census Bureau ran each applicant's name, date of birth, and Social Security number through the criminal history database of the Federal Bureau of Investigation (“FBI”). (Id. ). The Plaintiffs allege that this “namecheck” process often returned multiple “hits” for a single applicant and resulted in false positives for applicants who had common names or had used aliases in the past. (See ECF No. 176 (Pls.' Mem. of Law in Supp. of Mot. for Class Cert. (“Pls.' Class Cert. Mem.”)) at 15).
If the FBI database returned a criminal history for a particular applicant, LCO staff members would review its contents using certain criteria to determine whether to treat the applicant as immediately available for hire, or to request further information from the applicant. (Patterson Decl., ¶¶ 4–5). According to the Plaintiffs, this namecheck adjudication did not consider the disposition of the applicant's cases, the amount of time that had passed since the arrests occurred, the nature of the applicant's offenses, or whether the position the applicant sought involved interaction with the public. (ECF No. 125 (Second Am. Compl. (“SAC”)), ¶¶ 2, 15–16). If the LCO staff determined that more information was necessary, the Census Bureau sent a form letter (the “30–day Letter”) requiring the applicant to provide “official court documentation on any and all arrest(s) and/or conviction(s)” within thirty days. (Patterson Decl., ¶ 5; Miazad Decl. I, Ex. 70 at USA43538). If the applicant disputed the “identity of the arrest record,” the 30–day Letter advised the applicant to submit a set of fingerprints within that period. (Miazad Decl. I, Ex. 70 at USA43538).
The Plaintiffs estimate that nearly 854,000 of the approximately 3.8 million applicants who applied for temporary employment during the 2010 Decennial Census received a 30–day Letter. (Id., Ex. 68). Only a small percentage of those applicants responded to the letter. (Id. ). Overall, the Plaintiffs contend that the 30–day Letter policy prevented ninety-three percent of the applicants with arrest records—approximately 700,000 people—from being considered for employment in connection with the 2010 Census. (SAC ¶ 2).
If an applicant did respond to the 30–day Letter, his application proceeded to the “adjudication” stage of the screening process. (Patterson Decl., ¶ 6–7). At this stage, the Census Bureau applied certain criteria (“Adjudication Criteria”) to determine whether the applicant was suitable for employment. (Id. ). Pursuant to these criteria, the Census Bureau automatically excluded applicants with prior arrests for nearly all felonies and most misdemeanors. (Miazad Decl. I, Ex. 92). To the extent that the criteria allowed any discretion, Census Bureau employees were instructed to “[d]efer anything [they felt] strongly about” or as to which they could not “make an unbiased decision.” (SAC ¶ 26). The Plaintiffs contend that the Adjudication Criteria were “arbitrarily drawn, dramatically overinclusive, unguided by professional principles or social science, and based largely on irrational instincts.” (Pls.' Class Cert. Mem. at 21). Moreover, because the adjudication stage often took months to complete, the Plaintiffs allege that many applicants who eventually were deemed “eligible for hire” nonetheless were never considered for a job because the Census Bureau had already filled all of the temporary positions by the time the applicants received clearance. (SAC ¶ 29).
3. Final Selection ProcessIf an applicant was deemed “eligible for hire” following the background check, LCO staff members entered the applicant's information into an electronic data management system known as the Decennial Applicant Personnel Pay System (“DAPPS”). (Willis Decl. I, ¶ 4). Once an applicant appeared in DAPPS, the Census Bureau assigned his application a numerical label, known as a “geocode,” that corresponded to the geographic area in which he lived. (Id. ¶¶ 9–10). LCO staff members could use DAPPS to filter applicants by geocode in order to limit their searches to applicants residing within the LCO's jurisdiction. (Id. ¶ 29).
To begin the hiring process, LCO staff members completed a “Requisition to Hire” form that identified the number of positions available, the applicable geographic area, and any required qualifying criteria, including hours of availability, foreign language proficiency, and access to transportation. Based on the information in the Requisition to Hire form, DAPPS generated a “selection record” that listed all qualified applicants within the LCO's geographic jurisdiction from highest to lowest test scores. (Id. ¶ 13). Each selection record listed a maximum of fifty applicants, so candidates whose test scores did not place them among the top fifty qualified applicants did not appear on the selection record. (Id. ¶ 26). Many selection records listed fewer than fifty candidates. (See id., Exs. A–H).
The first group of applicants on the selection record included all qualified candidates who received “veteran preferences” because they had significant service-related disabilities. (Id. ¶¶ 11, 14–15). LCO staff members were required to select these preference-eligible applicants over other applicants on the selection record, unless they provided a written justification for their belief that the preference-eligible applicant was not qualified for the position. (Id. ¶ 23). If there were not enough qualified preference-eligible applicants to fill the open positions, LCO staff members were required to offer interviews for the remaining positions to the highest listed applicants on the selection record. (Id. ¶ 24). If the position called for a particular foreign language proficiency, qualified non-citizen applicants appeared at the bottom of the selection record. (See Willis Decl. II, Ex. A at USA 1734). LCO staff members were permitted to offer interviews to these non-citizen candidates only if no qualified citizen applicants were available. (Id. at USA 1719).
LCO staff members were instructed to employ a “rule of three” when making hiring decisions. Although they had some discretion in deciding whether to extend an offer following an interview, the “rule of three” required that LCO staff always select from the top three available applicants. For example, if staff members interviewed the first, second, and third candidates listed on the selection record, they could choose to offer the position to the third candidate over the first and second candidates. However, the LCO staff members would have to offer a position to one of these three candidates before moving on to the fourth listed candidate. If the selected candidate declined the offer, the LCO staff could continue down the selection record, but at all times were required to offer the open position to one of the top three available applicants. (See ECF No. 247 (Decl. of Ossai Miazad, dated Feb. 10, 2014 (“Miazad Decl. II”)), Ex. 124 at USA5147–48).
C. Named Plaintiffs
The Second Amended Complaint names eight plaintiffs as putative class representatives: Evelyn Houser, Anthony Gonzalez, Ignacio Riesco, Precious Daniels, Felicia Rickett–Samuels, Chynell Scott, Vivian Kargbo and Scotty Desphy. (SAC at 2). Each of these named plaintiffs applied for, but ultimately was denied, temporary employment with the Census Bureau during the 2010 Decennial Census.
1. Evelyn HouserEvelyn Houser (“Houser”) is an African–American resident of Philadelphia, Pennsylvania. (Miazad Decl. I, Ex. 20 (“Houser Dep.”) at 4:23–24, 8:9–11). She worked as a temporary employee for the Census Bureau during the 1990 Decennial Census, and then reapplied for a similar position in January 2009. (Id. at 31:9–16; Patterson Decl., Ex. I at USA3880). She indicated in her application that she was a United States citizen and did not speak any foreign languages. (Patterson Deck, Ex. I at USA3878, USA3882). On the written exam, Houser earned a score of 72. (Id. at USA3886).
Two months after submitting her application, Houser received a 30–day Letter based on the results of her criminal background check. (Houser Dep. 73:18–20, 85:22–86:19). On May 21, 2009, the Census Bureau informed Houser that she had been disqualified because she “failed to provide the requested information within the 30[-]day window provided.” (Miazad Decl. I, Ex. 26).
LCO staff members in Houser's geographic region reviewed a total of eighty-four selection records after the date on which she applied. (Willis Decl. I, ¶ 44). Twenty-five of those selection records related to applicants with foreign language abilities. (Id. ¶ 45). The remaining fifty-nine selection records included candidates with test scores ranging from 70 to 105. (See id., Ex. A). Accordingly, had Houser passed the background check, her score would have been high enough to place her on at least one of those selection records. However, none of the applicants ultimately hired from those selection records received exam scores lower than 78. (Id. ¶ 47).
2. Anthony GonzalezAnthony Gonzalez (“Gonzalez”) is a Latino resident of Riverview, Florida. (Miazad Decl. I, Ex. 27 (“Gonzalez Dep.”) at 7:16–18). He is a United States citizen who speaks fluent Spanish. (Patterson Decl., Ex. N at USA3852, USA3855). Gonzalez applied for temporary employment with the 2010 Decennial Census in February 2010, and received a score of 92 on the qualifying exam. (Id. at USA3859). On or about March 11, 2010, Gonzalez received a 30–day Letter from the Census Bureau, to which he promptly responded. (Miazad Decl. I, Exs. 29, 30). Gonzalez nevertheless did not hear back from the Census Bureau. (Gonzalez Dep. 181:4–13).
DAPPS generated a total of twenty-three selection records encompassing Gonzalez's geocode after the date on which he applied. (Willis Decl. I, ¶ 86). Two of those selection records pertained to positions requiring fluency in Haitian–Creole or French. Candidates who appeared on the remaining selection records all earned scores of 93 or above on the qualifying exam, with the exception of a few applicants who qualified for veterans' preferences. (See id., Ex. F). Gonzalez thus did not score high enough to have been listed on any of the selection records.
3. Ignacio RiescoIgnacio Riesco (“Riesco”) is a Latino resident of Orlando, Florida. (Miazad Decl. I, Ex. 31 (“Riesco Dep.”) at 22:16–19, 97:18–19). He applied for a temporary position with the Census Bureau in April 2010. (Patterson Deck, Ex. P). In his application, Riesco indicated that he spoke fluent Portuguese and Spanish and was a lawful permanent resident, but not a United States citizen. (Id. at USA5595–96). Riesco received a score of 93 on the qualifying exam. (Willis Deck I, ¶ 97).
After Riesco submitted his application, the Census Bureau sent him a 30–day Letter based on the results of his background check. (Miazad Deck I, Ex. 35). Within a few days of receiving the letter, Riesco sent a copy of the disposition of the criminal charges that had triggered the letter, showing that all charges had been dropped. (Riesco Dep. 91:22–24). Despite his compliance with the 30–day Letter's directives, Riesco never heard anything further from the Census Bureau. (Id. at 122:2–7).
Because Riesco submitted his application near the end of the Census Bureau's hiring period, the LCO in Riesco's area generated only one selection record after he applied. (Willis Deck I, ¶ 99). The lowest scoring candidate that appeared on that selection record received a score of 95 on the qualifying test. (See id., Ex. H). Riesco thus would not have appeared on that selection record, even if he had been deemed eligible for hire.
4. Precious DanielsPrecious Daniels (“Daniels”) is an African–American resident of Detroit, Michigan. (Miazad Deck I, Ex. 1 (“Daniels Dep.”) at 6:9–12, 156:5–8). She applied for temporary employment with the Census Bureau in January 2010. (Patterson Deck, Ex M). Although she apparently testified that she was willing to work forty hours per week if necessary, her application stated that she would prefer to work twenty-eight hours per week. (Id. at USA3796). She further indicated that she did not speak any languages other than English. (Id. ). Daniels received a score of 83 on her written test. (Willis Decl. I, ¶ 78).
On February 16, 2010, the Census Bureau sent Daniels a 30–day Letter based on the results of her background check. (Miazad Decl. I, Ex. 3). Daniels promptly submitted the requested information, but the Census Bureau rejected her application in March 2010 “based on the nature of the facts disclosed” in her response. (Id., Ex. 5).
LCOs encompassing Daniels' geocode reviewed a total of thirty-four selection records after Daniels applied for employment with the Census Bureau. (Willis Decl. I, ¶ 79). Eleven of those records related to positions that required foreign language capabilities. (Id. ¶ 80). The remaining twenty-three selection records included candidates with test scores ranging from 72 to 107, some of whom, like Daniels, indicated a preference for working fewer than forty hours per week. (See id., Ex. E). Indeed, several applicants who scored below Daniels ultimately were hired from these selection records. (Id. at USA48405–06). According to the Census Bureau, however, none of the candidates hired from these selection records had both scored below Daniels on the qualifying exam and stated a preference for working fewer than forty hours per week. (Willis Decl. I, 183).
5. Felicia Rickett–SamuelsFelicia Rickett Samuels (“Rickett–Samuels”) is an African–American woman who lives in Stamford, Connecticut. (Miazad Decl. I, Ex. 36 (“Rickett–Samuels Dep.”) at 6:8–12, 7:3–4). She applied for temporary employment with the 2010 Decennial Census on January 13, 2009, and received a score of 88 on her written exam. (Willis Decl. I, ¶ 50). Her application indicated that she was a United States citizen and did not speak any foreign languages. (Patterson Decl., Ex. J at USA44224, USA44227).
Approximately two months after submitting her application, Rickett–Samuels received a 30–day Letter from the Census Bureau. (Miazad Decl. I, Ex. 40). She responded to that letter within the allotted time frame by providing a copy of a Certificate of Good Conduct and records from the Division of Criminal Justice Services related to the charges that triggered the letter. (Id., Ex. 38). Soon after Rickett–Samuels responded, the Census Bureau rejected her application based on the information she provided. (Rickett–Samuels Dep. 197:10–21).
LCO staff members ran a total of twenty-two selection records encompassing Rickett–Samuels' geocode after she applied. (Willis Decl. I, ¶ 51). Of those records, eleven were limited to applicants with foreign language abilities. (Id. ¶ 52). The remaining eleven selection records included candidates with scores ranging from 85 to 110. (See id., Ex. B). Rickett–Samuels' score thus was high enough to have placed her on one of those selection records. Nonetheless, according to the Census Bureau, none of the applicants ultimately hired from those selection records scored below 97. (Id. ¶ 58).
6. Chynell ScottChynell Scott (“Scott”) is an African–American resident of Philadelphia, Pennsylvania. (Miazad Decl. I, Ex. 41 (“Scott Dep.”) at 4:15–19, 14:23–24). She previously worked for the Census Bureau in various capacities in 2000 and 2004. (Id. at 53:2–57:23). In December 2009, she applied for a temporary position with the Census Bureau as part of the 2010 Decennial Census. (Id. at 76:9–13). Although the documents filed in this case do not indicate her test score, Scott apparently testified at her deposition that her score was approximately 100.
In her application, Scott was asked whether, in the last ten years, she had ever been “convicted.” She responded “no.” (Patterson Deck, Exs. H at USA43462, K at USA43550). On December 16, 2009, the Census Bureau sent Scott a 30–day Letter based on the results of a background check that revealed a prior arrest record. (Miazad Decl. I, Ex. 44). She responded to that letter by submitting a “Case Summary” from the Pennsylvania Court of Common Pleas indicating that she had pleaded guilty to disorderly conduct approximately two months before she completed her Census application. (Id., Ex. 42). Scott later testified that she did not know that her guilty plea had resulted in a “conviction” because the judge had reduced her charges to a fine. (Scott Dep. 89:3–24). On February 3, 2010, the Census Bureau rejected Scott's application “based on the nature of the facts [she] disclosed.” (Miazad Decl. II, Ex. 146).
The selection records for Scott's geographic region included candidates who scored between 70 and 102 on the qualifying exam. (See Willis Decl. I, Ex. C). With a score close to 100, Scott undoubtedly would have been listed on at least one of those selection records had she passed the criminal background check.
7. Vivian KargboVivian Kargbo (“Kargbo”) is an African–American woman who lives in Boston, Massachusetts. (Miazad Decl. I, Ex. 14 (“Kargbo Dep.”) at 7:20–24, 12:16–17). She applied for a temporary position with the Census Bureau in March 2010. (Patterson Decl., Ex. O). In her application, Kargbo indicated that she was a lawful permanent resident of the United States and did not have any foreign language skills. (Id. at USA43530, USA43533). She received a score of 80 on her qualifying exam. (Id. at USA43537).
On or about March 24, 2010, Kargbo received a 30–day Letter from the Census Bureau asking for further information regarding an arrest record that bore her name. (Miazad Decl. I, Ex. 18). Kargbo disputed the identity of the arrest record and sent the Census Bureau a copy of her fingerprints to support her challenge. (Id., Ex. 19). Kargbo never heard anything further from the Census Bureau despite her timely response to the 30–day Letter. (Kargbo Dep. 126:18–20).
LCO staff members reviewed two selection records encompassing Kargbo's geocode after the date of her application. (Willis Decl. I, ¶ 94). These two selection records included candidates with scores ranging from 92 to 105. (See id., Ex. G). Kargbo's score of 80 thus would not have placed her on either selection record. The lowest-scoring applicant selected from these records earned a score of 97 on the written exam. (Id. ¶ 94).
8. Scotty DesphyScotty Desphy (“Desphy”) is an African–American resident of Philadelphia, Pennsylvania. (Miazad Decl. I, Ex. 7 (“Desphy Dep.”) at 7:3–6, 19–23). She previously had worked for the Census Bureau during the 2000 Decennial Census, and applied again in December 2009. (Id. at 24:5–7, 25:15–17). Desphy initially earned a score of 75 on her qualifying exam. (Willis Decl. I, ¶ 68). After submitting her application and completing the exam, Desphy received a 30–day Letter based on the results of her background check. (Desphy Dep. 77:21–25, 79:2–7). Desphy responded to the 30–day Letter within the requisite time frame. (Miazad Decl. I, Ex. 10). She subsequently received an additional letter from the Census Bureau requesting a further explanation and two character references. (Id., Ex. 11). In response, Desphy sent the Census Bureau a letter detailing the circumstances of her arrest along with several letters of recommendation and a copy of her fingerprints. (Id., Ex. 12). In addition, she retook the written exam, increasing her score to 83. (Willis Decl. I, ¶ 68). Nevertheless, in March 2010, the Census Bureau rejected Desphy's employment application “based on the nature of the facts disclosed” in her correspondence. (Miazad Decl. I, Ex. 13).
LCO staff members reviewed a total of 123 selection records encompassing Desphy's geocode after the date of her application. (Willis Decl. I, ¶ 69). Fifty-one of those selection records were limited to applicants with foreign language skills, which Desphy did not possess. (Id. ¶ 70). The remaining selection records included candidates with scores ranging from 70 to 100. (See id., Ex. D). Desphy's score of 83 thus would have placed her on at least one of the selection records produced in her geographic region. According to the Census Bureau, however, no candidate with a score lower than 88 was hired from the selection records that did not require foreign language fluency. (Id. ¶ 75).
II. Title VII Disparate Impact FrameworkThe Plaintiffs allege that the Census Bureau used racially discriminatory hiring procedures in violation of Title VII. As amended, Title VII prohibits employers from discriminating on the basis of race, color, religion, sex, or national origin. In addition to banning intentional employment discrimination, Title VII prohibits some employment practices “that are not intended to discriminate but in fact have a disproportionately adverse effect on minorities.” Ricci v. DeStefano, 557 U.S. 557, 577, 129 S.Ct. 2658, 174 L.Ed.2d 490 (2009).
To establish a prima facie case under a “disparate impact” theory of liability, Title VII plaintiffs first must show that their employer uses “a particular employment practice that causes a disparate impact on the basis of race, color, religion, sex, or national origin.” 42 U.S.C. § 2000e–2(k)(1)(A)(i). The Second Circuit has described this provision as requiring plaintiffs to “(1) identify a specific employment practice or policy; (2) demonstrate that a disparity exists; and (3) establish a causal relationship between the two.” Chin v. Port Auth. of N.Y. & N.J., 685 F.3d 135, 151 (2d Cir.2012), cert. denied, ––– U.S. ––––, 133 S.Ct. 1724, 185 L.Ed.2d 785 (2013) (internal quotation marks and citations omitted).
Once plaintiffs have established a prima facie Title VII violation on a disparate impact theory, an employer may defend against liability by demonstrating that the challenged employment practice is “job related for the position in question and consistent with business necessity.” 42 U.S.C. § 2000e–2(k)(1)(A)(i). Even if the employer meets that burden, plaintiffs may succeed if they can show that the employer has refused to adopt an available alternative employment practice that would reduce the level of disparate impact while still serving the employer's legitimate needs. Id. §§ 2000e–2(k)(1)(A)(ii), (k)(1)(C).
If the trier of fact finds that the plaintiffs have established a Title VII violation on a disparate impact theory, each plaintiff seeking individual relief such as backpay “need only show that he or she suffered an adverse employment decision ‘and therefore was a potential victim of the proved ... discrimination.’ ” Robinson v. Metro–North Commuter R.R. Co., 267 F.3d 147, 159 (2d Cir.2001) (quoting Int'l Bhd. of Teamsters v. United States, 431 U.S. 324, 362, 97 S.Ct. 1843, 52 L.Ed.2d 396 (1977) ). The employer then may rebut the plaintiffs' presumptive entitlement to backpay “by proving that the [plaintiff] would not have been hired even absent discrimination.” Cohen v. West Haven Bd. of Police Comm'rs, 638 F.2d 496, 502 (2d Cir.1980). If the employer cannot make such a showing, or if the issue remains uncertain because the nature of the employer's hiring practices makes it impossible to determine a “but for” outcome, the plaintiff is entitled to individualized relief, including backpay. Chin, 685 F.3d at 151–52 ; Cohen, 638 F.2d at 502.
In this case, the Plaintiffs claim that the 30–day Letter and the Adjudication Criteria had a disparate impact on African–American and Latino applicants, were not job-related or justified by business necessity, and were not the least discriminatory hiring protocols that the Census Bureau could employ to meet its business needs. The Plaintiffs contend that every putative class member thus is entitled to backpay. The Census Bureau disputes each of the Plaintiffs' contentions.
III. Motion to Dismiss for Lack of Subject–Matter JurisdictionA. Standard of Review
The Census Bureau brings its motion to dismiss pursuant to Rule 12(b)(1) of the Federal Rules of Civil Procedure. Under Rule 12(b)(1), a court must dismiss a complaint if the court lacks subject-matter jurisdiction over the claims asserted. In resolving the issue of subject-matter jurisdiction, a court is not limited to the face of the complaint and may consider evidence outside the pleadings. Phifer v. City of N.Y., 289 F.3d 49, 55 (2d Cir.2002). The plaintiff has the burden of demonstrating the court's subject-matter jurisdiction by a preponderance of the evidence. Id. (citing Makarova v. United States, 201 F.3d 110, 113 (2d Cir.2000) ); Shipping Fin. Servs. Corp. v. Drakos, 140 F.3d 129, 131 (2d Cir.1998) (“[J]urisdiction must be shown affirmatively, and that showing is not made by drawing from the pleadings inferences favorable to the party asserting it.”).
Although challenges to plaintiffs' standing sometimes are brought under Rule 12(b)(6), Rule 12(b)(1) is the more appropriate procedural mechanism for raising such a challenge. See Alliance for Envtl. Renewal, Inc. v. Pyramid Crossgates Co., 436 F.3d 82, 88 n. 6 (2d Cir.2006). The distinction between the two rules is important because a dismissal under Rule 12(b)(6) is an adjudication on the merits with preclusive effect, while a dismissal under Rule 12(b)(1) is not. Id. (citing Federated Dep't Stores, Inc. v. Moitie, 452 U.S. 394, 399 n. 3, 101 S.Ct. 2424, 69 L.Ed.2d 103 (1981) ; Nowak v. Ironworkers Local 6 Pension Fund, 81 F.3d 1182, 1188 (2d Cir.1996) ).
B. Standing
Under Article III of the Constitution, the subject-matter jurisdiction of federal courts is limited to resolving “cases” and “controversies.” U.S. Const. art. III, § 2. Standing “is an essential and unchanging part of the case-or-controversy requirement of Article III.” Lujan v. Defenders of Wildlife, 504 U.S. 555, 560, 112 S.Ct. 2130, 119 L.Ed.2d 351 (1992). As the Supreme Court explained in Lujan:
the irreducible constitutional minimum of standing contains three elements. First, the plaintiff must have suffered an injury in fact—an invasion of a legally protected interest which is (a) concrete and particularized and (b) actual or imminent, not conjectural or hypothetical. Second, there must be a causal connection between the injury and the conduct complained of—the injury has to be fairly traceable to the challenged action of the defendant, and not the result of the independent action of some third party not before the court. Third, it must be likely, as opposed to merely speculative, that the injury will be redressed by a favorable decision.
Id. at 560–61, 112 S.Ct. 2130 (citations, internal quotation marks, ellipses, and brackets omitted). The plaintiff carries the burden to establish these elements of constitutional standing, which are to be “evaluated at the time the complaint is filed.” Access 4 All, Inc. v. Trump Int'l Hotel & Tower Condo., 458 F.Supp.2d 160, 167 (S.D.N.Y.2006) ; see also Disabled in Action of Metro. N.Y. v. Trump Int'l Hotel & Tower, No. 01 Civ. 5518(MBM), 2003 WL 1751785, at *7 (S.D.N.Y. Apr. 2, 2003) (“Events occurring after the lawsuit has been filed may be relevant to whether the claim has become moot but are not relevant to whether a plaintiff has standing in the first instance.”).
Article III's standing requirements apply equally to class actions. As the Supreme Court has held, “if none of the named plaintiffs purporting to represent a class establishes the requisite of a case or controversy with the defendant[ ], none may seek relief on behalf of himself or any other member of the class.” O'Shea v. Littleton, 414 U.S. 488, 494, 94 S.Ct. 669, 38 L.Ed.2d 674 (1974). To demonstrate standing, each of the named plaintiffs “must allege and show that they personally have been injured, not that injury has been suffered by other, unidentified members of the class to which they belong and which they purport to represent.” Warth v. Seldin, 422 U.S. 490, 502, 95 S.Ct. 2197, 45 L.Ed.2d 343 (1975) ; see also Lewis v. Casey, 518 U.S. 343, 357, 116 S.Ct. 2174, 135 L.Ed.2d 606 (1996) (Article III standing requirements are “no less true with respect to class actions than with respect to other suits”).
The Census Bureau contends that the named plaintiffs in this action lack Article III standing because they cannot show that (1) they have personally suffered any injury-in-fact as a result of the Census Bureau's use of the 30–day Letter or its application of the Adjudication Criteria, or (2) the relief sought would redress their alleged injuries.
1. Injury–In–FactIn determining whether a plaintiff has suffered an injury-in-fact, courts must “assess whether the injury affect[s] the plaintiff in a personal and individual way, to confirm that the plaintiff has a personal stake in the controversy and avoid having the federal courts serve as merely publicly funded forums for the ventilation of public grievances or the refinement of jurisprudential understanding.” Baur v. Veneman, 352 F.3d 625, 632 (2d Cir.2003) (citation and internal quotation marks omitted); see also Sullivan v. Syracuse Hous. Auth., 962 F.2d 1101, 1106 (2d Cir.1992) ( “[C]ourts generally should refrain from adjudicating abstract questions of wide public significance which amount to generalized grievances....”) (internal quotation marks omitted).
Neither the Supreme Court nor the Second Circuit has expounded further on the meaning of the injury-in-fact requirement in Title VII cases. Several other circuit courts, however, have suggested that to establish an injury-in-fact, a Title VII plaintiff must at least demonstrate that he met the minimum qualifications for the position ultimately denied. See, e.g., Bates v. United Parcel Serv., Inc., 465 F.3d 1069, 1078 (9th Cir.2006) (determining whether plaintiff was injured requires court to examine whether he was “qualified” for the position “in the sense that, aside from the [employment] standard he is challenging and all prerequisites connected to that standard, he meets the basic job requirements for the desired position”); Coe v. Yellow Freight Sys., Inc., 646 F.2d 444, 451 (10th Cir.1981) (to establish standing, Title VII plaintiffs must show that “they were qualified for the positions that they sought”); Jones v. Mukasey, 565 F.Supp.2d 68, 81 (D.D.C.2008) (same). As one court has put it, “we assume that an unqualified plaintiff was not hired or promoted for the obvious reason—that he was unqualified. Such a plaintiff would have no standing to sue under Title VII, for he could not claim that he was injured, much less affected,” by the allegedly discriminatory employment practice. Melendez v. Ill. Bell Tel. Co., 79 F.3d 661, 668 (7th Cir.1996). Although there appears to be little, if any, authority in this Circuit addressing this “minimum qualification” requirement, the Plaintiffs do not dispute its validity. (See ECF No. 246 (Pls.' Mem. of Law in Opp. to Mot. to Dismiss (“Pl.'s Dismissal Mem.”) at 18 (“Plaintiffs need only show that they met the basic job requirements for the job.”)) (emphasis in original)).
The decisions applying a “minimum qualification” requirement are well-reasoned. Applicants who fail to meet the basic eligibility requirements for employment cannot demonstrate that they suffered a “particularized” personal injury by being denied such employment. Thus, an individual who did not satisfy the eligibility requirements for employment with the Census Bureau would lack constitutional standing to serve as a named plaintiff in a lawsuit challenging the Census Bureau's allegedly discriminatory hiring practices.
That the named plaintiffs must establish their eligibility for hire, however, does not mean, as the Census Bureau contends, that they must show that they ultimately would have been hired. Rather, the Supreme Court has made clear that “[w]hen the government erects a barrier that makes it more difficult for members of one group to obtain a benefit than it is for members of another group, ... [t]he ‘injury in fact’ ... is the denial of equal treatment resulting from the imposition of the barrier, not the ultimate inability to obtain the benefit.” Ne. Fla. Chapter of Associated Gen. Contractors of Am. v. Jacksonville, 508 U.S. 656, 666, 113 S.Ct. 2297, 124 L.Ed.2d 586 (1993) ; see also Connecticut v. Teal, 457 U.S. 440, 450, 102 S.Ct. 2525, 73 L.Ed.2d 130 (1982) (“In considering claims of disparate impact ... this Court has consistently focused on employment and promotion requirements that create a discriminatory bar to opportunities. ”) (emphasis in original); Regents of Univ. of Cal. v. Bakke, 438 U.S. 265, 280 n. 14, 98 S.Ct. 2733, 57 L.Ed.2d 750 (1978) (denial of opportunity to compete for admission to university constitutes sufficient injury-in-fact, even if plaintiff ultimately may not have been admitted). Under that standard, a plaintiff seeking to establish standing “need not allege that he would have obtained the benefit but for the barrier,” but simply that the barrier “prevented [him] from competing on an equal footing.” Jacksonville, 508 U.S. at 666–67, 113 S.Ct. 2297. Although that standard was announced in the equal protection context, it should, if anything, apply with even greater force in Title VII cases, since federal courts generally are less reluctant to overcome jurisdictional hurdles when faced with statutory, rather than constitutional, merits questions. Cf. Reno v. Catholic Soc. Servs., Inc., 509 U.S. 43, 82 n. 4, 113 S.Ct. 2485, 125 L.Ed.2d 38 (1993) (Stevens, J., dissenting) (“Jacksonville is, of course, an equal protection case, while respondents in this case are seeking a statutory benefit. If this distinction has any relevance ..., then it should mitigate in favor of finding ripeness here; I assume we should be more reluctant to overcome jurisdictional hurdles to decide constitutional issues than to effectuate statutory programs.”).
The Census Bureau cites a number of cases from other circuits that allegedly support its standing argument. None of these cases, however, goes beyond requiring plaintiffs to show that they were eligible to be considered for the position. For example, in Melendez v. Illinois Bell Telephone Company, the Seventh Circuit noted that a Title VII plaintiff could satisfy the “qualification requirement” by showing that he met “all of the objective requirements” for the position he sought, regardless of whether he would have been considered for the position. 79 F.3d at 668, 669 n. 8. In Rich v. Martin Marietta Corp., the Tenth Circuit “emphasize[d] that each plaintiff need not prove that he was the most qualified person,” so long as he could show that he was at least eligible to be considered for the position. 522 F.2d 333, 347–48 (10th Cir.1975). And in Bates v. United Parcel Service, Inc., the Ninth Circuit held that the plaintiff had standing to challenge an allegedly discriminatory hiring test because his “qualifications [were] sufficient to allow him to proceed to the next step” of the hiring process. 465 F.3d 1069, 1078 (9th Cir.2006) (internal quotation marks and brackets omitted). Neither these cases, nor any of the other cases cited by the Census Bureau, require a plaintiff to show that he would have made the employer's final cut.
In sum, I find that the named plaintiffs must establish that they were eligible to be considered for the positions they sought, but need not further demonstrate that they would have been hired after being evaluated in comparison to other applicants. Thus, the question here is whether the Census Bureau's allegedly discriminatory hiring practices placed any of the named plaintiffs on an unequal footing in terms of their ability to compete for employment.
Applying this standard, it is clear that five of the eight named plaintiffs—Houser, Daniels, Rickett–Samuels, Scott, and Desphy—have standing to sue under Title VII. It is undisputed that each of these class representatives possessed the bare minimum qualifications for employment with the Census Bureau: they all were older than eighteen, possessed Social Security numbers, and had scored above 70 on the qualifying exam. In addition, these plaintiffs all obtained scores within the range of scores that appeared on selection records in their geographic region. Thus, had these five applicants not been subjected to the allegedly discriminatory criminal background check process, they each would have appeared on at least one of the selection records for their geographic region. Simply by appearing on a selection record, these individuals would have had an opportunity (however slim) to compete for employment. Whether the Census Bureau likely would have reached their name on the rank-ordered selection record before it completed the hiring process makes no difference. These five plaintiffs have established that they were eligible to be considered for employment but were denied the opportunity to compete with other applicants. That showing is sufficient to confer standing under Title VII.The three remaining named plaintiffs—Gonzalez, Riesco, and Kargbo—each obtained scores that fell below the range of scores on selection records in their geographic regions. Thus, even if they had passed the criminal background check without delay, they would not have appeared on any of the selection records and would not have been eligible to compete for employment. And although these applicants could have retaken the exam in an effort to improve their scores, they chose not to do so. Given their existing test scores, they could not have been considered for hire, regardless of the results of their background check. They consequently cannot demonstrate an injury-in-fact sufficient to prove Article III standing, and must be dismissed from this action.
The Plaintiffs argue that applicants who were eliminated from the hiring pool based on the results of their criminal background check would have “had no reason or opportunity to retake the test.” (Pl.'s Dismissal Mem. at 22). That logic does not apply to Gonzalez, Riesco or Kargbo. These three applicants each responded to the 30–day Letter within the allotted time frame, which, as the letter indicated, was sufficient to keep their application “active.” (Miazad Decl. I, Ex. 70 at USA 43538). Contrary to the Plaintiffs' contention, nothing would have prevented them from retaking the exam while they awaited the results of their background checks.
2. Redressability
Article III also requires plaintiffs to show that a favorable decision could redress their injuries. The Second Circuit has described redressability as the “non-speculative likelihood that the injury can be remedied by the requested relief.” W.R. Huff Asset Mgmt. Co., LLC v. Deloitte & Touche LLP, 549 F.3d 100, 106–07 (2d Cir.2008).
In this case, the Plaintiffs request several forms of relief, including (a) a declaratory judgment that the Census Bureau's criminal background check process violates Title VII; (b) preliminary and permanent injunctive relief enjoining the Census Bureau from “engaging in each of the unlawful policies, practices, customs and usages” set forth in the Second Amended Complaint; (c) an order directing the Census Bureau to use employment policies, practices and programs that eliminate the effects of its allegedly unlawful employment practices; and (d) backpay in the form of lost wages and lost benefits. (See SAC ¶ 135–40). To satisfy the redressability prong, the putative class representatives each must show that at least one of these forms of relief would redress their alleged injuries.
The Plaintiffs also seek to recover costs, attorneys' fees, and interest. This element of the Plaintiffs' prayer for relief, standing alone, plainly would not satisfy the redressability requirement. As the Supreme Court has made clear, “a plaintiff cannot achieve standing to litigate a substantive issue by bringing suit for the cost of bringing suit.” Steel Co. v. Citizens for a Better Env't, 523 U.S. 83, 107, 118 S.Ct. 1003, 140 L.Ed.2d 210 (1998).
In ruling on the Census Bureau's first motion to dismiss this case, I determined that the declaratory and injunctive relief requested could redress the named plaintiffs' alleged injuries. Johnson v. Bryson, 851 F.Supp.2d 688, 700 (S.D.N.Y.2012). The Census Bureau now urges the Court to revisit that decision, arguing that the named plaintiffs' injuries could not be redressed by the requested relief because each named plaintiff was “precluded from selection for reasons entirely independent of the challenged policies and procedures.” (ECF No. 226 (Def.'s Mem. of Law in Supp. of Mot. to Dismiss) at 25). In repeating this refrain, the Census Bureau essentially has just repackaged its arguments related to the named plaintiffs' injuries-in-fact; it has said nothing about whether the requested injunction would redress those injuries. Assuming that the named plaintiffs did suffer cognizable injuries as a result of the Census Bureau's allegedly discriminatory hiring practices, they certainly would be entitled to—and their injuries could be redressed by—the requested injunctive and declaratory relief. The Census Bureau's redressability argument concerning injunctive relief therefore fails with respect to the five named plaintiffs who satisfy the injury-in-fact requirement.
The named plaintiffs' alleged injuries also may be redressed through an award of backpay. As both parties agree, members of a Title VII class in this Circuit are presumptively entitled to backpay. See Cohen, 638 F.2d at 502. To establish this presumptive entitlement to backpay, a Title VII plaintiff need only establish that he applied for the job and was not hired. Ass'n Against Discrimination in Emp't, Inc. v. City of Bridgeport, 647 F.2d 256, 289 (2d Cir.1981). The named plaintiffs plainly have made such a showing.
As the Census Bureau correctly notes, an employer may rebut a claimant's prima facie claim of entitlement to backpay by “proving that the class member would not have been hired even absent discrimination.” Id. This might arise, “for example, because no vacancies existed or because the claimant failed to meet nondiscriminatory prerequisites for employment.” Id. In an attempt to make the requisite showing, the Census Bureau once again reiterates its view that the named plaintiffs ultimately would not have been hired, and therefore are not entitled to backpay. Again, this argument speaks to whether the named plaintiffs have suffered an injury-in-fact sufficient to confer Article III standing, not to whether that injury is redressable by backpay. At the risk of sounding like a broken record, that the named plaintiffs ultimately may not have been hired is of no consequence insofar as standing is concerned, so long as they can show, as five have here, that they met the prerequisites for employment with the Census Bureau and could have been considered based on their test scores.
Because Houser, Daniels, Rickett–Samuels, Scott, and Desphy each can demonstrate that they suffered an injury-in-fact that can be redressed by a favorable disposition, they have standing to proceed as class representatives. Accordingly, the Census Bureau's motion to dismiss this action in its entirety must be denied. As noted above, however, Gonzalez, Riesco and Kargbo lack standing to pursue this action. Accordingly, their claims must be dismissed for lack of subject matter jurisdiction.
IV. Class CertificationSince Houser, Daniels, Rickett–Samuels, Scott, and Desphy have individual standing to bring this suit, the Court must consider whether the action should be certified as a class action under Rule 23 of the Federal Rules of Civil Procedure. The Rule 23 class action is “an exception to the usual rule that litigation is conducted by and on behalf of the individual named parties only.” Califano v. Yamasaki, 442 U.S. 682, 700–01, 99 S.Ct. 2545, 61 L.Ed.2d 176 (1979). Rule 23 permits named plaintiffs to proceed on behalf of a class of absent class members in three distinct situations: (a) when the rights of either the potential class members or the party opposing the class would be harmed by piecemeal litigation, Fed.R.Civ.P. 23(b)(1) ; (b) where the class seeks injunctive or declaratory relief against a party who has “acted or refused to act on grounds that apply generally to the class,” Fed.R.Civ.P. 23(b)(2) ; and (c) where (i) questions common to the class predominate over individual questions and (ii) class adjudication would be superior to other available methods of resolving the controversy, Fed.R.Civ.P. 23(b)(3). This last category encompasses cases in which class adjudication would “achieve economies of time, effort, and expense, and promote uniformity of decision as to persons similarly situated, without sacrificing procedural fairness or bringing about other undesirable results.” Fed.R.Civ.P. 23(b)(3) advisory committee notes (1966).
The Plaintiffs urge the Court to take a “hybrid approach” to the certification process by applying different provisions of Rule 23 to different stages of the litigation. For purposes of determining the Census Bureau's liability and affording classwide injunctive relief, the Plaintiffs request that the Court certify a class under Rule 23(b)(2) consisting of “all African American and Latino applicants who applied for temporary employment during the 2010 decennial and were harmed by ... [the Census Bureau's] use of the 30–day [L]etter as a screening device[,] ... [the Census Bureau's] use of adjudication criteria to screen applicants,” or both. (Pls.' Class Cert. Mem. at 26). For the monetary damages phase, the Plaintiffs request certification of several subclasses under Rule 23(b)(3), which would include: (a) individuals barred from employment “based solely on the procedural requirements imposed by the 30–day [L]etter;” (b) individuals barred from employment “based solely on the delay in adjudicating” their applications; and (c) individuals barred from employment “based on exclusions that are not job related.” (Id. ). The Plaintiffs contend that this hybrid approach is appropriate because once liability is determined, calculation of damages will be largely formulaic. (See Pls.' Class Cert. Mem. at 27, 38).
As an alternative to this hybrid approach, the Plaintiffs propose that the Court exercise its discretion under Rule 23(c)(4) to isolate the liability and injunctive relief questions, certify a single class under Rule 23(b)(2) to address those issues, and leave damages calculations for individualized hearings. (Id. at 44–45).
Because both the proposed liability class and the proposed damages subclasses must independently meet all the requirements of Rule 23, I will consider them separately.
A. Proposed Liability Class
1. Rule 23(a)To certify the proposed liability class, the Court must first be satisfied that the Plaintiffs have met the requirements of Rule 23(a) by establishing that “(a) the class is so numerous that joinder of all members is impracticable, (b) there are questions of law or fact common to the class, (c) the claims or defenses of the representative parties are typical of the claims or defenses of the class, and (d) the representative parties will fairly and adequately protect the interests of the class.” Fed.R.Civ.P. 23(a). As set forth below, the Plaintiffs have met their burden in this regard.
a. Numerosity
Rule 23(a)(1) requires the Plaintiffs to demonstrate that the proposed class is so numerous that joinder of all class members would be “impracticable.” “Generally, a class composed of more than forty members satisfies the numerosity requirement.” See Noble v. 93 Univ. Place Corp., 224 F.R.D. 330, 338 (S.D.N.Y.2004). Although plaintiffs bear the burden of proving numerosity, they “need not present a precise calculation of the number of class members,” and a court may rely on reasonable inferences from available facts in drawing its conclusion. Id. (citing Robidoux v. Celani, 987 F.2d 931, 935 (2d Cir.1993) ; McNeill v. N.Y.C. Hous. Auth., 719 F.Supp. 233, 252 (S.D.N.Y.1989) ).
Although they have not ascertained the exact size of the putative class, the Plaintiffs indicate that there are thousands of minority class members who were barred from employment because of the Census Bureau's allegedly discriminatory hiring practices. Indeed, the Plaintiffs' experts estimate that more than 250,000 African–American and 200,000 Latino applicants received a 30–day Letter. (ECF No. 166, Ex. C (Decl. of Mark Bendick, Jr., dated June 28, 2013), ¶ 32). They further estimate that less than one percent of all applicants who received 30–day Letters ultimately were hired. (Id., Ex. A (Decl. of Maria Kozhevnikova, dated June 28, 2013) at Table 1). It therefore is obvious that the proposed liability class would contain far too many individuals to manage under the typical joinder rules. Indeed, the Census Bureau does not appear to dispute that conclusion.
b. Commonality
Commonality requires plaintiffs to show that “there are questions of law or fact common to the class.” Fed.R.Civ.P. 23(a)(2). The “Rule does not require that all questions of law or fact raised in the litigation be common[;] ... indeed, even a single question of law or fact common to the members of the class will satisfy the commonality requirement.” Wal–Mart Stores, Inc. v. Dukes, ––– U.S. ––––, 131 S.Ct. 2541, 2562, 180 L.Ed.2d 374 (2011) (internal quotation marks, brackets, and citations omitted).
To satisfy the commonality requirement, the lead plaintiffs in a putative class action must demonstrate that all of the class members “have suffered the same injury.” Dukes, 131 S.Ct. at 2551 (quoting Gen. Tel. Co. of Sw. v. Falcon, 457 U.S. 147, 157, 102 S.Ct. 2364, 72 L.Ed.2d 740 (1982) ). As the Supreme Court recently stated in Dukes, “[t]his does not mean merely that [the plaintiffs must] have all suffered a violation of the same provision of law.... Their claims must depend upon a common contention ... of such a nature that it is capable of classwide resolution—which means that determination of its truth or falsity will resolve an issue that is central to the validity of each one of the claims in one stroke.” Id. In other words, a properly certified class action must not only raise common questions, but must have the “capacity ... to generate common answers.” Id. (quoting Nagareda, Class Certification in the Age of Aggregate Proof, 84 N.Y.U. L.Rev. 97, 132 (2009) ).
The plaintiffs in Dukes sought certification of a nationwide class of current and former female Wal–Mart employees alleging gender discrimination with respect to pay and promotions. The Court ultimately determined that the plaintiffs had not satisfied the commonality requirement because Wal–Mart did not use a uniform “testing procedure or other companywide evaluation” in making pay and promotion decisions, but rather left those decisions to regional managers' discretion. Id. at 2553. This discretion, the Court determined, was “just the opposite of a uniform employment practice that would provide commonality.” Id. at 2554.
Dukes unquestionably “upped the ante on ‘commonality’ ” for plaintiffs challenging employers' allegedly discriminatory subjective decisionmaking. Haddock v. Nationwide Fin. Servs., Inc., 293 F.R.D. 272, 279 (D.Conn.2013). What distinguishes this case from Dukes, however, is the fact that the Census Bureau, unlike Wal–Mart, employed uniform, non-discretionary hiring instruments—the 30–day Letter and the Adjudication Criteria—to exclude applicants from the hiring pool. Addressing this very situation, the Dukes court made clear that when an employer uses the same “testing procedure to evaluate [all] applicants for employment,” “a class action on behalf of every applicant or employee who might have been prejudiced” by the allegedly biased procedure “clearly would satisfy the commonality” requirement. Dukes, 131 S.Ct. at 2553 (quoting Falcon, 457 U.S. at 159 n. 15, 102 S.Ct. 2364 ).
The fact that the Census Bureau conducted its hiring efforts out of a series of regional offices, admittedly, adds a slight wrinkle to the commonality analysis. As the Census Bureau emphasizes, applicant characteristics and the rate at which they received 30–day Letters varied from one LCO to the next. The commonality requirement, however, does not require that all class members' circumstances be identical, so long as their claims can be proven or disproven on a classwide basis. Here, notwithstanding the potential variance among LCOs, commonality will exist at the very least if it is possible to prove either that (i) the challenged hiring practices had a disparate impact on minority applicants across all LCOs, or (ii) the challenged hiring practices had no disparate impact in any LCO. In addition, commonality will exist if the Census Bureau has defenses that apply to all applicants in all LCOs, or if the Plaintiffs can rebut those defenses on a similar classwide basis.
Contrary to the Census Bureau's contentions, both sides here appear to agree that the central questions in this case each have a common, classwide answer; the only point on which the parties disagree is the answers themselves. For example, to answer the disparate impact question, the Plaintiffs have submitted an expert report from labor economist Dr. Marc Bendick, Jr., which suggests that the Census Bureau's hiring procedures adversely affected African–American and Latino applicants nationwide. (ECF No. 166, Ex. C). The Census Bureau disputes the validity of Dr. Bendick's analyses, and in response has proffered the report of their own statistical expert, Dr. Bernard Siskin. Dr. Siskin's report highlights the statistical differences among LCOs and criticizes Dr. Bendick for his failure to control for those differences in his study. Critically, however, upon examining the results by individual locality, Dr. Siskin concludes that the challenged practices had no statistically significant disparate impact in any LCO. (See ECF No. 201, ¶¶ 50–51). Thus, both experts agree that there is one nationwide answer to the question of disparate impact.
This case also presents a common question as to whether the Census Bureau's allegedly discriminatory criminal background procedures were justified by legitimate business needs. To answer this question, the Census Bureau has submitted the expert report of Dr. James Outtz, an industrial-organizational psychologist, who defends the agency's hiring procedures on the grounds that they were necessary to gain public trust and minimize public safety risks. (ECF No. 202). The Plaintiffs, in turn, have submitted a report from their own industrial-organizational expert, Dr. Kathleen Lundquist, to dispute those defenses. (ECF No. 166, Ex. B). Once again, a single nationwide answer to this question seems likely.
In the end, though both sides have spent a great deal of energy arguing the validity of their experts' analyses, the Court need not resolve such “battles of the experts” at this juncture. The question at this preliminary stage of the litigation is not whether the challenged hiring procedures actually had a disparate impact or were justified by business necessity, but merely whether those questions can be resolved on a classwide basis. See In re Magnetic Audiotape Antitrust Litig., No. 99 Civ. 1580 (LMM), 2001 WL 619305, at *4 (S.D.N.Y. June 1, 2001) (“[O]n a motion for class certification, the Court only evaluates whether the method by which plaintiffs propose to prove class-wide impact could prove such impact, not whether plaintiffs in fact can prove class-wide impact.”); see also In re Aftermarket Auto. Lighting Prods. Antitrust Litig., 276 F.R.D. 364, 373–74 (C.D.Cal.2011) (post-Dukes case concluding that courts “need not cho[o]se between experts” at the class certification stage because inquiry is limited to determining whether merits issues can be resolved through “generalized proof common to the class”) (internal citations omitted). Although the experts obviously reach different conclusions regarding the merits in this case, the fact that both sides' experts are able to provide classwide answers to the liability question suffices to satisfy the commonality requirement.
In reaching this conclusion, I am mindful that class certification questions often may overlap with questions related to the merits of plaintiffs' underlying claims, and that the Court's duty to perform a “rigorous analysis” of the issues pertaining to certification is not lessened by this overlap. Dukes, 131 S.Ct. at 2551–52 (quoting Falcon, 457 U.S. at 161, 102 S.Ct. 2364 ). Nonetheless, a district court has the obligation to “assure that a class certification motion does not become a pretext for a partial trial of the merits,” and thus “should not assess any aspect of the merits unrelated to a Rule 23 requirement.” In re Initial Pub. Offerings Sec. Litig., 471 F.3d 24, 41 (2d Cir.2006). Here, the evidence strongly suggests that the Plaintiffs' liability claims are capable of classwide resolution. It therefore would be inappropriate for me to opine as to the validity of the experts' merits conclusions at this preliminary stage.
After filing its opposition to the Plaintiffs' motion for class certification, the Census Bureau filed a separate motion in limine to exclude the testimony of Dr. Lundquist for purposes of class certification. (ECF No. 207). In that motion, the Census Bureau contends that Dr. Lundquist used a flawed methodology to reach her conclusions regarding the “job-relatedness” of the Census Bureau's hiring procedures, and failed to conduct any analysis to support her assertions regarding adverse impact. Because it is unnecessary to consider the substance of Dr. Lundquist's conclusions in order to resolve the class certification motion, the concerns raised in the Census Bureau's motion in limine are moot.
Of course, as this litigation unfolds, further expert testimony may show that the differences among the various hiring regions were so significant as to preclude classwide determination of the liability question. Should that occur, the Court will have to revisit the commonality issue to determine whether it is possible to create subclasses for different LCOs and to appoint class representatives for each subclass. See Fed.R.Civ.P. 23(c)(1)(c) (court may alter or amend certification order at any time before final judgment); 23(c)(5) (court may divide certified class into subclasses, each to be treated as a separate class under Rule 23 ). The mere possibility that this might prove necessary, however, is not an obstacle to certifying a single nationwide class at this stage. See In re Omnicom Grp., Inc. Secs. Litig., No. 02 Civ. 4483(RCC), 2007 WL 1280640, at *8 (S.D.N.Y. Apr. 30, 2007) (if common liability issues later “metastasize into markedly different questions of law and fact, the Court has the option” to amend the certification order or create subclasses “at a later time”); see generally Sumitomo Copper Litig. v. Credit Lyonnais Rouse, Ltd., 262 F.3d 134, 139 (2d Cir.2001) (“[T]he district court is often in the best position to assess the propriety of the class and has the ability ... to alter or modify the class, create subclasses, and decertify the class whenever warranted.”).
In sum, I find that the proposed liability class satisfies the commonality requirement of Rule 23(a). Each of the proposed class members' claims raises at least one common question: whether the 30–day Letter, the Adjudication Criteria, or both, were racially biased. That question has the capacity to “generate common answers that will drive the resolution of the litigation.” Dukes, 131 S.Ct. at 2551. If it ultimately is determined, as the Census Bureau contends, that neither the 30–day Letter nor the Adjudication Criteria had a disparate impact on minority applicants, then the claims of every class member, whether for damages, injunctive relief, or both, will fail and the Court will enter judgment in favor of the Census Bureau—“a course [the Census Bureau] should welcome, as all class members ... [then] would be bound by the judgment.” Butler v. Sears, Roebuck and Co., 727 F.3d 796, 799 (7th Cir.2013), cert. denied, ––– U.S. ––––, 134 S.Ct. 1277, 188 L.Ed.2d 298 (2014). If, however, disparate impact is found, then the class members' claims will go forward.
c. Typicality
Rule 23(a) further requires that “the claims or defenses of the representative parties [be] typical of the claims or defenses of the class.” This requirement is satisfied when the lead plaintiffs' claims arise from the same series of events and find support in the same legal theories as the claims of all of the remaining class members. In re Drexel Burnham Lambert Grp., Inc., 960 F.2d 285, 291 (2d Cir.1992) ; In re Prudential Secs. Inc. Ltd. P'ships Litig., 163 F.R.D. 200, 207–08 (S.D.N.Y.1995). “Typicality ... does not require that the situations of the named representatives and the class members be identical,” In re Oxford Health Plans, Inc. Secs. Litig., 199 F.R.D. 119, 123 (S.D.N.Y.2001), but merely that the class representatives have the “incentive to prove all the elements of the cause of action which would be presented by the individual members of the class were they initiating individualized actions,” In re NASDAQ Market–Makers Antitrust Litig., 169 F.R.D. 493, 510 (S.D.N.Y.1996). The “crux” of the typicality requirement is simply to “ensure that ‘maintenance of a class action is economical and that the named plaintiff's claim and the class claims are so interrelated that the interests of the class members will be fairly and adequately protected in their absence.’ ” Marisol A. v. Giuliani, 126 F.3d 372, 376 (2d Cir.1997) (quoting Falcon, 457 U.S. at 157 n. 13, 102 S.Ct. 2364 ) (brackets omitted).
The Plaintiffs claim that the Census Bureau's hiring practices had a disparate impact on both African–American and Latino applicants, and seek to certify a single liability class that includes applicants belonging to both minority groups. Although both groups assert the same disparate impact theory, they will rely on separate sets of statistical data to support their claims. Accordingly, an African–American plaintiff's claims will not be typical of a Latino plaintiff's claims, and vice versa. Thus, to satisfy the typicality requirement for the proposed class as a whole, the named plaintiffs must include both African–American and Latino individuals. Here, however, the only two Latino class representatives named in the Second Amended Complaint—Gonzalez and Riesco—must be dismissed from the suit on standing grounds. Consequently, the absent Latino class members presently have no representative who can fairly and adequately represent their interests in this litigation. Since the remaining named plaintiffs' claims are not typical of the claims that would be raised by absent Latino class members, it would be inappropriate to certify a class that includes Latino applicants. The proposed class definition therefore must be amended—at least temporarily—to exclude Latino class members. If, however, the Plaintiffs are able to find a Latino representative who satisfies Article Ill's standing requirements, they may move to amend their complaint to assert that individual's claims. At that time, the Court would have discretion to amend the class definition to include Latino class members. See Fed.R.Civ.P. 23(c)(1)(C) (“An order that grants or denies class certification may be altered or amended before final judgment.”).
Turning to the remaining named plaintiffs, the typicality inquiry centers on whether their legal challenges are based on the same factual circumstances that would be raised by absent class members. The Census Bureau first contends that the Plaintiffs cannot satisfy the typicality requirement in this regard because the proposed liability class challenges both the 30–day Letter and the Adjudication Criteria, but no single named plaintiff was injured by both hiring instruments. In the Census Bureau's view, the claims of those class representatives who responded to the 30–day Letter and ultimately were excluded based on the Adjudication Criteria are not typical of those class members who never responded, or whose responses were untimely. In that same vein, the Census Bureau argues that those who did not respond to the 30–day Letter do not have claims that are typical of those who did but later were excluded based on the Adjudication Criteria.
The Census Bureau also argues that none of the named plaintiffs' claims are typical of those who were excluded based on the use of non-job-related Adjudication Criteria. To support this contention, the Census Bureau points to deposition testimony from the Plaintiffs' expert, Dr. Lundquist, who conceded that the offenses committed by the named plaintiffs who were excluded in the adjudication phase were “job-related” in a technical sense, but occurred too far in the past to have had an effect on the applicant's current qualifications. (ECF No. 203 (Def.'s Mem. of Law in Opp. to Mot. for Class Cert. (“Def.'s Class Cert. Mem.”)) at 34–35). This argument is unavailing. The Plaintiffs here challenge not only the Adjudication Criteria, but also the time frames within which they applied. The fact that some named plaintiffs may have, at one point in their history, committed “job-related offenses” does not weaken their claim that they were excluded for reasons that would have no effect on their job qualifications. Cf. Green v. Mo. Pac. R. Co., 523 F.2d 1290, 1298 (8th Cir.1975) (“We cannot conceive of any business necessity that would automatically place every individual convicted of any offense ... in the permanent ranks of the unemployed.... To deny job opportunities to these individuals because of some conduct which may be remote in time ... is an unnecessarily harsh and unjust burden.”). The claims raised by the named plaintiffs who were excluded at the adjudication phase thus are typical of the class claims related to the use of the Adjudication Criteria.
In this case, it would be impossible for a single plaintiff to have been legally “harmed” by both the 30–day Letter and the Adjudication Criteria. Those who either did not respond, or failed to respond in a timely manner, to the 30–day Letter were never subjected to the Adjudication Criteria, and those who were subjected to the Adjudication Criteria necessarily must have responded to, and therefore suffered no harm by virtue of the 30–day Letter. The fact that no single named representative has claims that are perfectly typical of all other class members does not, however, defeat typicality, so long as the named plaintiffs collectively satisfy the requirement. See, e.g., Ray M. v. N.Y. Bd. of Educ., 884 F.Supp. 696, 706 (E.D.N.Y.1995) (to satisfy the typicality prong, “the named plaintiffs [must], ‘in the aggregate, assert all of the claims asserted on behalf of class members' ”) (quoting Kenavan v. Empire Blue Cross and Blue Shield, No. 91 Civ. 2393 (KMW), 1993 WL 128012, at *4 (S.D.N.Y. Apr. 19, 1993) ).
Here, the factual circumstances raised by Houser, Daniels, Rickett–Samuels, Scott, and Desphy collectively satisfy the typicality requirement with respect to all African–American putative class members. Together, this group includes one individual who was harmed by the use of the 30–day Letter as a screening device (Houser), and four individuals who were harmed by the allegedly discriminatory Adjudication Criteria (Daniels, Rickett–Samuels, Scott, and Desphy). At least with respect to the issues of liability and injunctive relief, the class as a whole will prevail if these representatives can prove at trial that both the 30–day Letter and the Adjudication Criteria had a disparate impact on African–American applicants. If only portions of that claim succeed—for example, if the Court finds that the 30–day Letter had a disparate impact on African–American applicants but the Adjudication Criteria did not—it will have the discretion to create subclasses for purposes of fashioning injunctive relief. The important question at this stage simply is whether the interests of unnamed class members will be fairly and adequately represented by those who litigate their claims in court. “At bottom, the typicality requirement concerns the fairness of allowing an entire class's claim[s] to rise or fall with the fate of the named representative[s'] claims; thus, th [ose] representative[s'] claims must be typical of the class so as to prevent a false prophet from bearing the standard for an entire class of claims.” Rapcinsky v. Skinnygirl Cocktails, LLC, No. 11 Civ. 6546(JPO), 2013 WL 93636, at *5 (S.D.N.Y. Jan. 9, 2013). Viewing the claims of Houser, Daniels, Rickett–Samuels, Scott, and Desphy in the aggregate, there is no reason to believe that any absent African–American class members' claims will go unrepresented during the liability phase of the litigation.
The Census Bureau also contends that the individualized nature of the adjudication process precludes any of the named plaintiffs who were excluded during the adjudication phase from establishing typicality. This argument is equally untenable. Although the adjudication process may have involved “individualized” considerations in the sense that each applicant was evaluated on the basis of that individual's specific criminal history, the same set of objective Adjudication Criteria applied to all applicants nationwide, and those Adjudication Criteria required that LCO staff members automatically exclude applicants who had committed certain crimes without any further “individualized” consideration. At the liability phase, the Plaintiffs will need to prove only that those Adjudication Criteria disproportionately impacted African–American applicants and were not justified by business necessity. The individualized nature of each class member's criminal background will have no effect on these two inquiries.
Finally, the Census Bureau contends that the claims of one of the named plaintiffs, Chynell Scott, are not typical of those asserted by other putative class members because, according to the Census Bureau, Scott was not deemed ineligible because she failed the adjudication phase, but because she misrepresented facts in her initial application. In the Census Bureaus's view, the “unique defense” that it might have with respect to Scott's claim renders that claim atypical of other proposed class members' claims. (Def.'s Class Cert. Mem. at 35–36).
Although the Census Bureau's unique defense to Scott's claim may prove meritorious, it would not defeat typicality unless it truly “threaten [ed] to become the focus of the litigation.” Baffa v. Donaldson, Lufkin & Jenrette Secs. Corp., 222 F.3d 52, 59 (2d Cir.2000) (quoting Gary Plastic Packaging Corp. v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 903 F.2d 176, 180 (2d Cir.1990) (internal citation omitted)). Particularly with respect to the liability and injunctive relief phases, the potential defense against Scott's claims will have little if any relevance to the outcome of the litigation. So long as some of the class representatives can succeed in proving disparate impact and lack of business justification, the class claims for liability and injunctive relief will prevail. Accordingly, this challenge, too, must fail.
d. Adequacy
To satisfy the adequacy requirement of Rule 23, “[i] there should be no conflict between the interests of the class and the named plaintiff[s] nor should there be collusion among the litigants; and [ii] the parties' attorney [s] must be qualified, experienced, and generally able to conduct the proposed litigation.” Jolly Roger Offshore Fund Ltd. v. BKF Capital Grp., Inc., No. 07 Civ. 3923(RWS), 2007 WL 2363610, at *5 (S.D.N.Y. Aug. 16, 2007) (quoting, inter alia, Jackson v. Foley, 156 F.R.D. 538, 543 (E.D.N.Y.1994) ).
The Census Bureau understandably does not dispute the qualifications of the team of lawyers representing the Plaintiffs. Class Counsel in this case includes the law firm of Outten & Golden, as well as attorneys from various public interest groups including the Center for Constitutional Rights, Community Legal Services of Philadelphia, Community Service Society and Latino Justice PRLDEF of New York, the Indian Law Resource Center of Helena, Montana, and the Lawyers' Committee for Civil Rights and Public Citizen Litigation Group of Washington, D.C. Collectively, these attorneys bring to the case a wealth of class action litigation experience, much of which relates to claims of employment and race discrimination. Moreover, this team has devoted at least four years of time and substantial resources to litigating this case. There is no reason to believe that they cannot provide adequate legal representation to the Plaintiffs.
The Census Bureau does, however, dispute the named plaintiffs' adequacy to serve as class representatives. Specifically, the Census Bureau contends that the named plaintiffs have a weak incentive to litigate the issue of liability on behalf of the proposed class because none of them would have been hired had they passed the criminal background check and, therefore, none has a viable backpay claim. This contention, reminiscent of the Census Bureau's standing arguments, is unpersuasive.
A plaintiff is not disqualified as class representative simply because “he may fail to prove his case or [because] the defendant may have good defenses.” Robinson v. Sheriff of Cook Cnty., 167 F.3d 1155, 1158 (7th Cir.1999) (emphasis in original). Only if the named plaintiff's claim is a “clear loser” should he be excluded on adequacy grounds. Id. Whether the named plaintiffs in this case would have been hired in the absence of discrimination is a question that will arise only if there is a finding in favor of the Plaintiffs at the liability stage. At this preliminary stage of the litigation, the Court could do no more than speculate as to whether the Census Bureau would have hired any of the proposed representatives. Thus, even if that were the proper inquiry when deciding a plaintiffs' entitlement to monetary damages, one could not say at this stage that the named plaintiffs have a weak incentive to litigate the issue of liability. If anything, they may have a more powerful incentive to litigate their claims vigorously, since their claims may require stronger proof than those of applicants who unquestionably would have been hired but for the allegedly discriminatory practices. The fact that their damages claims ultimately may fail does not preclude them from serving as adequate representatives of the liability class.
In sum, I find that a liability class consisting, at least for the moment, solely of African–American claimants satisfies all of the prerequisites to class certification under Rule 23(a).
2. Rule 23(b)In addition to meeting the requirements of Rule 23(a), the Plaintiffs also must show that the proposed class satisfies at least one of the three alternative bases for a class action set forth in Rule 23(b). At the liability and injunctive relief stages, the Plaintiffs ask the Court to certify a single class under Rule 23(b)(2), consisting of “all African American and Latino applicants who applied for temporary employment during the 2010 decennial and were harmed by ... [the Census Bureau's] use of the 30–day [L]etter as a screening device[,] ... [or the Census Bureau's] use of adjudication criteria to screen applicants,” or both. (Pls.' Class Cert. Mem. at 26).
Rule 23(b)(2) allows class treatment in situations where the alleged wrongdoer “has acted or refused to act on grounds generally applicable to the class, thereby making appropriate final injunctive relief or corresponding declaratory relief with respect to the class as a whole.” The fundamental characteristic of a class properly certified under Rule 23(b)(2) is “the indivisible nature of the injunctive or declaratory remedy warranted—the notion that the conduct is such that it can be enjoined or declared unlawful only as to all of the class members or as to none of them.” Dukes, 131 S.Ct. at 2557 (quoting Nagareda, supra, at 132). Rule 23(b)(2) thus applies “when a single injunction or declaratory judgment would provide relief to each member of the class.” Id. Civil rights class actions charging parties with unlawful, class-based discrimination are “prime examples” of the type of case encompassed by Rule 23(b)(2). Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 614, 117 S.Ct. 2231, 138 L.Ed.2d 689 (1997).
Insofar as the Plaintiffs seek injunctive relief, they plainly satisfy the requirements of Rule 23(b)(2). In the event that the Court determines that the challenged hiring procedures had a disparate impact on minority applicants, there are a number of injunctive and declaratory remedies that the Court could order that would apply equally to all class members. For example, the Plaintiffs suggest that the Court could issue a decree enjoining the use of the 30–day Letter and mandating that the Census Bureau revise its Adjudication Criteria to ensure a more comprehensive, job-related screening process. At this stage in the litigation, it is not necessary to speculate about the precise contours of such relief. All that matters is that the Court has the equitable power and the practical ability to fashion some form of injunctive or declaratory relief that would apply to all members of the proposed (b)(2) class. There is no reason to believe that the Court will be unable craft such an appropriate remedy if and when it reaches that stage.
The Census Bureau contends that the Plaintiffs' claims for injunctive relief cannot be certified under Rule 23(b)(2) because the proposed class is “not ‘sufficiently cohesive to warrant adjudication by representation.’ ” (Def.'s Class Cert. Mem. at 39 (quoting Blackman v. District of Columbia, 633 F.3d 1088, 1094 (D.C.Cir.2011) )). Specifically, the Census Bureau argues that every putative class member would have had a different subjective reaction to the 30–day Letter and that those who responded to it were individually adjudicated based on their “infinitely different criminal histories.” (Id. ). In the Census Bureau's view, these individual differences destroy class “cohesiveness,” making class treatment inappropriate under Rule 23(b)(2). (Id. at 39–40).
Although class “cohesiveness” is not expressly required by the text of Rule 23(b)(2) or Supreme Court precedent interpreting that rule, some lower courts have denied certification on cohesiveness grounds due to the individualized nature of the injunctive relief that would be necessary to remedy each class member's injuries. For example, courts have raised cohesiveness concerns in mass tort cases in which plaintiffs' claims for injunctive relief include requests for future medical monitoring, since the need and desire for such monitoring would vary among individual class members. See, e.g., Gates v. Rohm & Haas Co., 655 F.3d 255, 264 (3d Cir.2011) ; Barnes v. Am. Tobacco Co., 161 F.3d 127, 146 (3d Cir.1998) ; In re Rezulin Prods. Liab. Litig., 210 F.R.D. 61, 75 (S.D.N.Y.2002) ; see also Amchem, 521 U.S. at 625, 117 S.Ct. 2231 (acknowledging concern over Rule 23(b)(2) certification in “ ‘mass accident’ cases[, which] are likely to present ‘significant questions, not only of damages but of liability and defenses of liability, ... affecting the individuals in different ways.’ ”) (quoting Fed.R.Civ.P. 23 advisory committee notes (1966)). Notably, however, even those cases do not suggest that class members all must have suffered the exact same injury in the exact same way. The cohesiveness element merely requires that plaintiffs' legal injuries be similar enough that they all can be remedied with a single “indivisible” injunction. See Barnes, 161 F.3d at 143 n. 18 (cohesiveness requirement exists because (b)(2) actions are intended to remedy “group, as opposed to individual injuries”). Although it may be true that the putative class members all had different reactions to the 30–day Letter, or that the Census Bureau individually adjudicated all class members who responded to the 30–day Letter, these differences do not affect the Court's ability to craft an injunction that would remedy all of the eligible class members' injuries in one fell swoop.
It bears mention that the proposed (b)(2) class, as currently defined, may ultimately encompass individuals who are not entitled to injunctive or declaratory relief. For example, the proposed class includes some individuals who allegedly were harmed only by the 30–day Letter, and some who allegedly were harmed only by the Adjudication Criteria. If only one of these two challenged hiring instruments is found to have disparately impacted minority applicants, those who fall outside the category of adversely-affected applicants will not be entitled to any relief, injunctive or otherwise. Should that occur, the Court will retain the power to amend the certification order, create subclasses that carve out class members who are entitled to relief, and craft injunctive relief accordingly. See Fed.R.Civ.P. 23(c)(1)(C), (c)(5) ; see also In re Sumitomo, 262 F.3d at 139 (noting the Second Circuit's “longstanding view that the district court ... has the ability ... to alter or modify the class, create subclasses, and decertify the class whenever warranted”). That this course of action may prove necessary in the future does not preclude certification at this early stage.
Accordingly, I find that a class limited to African–American applicants may be maintained under Rule 23(b)(2) for purposes of determining liability and affording injunctive and declaratory relief.B. Proposed Damages Subclasses: Rule 23(b)(3)
If the trier of fact ultimately concludes that the Census Bureau's hiring procedures had a disparate impact on minority applicants, the Court then will move on to the remedial phase of the case to determine the Plaintiffs' entitlement to damages. At this stage, the Plaintiffs propose that the Court certify three separate subclasses, each of which must independently satisfy all of the prerequisites of Rule 23(a) and at least one of the provisions of Rule 23(b). See Fed.R.Civ.P. 23(c)(5) (“[A] class may be divided into subclasses that are each treated as a class under this rule.”). The Plaintiffs request that the Court certify each of the damages subclasses under Rule 23(b)(3), which requires both that the proposed class members share common questions that are capable of generating common answers, and that those common questions “predominate over any questions affecting only individual members.” Fed.R.Civ.P. 23(b)(3). As the Supreme Court has noted, although the “same analytical principles” govern both Rule 23(a) and Rule 23(b), “Rule 23(b)(3)'s predominance criterion is even more demanding than [the commonality requirement of] Rule 23(a).” Comcast Corp. v. Behrend, –––U.S. ––––, 133 S.Ct. 1426, 185 L.Ed.2d 515 (2013) (citing Amchem, 521 U.S. at 623–24, 117 S.Ct. 2231 ).
The Plaintiffs do not seek certification of the damages subclasses under Rule 23(b)(2) —and wisely so. As the Supreme Court made clear in Dukes, “individualized monetary claims,” including claims for backpay, “belong in Rule 23(b)(3).” 131 S.Ct. at 2558. That decision sounded the death knell for claimants seeking to certify backpay claims under Rule 23(b)(2).
--------
The Supreme Court recently emphasized the stringency of the predominance requirement in Comcast Corp. v. Behrend, ––– U.S. ––––, 133 S.Ct. 1426, 185 L.Ed.2d 515 (2013). That case involved a class of current and former Comcast cable subscribers who sought damages for purported violations of the federal antitrust laws. The plaintiffs initially alleged damages under four theories of antitrust impact, and proposed a damages model that would have included damages under all four theories. The district court accepted only one of the alleged impact theories as capable of classwide proof, and rejected the remaining three. Id. at 1431. Notwithstanding the fact that the proposed damages model provided no method of isolating damages resulting from the one accepted theory of antitrust impact, the district court certified the proposed class under Rule 23(b)(3) for all purposes, including damages calculations. The Supreme Court eventually reversed that decision, holding that the district court had improperly certified the class “[i]n light of the [damages] model's inability to bridge the differences between supra-competitive prices in general and supra-competitive prices attributable to” the one theory of antitrust impact on which the plaintiffs would be permitted to proceed. Id. at 1435. Since the damages model itself did not isolate those class members who were entitled to damages under the accepted theory of liability, the Court concluded that individual questions of entitlement to damages would inevitably predominate over common questions at the remedial stage of the litigation. As the Court stated, in such circumstances, Rule 23(b)(3) could not “authorize treating subscribers ... as members of a single class.” Id.
The Census Bureau invokes Comcast to argue against Rule 23(b)(3) certification in this case. The gravamen of the Census Bureau's argument is that questions of individual entitlement to damages will necessarily overwhelm the litigation in the remedial phase because Title VII affords defendants a statutory right to rebut any presumptive entitlement to backpay with fact-specific, individualized defenses. In particular, the Census Bureau argues that it would be entitled to defend against each individual class member's claim by proving that the applicant would not have been hired even if they had survived the criminal background check. In its view, adjudicating these individual defenses would require the Court to engage in an exhaustive analysis of all the selection records pertaining to the positions sought by each applicant. As the Census Bureau explains, this analysis would require the Court to determine whether each individual applicant met the relevant citizenship requirements and scored high enough on the qualifying exam to have appeared on a selection record in the relevant geographic region, taking other applicants' veteran preferences into account. The Court then would have to apply the “rule of three” to determine whether the applicant could have been selected under any hypothetical circumstance. The Court also would have to consider the applicant's availability to work the hours required by the position, the applicant's willingness to work in the location specified, the applicant's ability to speak any foreign language required for the position, and, for a male applicant, his compliance with the federal selective service requirements. In addition, the Court would have to look to the date of each individual class member's application to determine the appropriate period of backpay, and then calculate the average number of hours worked in each geographic region to determine the appropriate backpay award. The Court also would need to consider whether each of the class members had mitigated their damages to a reasonable extent. Finally, the Court would have to determine whether any of the class members had committed job-related offenses in the past for which they properly could have been excluded. Having conducted part of this analysis for the eight named plaintiffs in this action, I can attest that the process would, indeed, require much time and effort. Given the highly individualized nature of this analysis, it is clear that individual questions would predominate over any common questions that might still linger at the damages phase of the litigation.
The Plaintiffs propose a shortcut that they contend could obviate the need to undertake the sort of detailed damages analysis proposed by the Census Bureau. Rather than conducting an individualized review of the damages sustained by each class member, the Plaintiffs would have the Court estimate the “hiring shortfall,” or the number of applicants who would have been hired absent the alleged discriminatory practices, estimate the aggregate wages that the shortfall hires would have earned had they been hired, and then distribute those aggregate wages on a pro rata basis. Citing pre-Comcast precedent from the Second Circuit, the Plaintiffs argue that this type of pro rata distribution of backpay is appropriate in situations when “the number of qualified class members exceeds the number of openings lost to the class through discrimination[,] and identification of the individuals entitled to relief would drag the court into a quagmire of hypothetical judgments' and result in mere guesswork.” Robinson, 267 F.3d at 161 n. 6 (quoting Catlett v. Mo. Highway & Transp. Comm'n, 828 F.2d 1260, 1267 (8th Cir.1987) ) (internal quotation marks omitted). The Plaintiffs contend that by applying this method, the Court could avoid the need to analyze each class member's entitlement to damages separately. Moreover, the Plaintiffs suggest that this model would eliminate all of the individualized questions that might otherwise preclude certification under Rule 23(b)(3).While simplicity is, of course, a virtue, the Plaintiffs' proposed damages model does not survive scrutiny under Comcast. Here, as in Comcast, the Plaintiffs' proposed damages model is vastly overinclusive, and doubtlessly includes individuals who are not entitled to backpay under the proposed theory of liability because they would not have been hired even absent the alleged discrimination. Moreover, as in Comcast, the Plaintiffs' pro rata model makes no attempt whatsoever to separate those class members who are entitled to damages from those who are not. Individual questions regarding each class member's entitlement to damages thus necessarily will predominate at the remedial stage. For these reasons, the Plaintiffs' proposed damages subclasses cannot be certified under Rule 23(b)(3).
C. Bifurcation under Rule 23(c)(4)
As an alternative to maintaining a single class action and certifying liability and damages classes separately under Rules 23(b)(2) and 23(b)(3), the Plaintiffs suggest that the Court exercise its discretion under Rule 23(c)(4) to certify a class only as to the issue of liability. Rule 23(c)(4) gives litigants the option, “when appropriate,” to bring or maintain a class action only with respect to particular issues. Fed.R.Civ.P. 23(c)(4). Prior to Comcast, the Second Circuit endorsed the use of that rule to bifurcate proceedings by certifying the liability and injunctive relief aspects of a proposed class action while leaving damages determinations to individual hearings. See In re Visa Check/MasterMoney Antitrust Litig., 280 F.3d 124, 141 (2d Cir.2001) (discussing decertification after liability phase to allow plaintiffs to litigate damages individually), overruled on other grounds by In re Initial Pub. Offerings Sec. Litig., 471 F.3d at 40 ; Robinson, 267 F.3d at 167–69 (holding that district court erred by refusing to bifurcate proceedings and certify the liability stage of Title VII claim for (b)(2) class treatment). Following Comcast, courts have taken varying positions as to whether—and if so, in what circumstance—this approach remains viable. See Jacob v. Duane Reade, Inc., 293 F.R.D. 578, 584–89 (S.D.N.Y.2013) (closely examining both pre- and post-Comcast case law). Some courts have taken the view that plaintiffs seeking class certification must offer “a damages model susceptible of measurement across the entire class,” and that this issue cannot be separated from questions concerning liability. See, e.g., Roach v. T.L. Cannon Corp., No. 3:10 Civ. 0591(TJM), 2013 WL 1316452, at *3 (N.D.N.Y. Mar. 29, 2013) (post-Comcast case refusing to certify a class to adjudicate common liability questions because damages could not be measured on a classwide basis). Others courts have held, however, that “where determinations on liability and damages have been bifurcated, the decision in Comcast —to reject certification of a liability and damages class because plaintiffs failed to establish that damages could be measured on a classwide basis—has limited application.” In re Whirlpool Corp. Front–Loading Washer Prods. Liab. Litig., 722 F.3d 838, 860 (6th Cir.2013) (internal citations omitted); see Miri v. Dillon, 292 F.R.D. 454, 464 (E.D.Mich.2013) (same); In re Motor Fuel Temperature Sales Practices Litig., 292 F.R.D. 652, 675 (D.Kan.2013) (same).
The Supreme Court's primary concern in Comcast was the plaintiffs' inability to provide a damages model that would allow the district court to award damages on a classwide basis. The plaintiffs in that case, however, did not request bifurcation, and nothing in the Court's ruling appears to have taken that option off the table in future lawsuits. Accordingly, it appears that Rule 23(c)(4) continues to provide a viable solution if damages cannot be determined on a classwide basis, so long the proposed class satisfies the requirements of Rule 23(a) and (b) with respect to liability. Accord Johnson v. Nextel Commc'ns, Inc., 293 F.R.D. 660, 669 (S.D.N.Y.2013) ; Indergit v. Rite Aid Corp., 293 F.R.D. 632, 659 (S.D.N.Y.2013) ; Jacob, 293 F.R.D. at 588–89. Of course, “[c]ourts should use Rule 23(c)(4)... only where resolution of the particular common issues would materially advance the disposition of the litigation as a whole.” In re Motor, 292 F.R.D. at 665 (internal quotation marks omitted). Courts thus should not use the bifurcation approach where, “despite the presence of a common [liability] issue, certification would not make the case more manageable.” Benner v. Becton Dickinson & Co., 214 F.R.D. 157, 169 (S.D.N.Y.2003) (internal quotation marks omitted).
In this case, certifying the Plaintiffs' proposed liability and injunctive relief class will materially advance the litigation and make the proceedings more manageable. By litigating these issues on a classwide basis, both the parties and the Court can avoid the extreme time and expense necessary to try each class member's claims individually. If the 30–day Letter and the Adjudication Criteria are found to have had a disparate impact on minority applicants, they necessarily will have caused a uniform injury to all members of the liability class. The only remaining questions at that point will be whether the Census Bureau can maintain personal defenses against individual plaintiffs' claims for backpay, and, if not, the amount of backpay owed. If and when the litigation reaches that stage, the Court will have a number of management tools at its disposal to help resolve these issues. For example, the Court could appoint a special master to preside over individual damages proceedings, or could decertify the class after the liability phase and provide notice to plaintiffs as to how to proceed to prove damages. In re Nassau Cnty. Strip Search Cases, 461 F.3d 219, 231 (2d Cir.2006). There is no need to decide at this time which avenue to pursue. What is important is that the Court has the tools to handle any management difficulties that may arise at the remedial phase of this litigation.
V. ConclusionFor the reasons set forth above, the Census Bureau's motion to dismiss for lack of subject matter jurisdiction, (ECF No. 225), and the Plaintiffs' motion for class certification, (ECF No. 166), are both GRANTED in part and DENIED in part. In addition, because Dr. Lundquist's expert report was not necessary to resolve the motion for class certification, the Census Bureau's motion to exclude her report for that purpose is denied as moot. (ECF No. 207).
To summarize, my rulings on the Census Bureau's motion to dismiss and the Plaintiffs' motion for class certification are as follows:
1. Plaintiffs Gonzalez, Riesco, and Kargbo lack standing and are dismissed from this case.
2. The Plaintiffs' class shall be limited to African–American applicants who sought temporary employment during the 2010 Decennial Census and claim to have been harmed by the Census Bureau's 30–day Letter, its Adjudication Criteria, or both. The class shall not include Latino applicants. If the Plaintiffs are able to identify a suitable Latino class representative, they may move to amend the Second Amended Complaint and the class certification order.
3. The Plaintiffs' class shall be certified under Rule 23(b)(2) for purposes of
determining liability and affording injunctive relief, but shall not be certified for purposes of resolving damages.
4. Plaintiffs Houser, Daniels, Rickett–Samuels, Scott, and Desphy shall be designated as class representatives. The Plaintiffs' current counsel shall be appointed as class counsel.ORDER AMENDING THE COURT'S JULY 1, 2014 MEMORANDUM DECISION AND ORDER
On June 28, 2013, Plaintiffs filed a Motion for Class Certification, seeking to certify the following classes pursuant to Rule 23(b)(2) for the liability phase of the case and for class-wide injunctive relief: All African American and Latino applicants who applied for temporary employment during the 2010 decennial and were harmed by one or both of the following employment practices: (1) Defendant's use of the 30–day letter as a screening device; (2) Defendant's use of adjudication criteria to screen applicants. ECF No. 176 at 26.
In their motion, Plaintiffs also requested certification for post-liability monetary relief claims under Rule 23(b)(3), or in the alternative of an issues class under Rule 23(c)(4), of the following subclasses: All African American and Latino applicants who applied for temporary employment during the 2010 decennial and were barred for one of the following reasons: (1) based solely on the procedural requirements imposed by the 30–day letter; (2) based solely on the delay in adjudicating the applicant; (3) based on exclusions that are not job related. ECF No. 176 at 26.
On December 16, 2013, Defendant filed a Motion to Dismiss for Lack of Subject Matter Jurisdiction (“Defendant's Motion to Dismiss”), seeking to dismiss Plaintiffs' Second Amended Complaint on the grounds that each of the named Plaintiffs lacked Article III standing to assert claims of disparate impact under Title VII of the Civil Rights Act of 1964, 42 U.S.C. §§ 2000e et seq. (“Title VII”). ECF No. 225.
On July 1, 2014, this Court issued a Memorandum Decision and Order on Plaintiffs' Motion for Class Certification and Defendant's Motion to Dismiss (“Order”). The Order dismissed the claims of one of the African–American named Plaintiffs, Vivian Kargbo, and the claims of both Latino named Plaintiffs, Ignacio Riesco and Anthony Gonzalez. For the reasons set forth in the Court's order, the Court concluded that Plaintiffs Kargbo, Riesco, and Gonzalez could not demonstrate an injury-in-fact sufficient to prove Article III standing. See ECF No. 265 at 238–39. The Court also granted certification under Rule 23(b)(2) of a class for purposes of determining liability and affording injunctive relief. Because the Court found that both of the Latino named Plaintiffs lacked standing, the class definition in the Order was limited to “African–American applicants who sought temporary employment during the 2010 Decennial Census and claim to have been harmed by the Census Bureau's 30–day Letter, its Adjudication Criteria, or both.” Id. at 254. The Order provided, however, that Plaintiffs could move to amend the Second Amended Complaint and class certification order in the event they were able to identify a suitable Latino class representative. Id.
On July 29, 2014, Defendant notified the Court that it had discovered an error in running the selection certificates for the geographic regions in which the named Plaintiffs could have been considered for temporary employment with the 2010 Decennial Census. ECF No. 278. Defendant indicated that, as a result of this error, one or more of the Latino plaintiffs who had been dismissed by the Court's July 1, 2014 Order may, in fact, have had standing to bring their claims. Id. at 1–2. Defendant thereafter informed the Court on August 6, 2014, that based on the newly produced selection certificates, Plaintiff Anthony Gonzalez would, in fact, have appeared on at least one selection certificate and therefore would have been eligible for hire under the standard set forth by the Court in its Order. ECF No. 280 at 2.
In light of Defendant's recent disclosure and its acknowledgment that, pursuant to the Order, Plaintiff Gonzalez would have appeared as eligible on at least one selection certificate for his geographic region, the Court hereby concludes that Plaintiff Gonzalez has Article III standing to proceed as a class representative. Further, the Court grants Plaintiffs' unopposed motion to join Edward Zahnle and Alexis Mateo as named plaintiffs and proposed class representatives for Latinos who applied for temporary employment with Defendant during the 2010 Decennial Census and claim they were thereafter harmed by the Census Bureau's 30–day letter. Notwithstanding Defendant's agreement to not oppose Plaintiffs' motion to join Edward Zahnle and Alexis Mateo as named plaintiffs and proposed class representatives, Defendant preserves all Article III standing arguments raised in Defendant's Motion to Dismiss.
Accordingly, the Court amends its July 1, 2014 Order insofar as that Order dismissed Plaintiff Gonzalez and limited the certified class to African–American applicants who sought temporary employment during the 2010 Decennial Census. Mr. Gonzalez is hereby reinstated as a class representative. The class definition is amended as follows:
The Plaintiffs' class shall be limited to (1) African–American applicants who sought temporary employment during the 2010 Decennial Census and claim to have been harmed by the Census Bureau's 30–day Letter, its Adjudication Criteria, or both; and (2) all Latino applicants who sought temporary employment during the 2010 Decennial Census and claim to have been harmed by the Census Bureau's 30–day Letter, its Adjudication Criteria, or both.
Nothing in this Order is intended to prevent the Government from exercising its right to challenge the Court's July 1, 2014 Order either through its motion for reconsideration, via appeal, or through additional motion practice or at trial as appropriate; nor does the Government waive any defenses as against the newly added Plaintiffs Gonzalez, Zahnle and Mateo; nor do Plaintiffs waive any corresponding right to oppose such defenses or other relief sought by Defendants.
The remainder of the Court's July 1, 2014 Order remains in effect in all other respects.
SO ORDERED.
29 NO. 1 Westlaw Journal Employment 4
Comments