Opinion
No. 13 C 1524
03-31-2020
JAMES ZOLLICOFFER and NORMAN GREEN, on behalf of themselves and similarly situated laborers, Plaintiffs, v. GOLD STANDARD BAKING, INC., PERSONNEL STAFFING GROUP, LLC d/b/a MOST VALUABLE PERSONNEL d/b/a MVP, Defendants.
OPINION AND ORDER
Plaintiffs Norman Green and James Zollicoffer, individually and on behalf of other similarly situated job applicants, believing that Defendants Gold Standard Baking, Inc. ("GSB") and Personnel Staffing Group, LLC, doing business as Most Valuable Personnel ("MVP"), systematically steered African Americans away from work assignments at GSB because GSB did not want African American workers bring this putative class action alleging race discrimination in violation of 42 U.S.C. § 1981. Plaintiffs now move to certify a class of African Americans who sought, but were largely denied, referrals to GSB as a result of Defendants' allegedly discriminatory policies. Plaintiffs move to certify the class under Federal Rule of Civil Procedure 23(b)(3). Defendants oppose class certification and move to strike the testimony of Plaintiffs' expert Dr. Marc Bendick.
Antwoin Hunt, a class representative since August 9, 2017, asked to withdraw as a named Plaintiff on December 21, 2019, well after the parties had submitted their briefs regarding Plaintiffs' motion for certification. The Court includes background information from Hunt as relevant to disposition of this motion, but it does not address any arguments concerning whether Hunt makes an adequate class representative.
After Hunt withdrew as a named Plaintiff in this case, the parties completed Hunt's deposition in connection with another case pending in this District and filed supplemental briefing with respect to the adequacy of the class representatives and counsel in this case. This Opinion incorporates the arguments raised in the supplemental briefing. The National Employment Law Project also filed an amicus brief in support of Plaintiffs' motion for class certification, which the Court considered over Defendants' objection.
Because the Court finds that Dr. Bendick's report satisfies the criteria for admissibility under Daubert and Rule 702, the Court denies Defendants' motion to strike Dr. Bendick's testimony. Additionally, the Court finds that Plaintiffs have carried their burden under Rule 23 to show that certification is appropriate and grants the motion for class certification.
BACKGROUND
I. Background on MVP and GSB
MVP, headquartered in Deerfield, Illinois, is a staffing agency that places employees in temporary positions with a variety of employers across the United States, including around the Chicago metropolitan area. MVP operated a branch office at 5637 W. Roosevelt Road, Cicero, Illinois ("Cicero office"), until December 2015. Thereafter, MVP sold the office to Elite Staffing. Under MVP's management, the Cicero office had approximately forty to forty-five client companies. The office staff included a manager, approximately six dispatchers who were responsible for managing payroll, recruitment, hiring, and assigning laborers, as well as van drivers, who transported workers and helped with recruitment. The manager, who oversaw the office's day-to-day operations, reported to MVP's president, Elijah Wilde, and Vice President of Operations, Darron Grottolo.
GSB is an industrial baking facility specializing in croissants and Danish pastries and located at 3700 Kedzie Avenue, Chicago, Illinois. GSB operates 24 hours a day, seven days a week, on three different shifts and relies in part on temporary laborers to maintain its workforce. From 2009 to 2015 (the "Class Period"), GSB contracted exclusively with MVP to supply temporary laborers. The contract provided that MVP employed laborers as temporary workers "for Client's benefit," paid laborers their wages, and maintained personnel and payroll records. Doc. 754 at 2. The contract also stated that GSB was responsible for supervising temporary laborers, and that "during the period Temporary Workers are performing services for Client, the duties and conduct of said Temporary Workers shall be wholly subject to the operational control and direction of Client, and, therefore, said Temporary Workers shall become the borrowed servants of Client for said period." Id. at 4.
MVP supplied workers to GSB for approximately ten different types of minimum-wage positions: packer, sanitation, curler, mixer, lamination, oven, cake sorter, pan feeder, icing aligner, and break/lunch person. While GSB preferred workers with experience, some of the assignments did not require any specific training or qualifications and MVP could fill these positions on a daily basis, also called "Dailies." Doc. 700-2 at 61-62. Other assignments, including the mixer, curler, cake sorter, and icing aligner positions, required some on-the-job training. MVP could not fill these positions on a daily basis and MVP provided workers hired in "temp to hire" positions that could potentially lead to full-time employment. Id. at 62.
This Opinion cites to the ECF header on documents rather than the original page number.
MVP relied on the same pool of laborers to fill both daily and temp to hire assignments. None of the positions required a criminal background check. The only requirement for all workers was to attend an orientation session at the Cicero office where laborers received instruction on GSB's "Good Manufacturing Practices" or "GMPs." Doc. 700-2 at 57, 64; Doc. 750 at 2. The GMPs prohibited laborers from wearing nail polish, jewelry, or earrings and required laborers to cover tattoos and wear a hairnet.
GSB's demand for laborers fluctuated day by day and week by week. Sometimes, there were no or very few open positions that needed filling. Other times GSB relied on MVP to provide more than a hundred laborers a day, sometimes requesting extra workers within twenty to thirty minutes. At times, GSB only needed workers for certain shifts that were difficult to fill, such as third shifts or weekend shifts. MVP employed a variety of tactics to meet GSB's needs. MVP stationed managers and supervisors, referred to as "Onsites," at GSB. Doc. 700-1 at 50. These employees took orders for laborers directly from GSB supervisors and personally recruited laborers through a variety of methods. The on-site supervisors kept notebooks with lists of workers they could call. They also worked hand-in-hand with dispatchers, who interfaced with walk-ins at MVP's Cicero office, and van drivers who went into neighborhoods to recruit laborers. MVP purchased advertisements in both English and Spanish in Spanish-language newspapers on an as-needed basis. MVP also gave flyers to dispatchers and van drivers to distribute near the Cicero office, close to the bakery, or in the neighborhoods where they lived.
Dispatchers did not refer individuals whom they thought might be affiliated with a gang. Although there was no formal training regarding how to identify gang affiliation, management instructed dispatchers to look at the candidates' behavior and appearance, the candidates' reaction when dispatchers told them about the requirements for work assignments, "whether their pants [were] hanging," and whether the candidate looked like he or she wanted to work. Doc. 700-2 at 63.
Additionally, GSB was not supposed to employ the same laborer for longer than ninety days unless GSB elected to hire that laborer for a permanent position. If GSB did not want a certain laborer to return, it could mark the individual as "Do Not Return" or "DNR." Id. at 66-67.
II. Discriminatory Policy
On February 27, 2013, Brian Lucas, Aronzo Davis, and Torrence Vaughans filed this putative class action against Defendants, alleging that MVP had a policy or practice of steering African Americans away from work assignments at GSB. None of the original plaintiffs remain in the case, and in their stead Green and Zollicoffer now act as class representatives on behalf of all African Americans who were allegedly denied work at GSB as a result of Defendants' discriminatory hiring policy. Plaintiffs offer the affidavits of several former MVP employees, summarized below, in support of their motion for certification. These employees testified that MVP's discriminatory policy was "common knowledge," Doc. 701-1 at 136, that management imposed it in a top-down manner, and that MVP's client, GSB, was the motivating force behind the policy. Plaintiffs also testified about their own experiences as job seekers at the Cicero office and how they witnessed MVP dispatchers giving preferential treatment to Latinos over African Americans. Finally, Plaintiffs offer the testimony of their expert witness, Dr. Marc Bendick, who concluded that there were substantial shortfalls in the number of African Americans hired during nearly the entire Class Period.
A. Maria Carretero
Maria Carretero worked at GSB from June 1999 until July 2014. She worked as a packer until September 2009 when GSB promoted her to third shift supervisor. As a supervisor, Carretero was responsible for overseeing employees of GSB as well as temporary laborers from MVP. Carretero testified that since she started as a supervisor, GSB had a strong preference for Latinos and "[i]t was common knowledge that GSB did not want to have African Americans at the factory." Doc. 700-1 at 136. Prior to 2013, Carretero testified that there were almost no African Americans at GSB as either employees or as temporary laborers, and that as far as she knew, "there was only one African American mechanic employed directly by GSB." Id. When MVP sent African Americans to fill temporary positions, GSB management would instruct on-site supervisors to DNR them. At some point in 2013, GSB began allowing African Americans to work as Dailies, but still did not want them to work as regular laborers.
Carretero testified to an incident involving Alejandro "Alex" Salgado—GSB's plant manager—in January 2014, when one of the production lines she was supervising became backed up. When Salgado inquired about the cause of the delay, Carretero explained that not everyone knew how to do their job. Although most of the workers on the line were Latinos, and there were a few African Americans, Salgado yelled "Quitan esos negros" ("get rid of those black people"). Id. At the end of the shift, GSB allowed all of the Latino workers to return, while marking all of the African Americans DNR.
Carretero also heard other GSB supervisors make derogatory remarks about African Americans. One supervisor referred to African Americans as "niggers" and stated that they were lazy and worthless. Id. Another referred to African Americans as "lazy, good-for-nothing 'negros.'" Id.
B. Rosa Ceja
Rosa Ceja began working for MVP in March 2014. MVP employed her as both an on-site supervisor at one of MVP's clients in Waukegan, Illinois, and as a dispatcher at various branch offices, including ones in Waukegan, Elmwood Park, and Niles, Illinois. Ceja testified that MVP accommodated clients who preferred Latino laborers over African Americans. She believed many of MVP's clients preferred Latinos because they were "less likely to complain about things like being injured on the job or being underpaid." Id. at 153. She described being "yelled at by representatives of client companies" for assigning them African Americans. She also testified that David Barnett, one of MVP's owners, warned her not to assign African Americans to certain clients "because MVP could lose the account." Id.
Ceja testified that when the Cicero office needed more laborers to fill positions, dispatchers would call the Elmwood Park office to request laborers, and vice-versa. This happened on almost a daily basis. When the Cicero office was trying to fill spots at GSB, GSB only wanted Latinos. Once, Grottolo called the Elmwood Park office and asked Ceja if MVP had "los que escuchan a La Ley," or "a La Que Buena," i.e. laborers who listened to the Spanish-language radio stations La Ley or La Que Buena, which Ceja understood to mean Latinos. MVP told dispatchers to use the terms bilingues (bilinguals) for Latino workers, and no bilingues (not bilingual), guapos (handsome ones), or morenas (dark ones) for African Americans. MVP management instructed her and other dispatchers to exclusively use code words to refer to a person's race and informed them that they could be fired for using terms like African American, Black, Hispanic, or Latino.
C. Daisy Corral
Daisy Corral worked as an on-site supervisor at MVP beginning in July 2014 and stayed with the office after its sale to Elite Staffing until June 2016. Corral testified that MVP occasionally assigned African Americans to GSB but that they were typically given daily assignments, rather than regular schedules that went to Latinos. In addition to being assigned weekend shifts, which relieved GSB of paying regular laborers overtime, African Americans often received assignments to "the worst jobs." Doc. 700-1 at 52. For example, African Americans were "disproportionately assigned to work in Danish Production or the oven room, where laborers are required to work in excessive and constant heat." Id.
Several GSB supervisors told Corral not to assign African Americans to their departments and complained if she did. When Corral sent her supervisor, Janet Rostro, an email suggesting that they rotate African Americans assigned to Danish production to other positions, "[n]othing happened." Id.
In the early fall of 2015, Rostro called a meeting for all on-site supervisors at GSB and indicated that GSB had instructed her to stop assigning African Americans. Rostro told the supervisors to advise African American laborers that GSB had a hiring freeze. Rostro also instructed supervisors not to advise African Americans about the mandatory orientation for people who wished to work at GSB. When the on-site supervisors complained and told Rostro that the new policy was illegal, she responded that this was what the customer wanted. The number of African American assigned to GSB subsequently declined. Corral described that MVP, and later Elite Staffing, maintained an Excel file titled "Employee Hours by Labor Account," that showed the number of hours worked per laborer assigned to GSB. Id. at 53. The number of hours worked by non-Latinos declined "dramatically" after the meeting. Id. In March 2016, Rostro told supervisors they could start bringing back African Americans laborers, but Corral testified there were still proportionately less African American workers.
D. Lissette Robles
Lisette Robles worked at MVP as a manager of the Cicero office from April 2011 until November 2013. The manager of the Cicero office handled "all marketing and advertising for employees on an as-needed basis." Doc. 700-1 at 2. During her time at the Cicero office, MVP primarily advertised, in both English and Spanish, in HOY, a Spanish-language newspaper. MVP would also distribute flyers on an as-needed basis, approximately half of which were in English and half in Spanish.
E. Pamela Sanchez
Pamela Sanchez worked for MVP as an on-site supervisor stationed at GSB from November 2013 to May 2014, and as a dispatcher at MVP's Cicero office from June 2014 until February 2015. While working as an on-site supervisor, Sanchez testified that GSB staff for every department told her that they did not want African Americans and preferred Latino laborers. Id. at 142. This included Salgado, who said "he wanted only Hispanic laborers" and "told [her] directly not to assign African Americans to work at GSB because they were lazy, they were trouble makers and their job performance is poor." Id. at 141.
Sanchez used many ways to fill positions according to GSB's preferences. Prior to her shift, MVP dispatchers would send her a list of prospective laborers. When Sanchez received the list, she would advise the dispatcher to cancel laborers with non-Hispanic names because she knew that GSB would reject them. Sanchez testified that MVP rarely assigned African Americans to GSB, and that if they did it was usually only on weekends—so that GSB could avoid paying its normal laborers overtime—or to third shifts which were harder to fill. On the rare occasion that Sanchez saw African Americans assigned to the third shift, GSB employees told her to DNR those workers.
Sanchez also personally recruited laborers in person or over the phone. Sanchez kept a list of laborers in a notebook whom she could call. Sanchez kept a person's contact information if they had previously completed a work assignment, or if they called the MVP office phone at GSB. Because of GSB's preference for Latinos, Sanchez only kept the names of Latino laborers based on their last name. Other on-site supervisors also kept notebooks and similarly kept the names and contact information for Latino laborers only.
Sanchez would also work with MVP van drivers, who generally went to predominantly Latino neighborhoods in Chicago, such as Little Village, to recruit workers. MVP also handed out flyers in Latino neighborhoods advertising work at a Chicago bakery. Sanchez testified these flyers were in Spanish and she never saw one in English.
Sanchez testified that in January 2014, there was a meeting with the on-site supervisors, dispatchers, and van drivers at the Cicero office after Salgado sent an email to MVP management complaining about unfulfilled job orders. Sanchez, as well as other supervisors and dispatchers, complained that they could not assign African Americans because GSB would DNR them. Barnett and Grottolo, who were both present at the meeting, indicated they would talk to the owner of GSB and Salgado about accepting workers regardless of their race. Sanchez testified that GSB supervisors continued to tell on-site supervisors to DNR African Americans after the meeting.
After Sanchez began working as a dispatcher, other dispatchers trained her to use two different methods to deal with walk-ins depending on whether they were African American or Latino. Dispatchers advised African Americans that they could only give them an application if there was a job opening and that they could come back at 4:00 a.m. to see if there was any work. Sanchez would sometimes assign African Americans to difficult or unpopular jobs, such as those that required long shifts, or those that involved working in freezing temperatures, also known as "cold sites." Id. at 145. For Latinos, by contrast, Sanchez and other dispatchers would immediately give them an application and put them on a work ticket or call clients to see if there were positions available.
Sanchez also testified regarding an incident with a particular worker named Kevin James. James, who is African American, completed an application at the Cicero office on October 15, 2014. Sanchez advised James that he could return the next morning to be assigned to one of two client companies, one of which was GSB. Sanchez sent a closing email to first-shift dispatchers confirming that James was scheduled to work the next morning. When James returned to the Cicero office, he waited from 4 a.m. until 9 a.m. but dispatchers never gave him an assignment. Morning dispatchers claimed that James never appeared, although Sanchez knew this to be false. When James returned that afternoon to complain to Sanchez, Sanchez pursued the issue with the Cicero office manager. The manager told Sanchez that she should not confirm African Americans for work and should instead tell them to come back in the morning to see if there were openings.
During the pendency of this case, Defendants alleged that Sanchez committed perjury, fabricated evidence at the direction of Plaintiffs' counsel, and violated a confidentiality agreement with MVP. The Court found that Defendants' allegations were appropriate to present to the trier of fact rather than the Court. A detailed description of the facts is set out in the January 16, 2019, Memorandum Report and Recommendation, Doc. 680, which the Court adopted on February 27, 2019.
F. Alex Salgado
On June 17, 2014, Salgado wrote the following e-mail to GSB president Yianny Caparos and CFO Mark Grosshans:
I'm typing this email with a huge level of frustration, we have been having a lot of issues with MVP now we are having problems no matter what shift/day it's just crazy, I just came from lines 1 and 2 packaging and the people they brought for clamshells are just standing, sleeping one of them with a watch on his hand and I called the lead upstairs to join me there and she was telling me that she is [sic] been getting all kinds of people that she knew those will not work but is what she gets from MVP main office [sic]. I spoke to Janet and she just told me that she will call the office, yesterday I ask Louis and Janet for a meeting regarding the
weekend mess. . . . (we have been having the house full with Afro-American people) and just not been working [sic].Doc. 700-3 at 17. Salgado also suggested that GSB work with a different staffing agency. Caparos responded that he would participate in discussions when he returned the following Thursday and that Salgado "need[ed] to hold Louis accountable." Id. Approximately two weeks later, on June 30, 2014, GSB issued Salgado a disciplinary letter, stating that his email had violated the company's affirmative action and anti-discrimination policies. GSB required Salgado to attend a training session and noted that it was the company's policy to provide equal employment opportunities to all qualified individuals regardless of their race or color.
G. Norman Green
Green first sought work at the Cicero office on October 10, 2013. On his first day, he completed an application and put down two phone numbers: one for his cell phone and one for a family member. Green asked to work immediately, and staff instructed him to place his name on the sign-in sheet. Green understood that MVP provided work assignments on a first-come, first-served basis and waited for approximately four hours for his name to be called. When he did not receive an assignment, the office staff instructed him to come back at 8 a.m. the next day. When Green returned the next day, he again waited for several hours without an assignment. This happened on several occasions, during which Green noticed that Latino workers would arrive after him, receive IDs, and receive work assignments before him, or that vans carrying Latino laborers would arrive and MVP sent those laborers to work immediately after they signed in.
On November 22, 2014, Green signed a declaration in which he erroneously stated that he first sought work at MVP beginning in July 2014.
Green visited the Cicero office approximately ten to fifteen times in 2014 but only received a few work assignments, none of which were to GSB. At his first assignment, to a meat packing factory, the company released Green after six hours into what was supposed to be an eight-hour shift. At what was supposed to be his second assignment to a chocolate factory, Green waited several hours before being told he had the incorrect type of boots even though an MVP dispatcher had already approved his boots. At his third assignment, Green borrowed bus fare to get to work after MVP's dispatcher promised him a ride home from a job. When Green was picked up from work close to 1 a.m., the driver dropped him off at a train station that had no trains running for several hours. The next day MVP called to ask if he wanted to go back. Green refused but said he was willing to work if they had a different assignment for him. He never received another assignment.
On January 5, 2017, the parties deposed Green as a part of this case. Green testified there were "lots of months" in 2013 when he did not have a working cell phone. Doc. 711-14 at 2. Green relied in part on prepaid phones, and when he had enough money to pay for a phone it was often easier to purchase a new one rather than place more money on his old phone. The new phone, in that case, would also have a new number. Green also testified that he met with his attorney "plenty of times" before he became a named Plaintiff in the case. Id. at 14. When asked about his allegations against GSB, Green testified that he knew "nothing about no bakery," and that he was not aware that he was suing GSB. Id. at 12.
H. James Zollicoffer
Zollicoffer first sought work in 2007 at MVP's branch office in Prospect Heights, Illinois, where he received regular assignments for several months. When Zollicoffer moved to Chicago around the summer of 2009, the Prospect Heights dispatcher directed him to the Cicero office. When he first visited the Cicero office, Zollicoffer signed in, completed an application, and waited several hours without receiving an assignment. Zollicoffer returned several times over the course of about two weeks, sometimes arriving at around 4 or 4:30 a.m. and waiting for an hour and a half to two hours. On one occasion he was approximately the tenth or eleventh person to sign in. MVP staff told Zollicoffer that they would call him or instructed him to check back to see if there were openings. Zollicoffer, however, observed that MVP consistently assigned work to Latino laborers but not to African Americans. Zollicoffer also called the office three to four times a week for approximately another month to inquire about work. He never received an assignment.
During Zollicoffer's February 2, 2018 deposition, he testified that he could not recall how many times he visited MVP in 2009, or whether he sought work at MVP in 2010, 2011, or 2015. Zollicoffer thought that he did not seek work at MVP in 2012 or 2013 because he was working elsewhere at the time. Zollicoffer also testified that he held a number of different jobs throughout the Class Period. Throughout 2012, for example, he worked for the staffing agencies Labor Solutions and Labor Temps, and in 2012 and 2013, he worked at the Georgia Nut company through MVP's Prospect Heights office. When Zollicoffer had consistent work elsewhere, he did not seek work at the Cicero office.
Zollicoffer testified that he pleaded guilty to several crimes in his life, including armed robbery in 1986, controlled substance crimes in 1997 and 1998, possession of a stolen vehicle in 1999, aggravated fleeing police in 2007, theft in 2008, and misdemeanor domestic battery in 2011. In 2011, Zollicoffer was incarcerated for approximately forty-five days. Zollicoffer thought that he may have been incarcerated in 2014 as well.
Counsel for Defendants also asked Zollicoffer whether he knew what it meant to be a class representative and what his role as a plaintiff entailed, to which Zollicoffer answered: "No." Doc. 713-5 at 3. Counsel also asked Zollicoffer regarding his claims against GSB:
Q: Are you claiming that GSB discriminated - what are you claiming GSB discriminated against you?
A: If it's through the company, I never said GSB. I said MVP.Doc. 758 at 186-88.
Q: Okay.
A: You're saying GSB. I never said GSB out of my mouth I don't recall saying it if I did. I'm saying the company that sent me there.
Q: Okay. Are you claiming that GSB, Gold Standard Baking, discriminated against you based on race?
A: Am I claiming that? No, I'm not claiming it.
. . .
Q: And other than that eavesdropping on that conversation you mentioned with Mr. Richardson, you don't have any basis for saying that GSB discriminated against African Americans, right?
A: None whatsoever.
I. Antwoin Hunt
Hunt, who withdrew as a named Plaintiff in December 2019, first began working for MVP beginning in the mid-2000s. In 2010, he completed a new application and sought work several times from 2010 to 2013, sometimes arriving at 4:00 or 4:30 a.m. before the doors opened. He consistently saw Latino laborers assigned work before African-Americans, even when they arrived late, just before assignees were sent out. Hunt received work assignments during this time period, but none were to GSB. Then, in August 2014, Hunt received his first assignment to GSB working overnights. During his time at GSB, Hunt noticed that most of the workers were Latino, and that all of the work schedules, signs, and instructions were entirely in Spanish. During his training, GSB also played a video in Spanish with English subtitles. A few months after his first assignment, GSB offered him a position as a regular temporary worker. Hunt had regular issues with one his supervisors, Guadalupe Garcia. She would, for example, give him his breaks in quick succession at the beginning of his shift instead of spacing them out as she did for the Latino laborers. Even though Hunt worked more than ninety days at GSB, the bakery never offered him a permanent position. In February 2015, Hunt showed up to work at his usual time when, without any warning, GSB staff did not allow him into the building. GSB marked him as DNR for poor performance. MVP placed Hunt at another client shortly thereafter but continued to drop him from the list of regular workers assigned to the client and failed to pay him the hourly wage he was promised.
On February 8, 2017 and February 19, 2018, the parties deposed Hunt as part of this case. In the second deposition, Defendants' counsel asked Hunt whether he had appeared as a witness or as a plaintiff against Defendants when he testified in the first deposition. Hunt replied: "I wasn't suing anyone. I was, I was here on a deposition because I felt that I was mistreated, so I hired my attorney right here, Chris, and the deposition came behind this." Doc. 711-10 at 2. Counsel also asked Hunt about a previous employment application in which he put down "seasonal temp to hire" as the reason he stopped working at his previous employer. When counsel asked Hunt why he put this down, he replied: "I'm filling blanks out so I don't leave [a] line short. People lie on applications all the time." Id. at 6. Hunt also admitted that at times, he would "stretch the truth." Id. at 8.
During the pendency of this case, the Court sanctioned Hunt for failing to disclose certain facts and documents during discovery. The May 29, 2018, Memorandum Report and Recommendation, Doc. 592, which the Court adopted on June 28, 2018, Doc. 595, sets forth a detailed description of the facts related to the sanctions. In connection with the sanctions order, Judge Kim ordered Hunt to personally pay $1,976, in addition to costs and reasonable attorneys' fees incurred in connection with his continued deposition, by January 7, 2019. Docs. 654-55.
In addition to this case, Hunt also served as a named Plaintiff in two other matters, Hunt v. Personnel Staffing Group, LLC, No. 16 C 11086 (N.D. Ill.), and Lucas v. Vee Pak, Inc., No. 12 C 9672 (N.D. Ill.). Hunt left his December 20, 2019 deposition in Vee Pak during the lunch break and did not return. That deposition also revealed that Hunt had failed to disclose certain prior litigation in which he had been involved. The following day, Plaintiffs in this case filed a motion to withdraw Hunt's appearance as a Named Plaintiff, explaining that he had moved to Texas and that the obligations of being a Named Plaintiff in all three cases had proved overwhelming. The Court granted the withdrawal request on January 7, 2020. Subsequently, Defendants in this case moved to reopen discovery to further investigate issues that arose during Hunt's December 20, 2019 deposition. Hunt sat for a continued deposition on February 16, 2020.
Of note during the December 20 and February 16 depositions, Hunt acknowledged not having much involvement in or understanding of this case. He suggested he had not had much interaction with his counsel. Hunt also testified that he had become involved in this case through Richard, whom Hunt thought may have been part of a workers advocacy group. Counsel for Defendants also asked Hunt about a cell phone he used during the relevant time period, from which Plaintiffs only disclosed a video Hunt took while working at GSB. Hunt indicated other potentially relevant information could be on that phone but that he did not know where it was. After the deposition, counsel for Plaintiffs represent that Hunt informed them it is in a storage unit under his control. Finally, the continued deposition revealed that one of the attorneys for Plaintiffs in this case, Christopher Williams, paid the entirety of the monetary sanction imposed on Hunt, with Williams and Hunt having an agreement for Hunt to repay the sanctions out of either settlement or personal funds. Hunt testified that he has only paid Williams $200, while Williams' ledger reflects two payments in the amounts of $200 and $150.
III. Expert Witness
A. Discovery Disputes Related to MVP Records
During discovery, Plaintiffs sought records from Defendants documenting the laborers who sought work at MVP and which laborers MVP assigned to GSB. This included the sign-in sheets that laborers used at the Cicero office. Plaintiffs intended to analyze the sign-in sheets to determine the number of African Americans who sought work at MVP, as well as the order in which people checked in—data that Plaintiffs could compare to the number of African Americans MVP actually assigned to GSB. In response, MVP indicated that it "had no policy or practice with respect to the use or maintenance of . . . sign-in sheets" and it "has not determined and/or collected information on the race or ethnicity of its applicants and employees." Doc. 644-1 at 26, 34.
Plaintiffs consulted with Dr. Marc Bendick, a labor economist, and determined they could analyze the racial makeup of MVP job applicants and referrals using laborers' addresses. They would do this in two ways. First, because MVP referred people to short-term, low-paid positions, it is typically not cost-effective for laborers to relocate or commute long distances for jobs. Thus, "the home addresses of laborers registered at the MVP Cicero office are expected to fall within a limited geographical area surrounding that office." Doc. 601-8 ¶ 11. This area, called the "reasonable recruitment area," Doc. 644-1 ¶ 10, could be used to estimate the expected representation of African Americans among those who sought work at MVP based on data from the US Census Bureau. Second, Plaintiffs could utilize a technique called geocoding to estimate the number of African Americans MVP actually referred to GSB. This involved converting home addresses to coordinates on a map, and then using Census Bureau data to calculate the probability that the individuals living at each address is of a certain race.
Plaintiffs subsequently sought records of the names and addresses of Cicero office job applicants and referrals to GSB. Initially, MVP indicated it was too burdensome to produce a list of job applicants restricted to the Class Period because of how it structured its database. The parties eventually came to an agreement, with the Court's approval, to resolve Plaintiffs' discovery request. MVP agreed to produce two lists: the first would include the name, address, and MVP ID number for laborers MVP assigned out of the Cicero office from when it opened in 2000 until September 1, 2014 (the "Master Address List"). The second list would include the names of laborers the Cicero office paid during the Class Period ("Class Period Assignment List"). The parties agreed that Plaintiffs' counsel would winnow down the first list by eliminating employees who were not also on the second list. The result would be a list with the pertinent information that Plaintiffs had requested, including the names and addresses of laborers who received work assignments from the Cicero office during the Class Period. Plaintiffs agreed they would return copies of the original, unfiltered Master Address List after they culled it.
Pursuant to this agreement, MVP produced the two Excel files to Plaintiffs. The Master Address List contained 64,142 rows of information, sometimes with multiple rows for a single employee, as well multiple addresses separated by commas under the same column, i.e. in comma-separated values ("CSV") format. The second list contained the names of employees MVP paid—and therefore assigned work—during the Class Period. Plaintiffs' counsel merged the two lists, removing employees from the Master Address List who were not on the Class Period Assignment List. The result was a list of 18,191 Cicero office referrals during the Class Period with corresponding home addresses ("List of 18,191 Referrals").
As addressed further below in the Analysis, Plaintiffs dispute that the Master Address List was in CSV format.
On October 7, 2014, Plaintiffs sent the List of 18,191 Referrals to Defendants and asked Defendants' counsel to review and confirm that the list accurately reflected the individuals to whom MVP assigned work from the Cicero office during the Class Period. Defendants did not follow up to verify or challenge the list's accuracy. Consequently, Plaintiffs' counsel provided the List of 18,191 Referrals to Dr. Bendick and represented that this was an accurate list of the employees assigned out of the Cicero office during the Class Period, and that Dr. Bendick could rely on it for his analysis. Later, on November 14, 2017, MVP produced two additional files to Plaintiffs. The first contained updated address information for all laborers MVP referred out of the Cicero office categorized by year, from 2009 to 2015 (the "Yearly Cicero Address Files"). The second contained the work hours for each laborer assigned to GSB also categorized by year ("Yearly GSB Hours Files").
Plaintiffs state that Defendants "refused [to] engage in a discussion" about the List of 18,191 Referrals and whether it could be used as a common data base on which both parties could rely. Doc. 644-1 ¶ 27. Defendants argue that this characterization is misleading, and counter that Plaintiffs' counsel never explained how they compiled the list. But Defendants do not refute that Plaintiffs' counsel sent Defendants an email on October 7, 2014, with the list of 18,191 Referrals attached, and specifically asked MVP's counsel to review and confirm that the list was accurate. Plaintiffs' counsel sent a follow-up email on January 15, 2015, indicating that they had not heard back from Defendants regarding the request. Further, Defendants concede that "MVP would not and did not stipulate to a common database of any sort." Doc. 652 at 8.
B. Dr. Bendick's Analysis
The first step in Dr. Bendick's analysis required identifying MVP's reasonable recruitment area. With the help of an assistant, William Glimpse, Dr. Bendick located each address from the List of 18,191 Referrals within a geographic zone called a Public Use Microdata Area (a "PUMA"). PUMAs are geographically contiguous areas encompassing at least 100,000 residents according to the decennial census. They are the smallest geographic zones for which the Census Bureau's American Community Survey ("ACS") releases data relevant to calculating the racial and ethnic characteristics of MVP job seekers. Dr. Bendick located the addresses within PUMAs for both the 2000 and 2010 Census. He then listed the PUMAs from largest to smallest according to the number of addresses located within the area and identified the largest PUMAs that accounted for 95% of potential laborers. This came out to nineteen PUMAs from the 2000 census and eighteen from the 2010 census that Dr. Bendick defined as the reasonable recruitment area.
This is technically what "geocoding" refers to: converting an address into latitudinal and longitudinal coordinates. Usually, this is to infer certain information about the people living at the address using Census Bureau information. See Kevin Fiscella & Allen M. Fremont, Use of Geocoding and Surname Analysis to Estimate Race and Ethnicity, Health Serv. Res. 41, 1482-1500 (Aug. 2006). In this case, the Court utilizes the term geocoding as it relates to converting an address into coordinates and then linking this to Census Bureau data to calculate the probability that the person living at the address is of a certain race. Id.
Second, Dr. Bendick counted the persons residing within the reasonable recruitment area who were likely to be qualified, available, and interested in MVP work referrals according the ACS. This included people who 1) were in the civilian work force, i.e. working or actively looking for work; 2) had no, little, or some education up through post-secondary school, but not people with college or graduate degrees; and 3) earned less than the median annual earnings in the Chicago metropolitan area. From 2009 to 2011, across PUMAs defined by the 2000 census, 26.8% of potential MVP job seekers were non-Hispanic African American ("NHAA"). From 2012 to 2013, across PUMAs defined by the 2010 census, 33.5% of potential MVP job seekers were NHAA.
Dr. Bendick's methodology here is a little unclear. The Census Bureau updates the ACS annually and releases statistical information in five-year data sets. The 2013 ACS data, for instance, was compiled according to data the Census Bureau collected from 2009 to 2013. At the time that Dr. Bendick calculated the reasonable recruitment area, this was the most recent ACS data available, and Dr. Bendick states that he solely used the 2013 data set for steps one and two. To the Court's understanding, the ACS only released data for 2009 to 2013 according to 2010 PUMAs. Only five-year data sets concluding in 2011 or earlier utilized the 2000 census PUMAs. Thus, if Dr. Bendick utilized ACS data for PUMAs defined by the 2000 census, it seems he would have needed to utilize the 2010 or 2011 ACS, which collected data going back to 2006 and 2007 respectively.
Third, Dr. Bendick combined the data for all five years, using the weighted average of the 26.8% and 33.5% figures. Dr. Bendick concluded that 29.5% of MVP job seekers from 2009 to 2013, determined according to MVP's reasonable recruitment area, were expected to be NHAA. This figure included two PUMAs from the 2000 Census and one PUMA from the 2010 Census that were not contiguous with the other PUMAs and that Dr. Bendick considered to be outliers. Excluding these PUMAs yielded a slightly higher expected representation figure of 30.6%, which Dr. Bendick considered a reasonable alternative expected representation figure.
Fourth, Dr. Bendick "consider[ed] the overall conservatism of the estimated availability." Doc. 701-1 at 74. Dr. Bendick considered his estimated expected representation figure to be conservative because 1) it excluded African Americans of Hispanic origin and 2) it included the outlier PUMAs that lowered the expected representation figure.
After estimating the expected availability of African Americans, Dr. Bendick calculated the estimated actual representation of NHAA among MVP referrals to GSB. To do this, Dr. Bendick used geocoding, converting each address to latitude/longitude coordinates and then linking the address to Census Bureau data about the neighborhood, specifically census blocks—geographic areas encompassing approximately forty-eight residents each. Dr. Bendick could then infer the probability a person referred to GSB was NHAA according to the proportion of NHAA living within the block in which the laborer's home was located.
Dr. Bendick began his analysis with a list of 11,255 referrals to GSB during the Class Period ("List of 11,255 Referrals"). Dr. Bendick compiled this list depending in part on the Yearly GSB Hours Files. Dr. Bendick computed a probability of being NHAA for approximately 72.4% of the referrals. He then applied this probability figure to all referrals in each year. Dr. Bendick estimated the actual representation of NHAA laborers as follows:
Dr. Bendick testified that he compiled the List of 11,255 Referrals based on the Yearly GSB Hours Files during his deposition. The Yearly GSB Hours Files, however, did not contain any addresses for referrals, and Dr. Bendick could not explain from where he pulled the addresses.
| Year | Non-Hispanic African Americans amongPersons Referred |
|---|---|
| 2009 | 1.7% |
| 2010 | 2.8% |
| 2011 | 3.9% |
| 2012 | 5.0% |
| 2013 | 9.9.% |
| 2014 | 28.2% |
| 2015 | 33.9% |
| Year | Expected NHAAReferrals | NHAA amongPersons Referred | Shortfall in NHAAReferrals | StandardDeviations | |||
|---|---|---|---|---|---|---|---|
| % | # | % | # | % | # | ||
| 2009 | 29.5% | 91 | 1.7% | 5 | 27.8% | 86 | 10.7 |
| 2010 | 29.5% | 251 | 2.8% | 23 | 26.7% | 228 | 17.1 |
| 2011 | 29.5% | 391 | 3.9% | 52 | 25.6% | 339 | 20.4 |
| 2012 | 29.5% | 803 | 5.0% | 135 | 24.5% | 667 | 28.1 |
| 2013 | 29.5% | 464 | 9.9.% | 156 | 19.6% | 308 | 17.0 |
| 2014 | 29.5% | 781 | 28.2% | 748 | 1.3% | 34 | 1.5 |
| 2015 | 29.5% | 539 | 33.9% | 618 | No shortfall | ||
| 2009-2014Total | 29.5% | 2,781 | 11.9% | 1,119 | 17.6% | 1,662 | 37.5 |
Dr. Bendick's Rebuttal Declaration contained figures slightly different than the figures in his Revised Declaration. The Court reproduces what it believes to be the most accurate figures here.
C. Dr. Bendick's Rebuttal Declaration
Defendants consulted with Dr. Carole Amidon, an economist specializing in analyzing labor economics, statistics, large databases and applied econometrics, to review Dr. Bendick's work. In their motion to strike Dr. Bendick's opinion, Defendants attached a report by Dr. Amidon criticizing several aspects of Dr. Bendick's analysis. Dr. Bendick subsequently submitted a rebuttal report, recalculating his estimates according to some of Dr. Amidon's critiques. Dr. Bendick concluded that the adjusted analysis still yields statistically significant shortfalls from 2009 to 2013.
Dr. Bendick writes that he incorporated Dr. Amidon's "suggested modifications" into his rebuttal declaration. Doc. 714-6 at 3. Defendants and Dr. Amidon dispute this characterization and argue that her critiques did not offer suggestions that would necessarily yield an accurate result.
| Shortfall Analysis from Dr. Bendick's Rebuttal Declaration | ||||
|---|---|---|---|---|
| Year | ExpectedNHAAReferrals | NHAA among PersonsReferred | Shortfall in NHAAReferrals | StandardDeviations |
| 2009 | 27.4% | 3.9% | 23.5% | 9.3 |
| 2010 | 27.4% | 5.3% | 22.1% | 14.5 |
| 2011 | 27.4% | 6.4% | 21.0% | 17.2 |
| 2012 | 27.4% | 7.4% | 20.0% | 22.4 |
| 2013 | 27.4% | 13.2% | 14.2% | 12.6 |
| 2014 | 27.4% | 34.6% | No shortfall | |
| 2015 | 27.4% | 39.0% | No shortfall | |
| 2009-2013Total | 27.4% | 19.3% | 35.6 |
Presumably the average representation figure for NHAA among GSB referrals is lower in Dr. Bendick's rebuttal declaration than it was in Dr. Bendick's original report because a larger number of addresses were incorporated in the analysis, placing greater weight on the years with a shortfall.
ANALYSIS
I. Motion to Strike Marc Bendick's Expert and Rebuttal Reports
Before considering the motion for certification, the Court considers Defendants' motion to strike Dr. Bendick's expert reports. Defendants argue that the Court should strike Dr. Bendick's Revised Declaration because it is insufficiently reliable under the standards set forth in Daubert and Federal Rule of Evidence 702. Defendants also argue that the Court should strike Bendick's rebuttal declaration for being procedurally and substantively improper.
A. Legal Standard
"[W]hen an expert's report or testimony is critical to class certification . . . a district court must conclusively rule on any challenge to the expert's qualifications or submissions prior to ruling on a class certification motion." Am. Honda Motor Co. v. Allen, 600 F.3d 813, 815-16 (7th Cir. 2010). Federal Rule of Evidence 702 and Daubert v. Merrell Dow Pharmaceuticals Inc., 509 U.S. 579 (1993), govern the admissibility of expert evidence. See Bielskis v. Louisville Ladder, Inc., 663 F.3d 887, 893 (7th Cir. 2011). Together, Rule 702 and Daubert provide that an expert's testimony is admissible if: 1) the witness is qualified, 2) the expert's methodology is reliable, and 3) the testimony will assist the trier of fact to understand the evidence or to determine a fact in issue. Myers v. Ill. Cent. R.R. Co., 629 F.3d 639, 644 (7th Cir. 2010). The Rule 702 inquiry "is a flexible one." Daubert, 509 U.S. at 594. "Determinations on admissibility should not supplant the adversarial process; 'shaky' expert testimony may be admissible, assailable by its opponents through cross-examination." Gayton v. McCoy, 593 F.3d 610, 616 (7th Cir. 2010). The proponent of the testimony bears the burden of proving that the proffered testimony meets these requirements, and the Seventh Circuit grants the district court "wide latitude in performing its gate-keeping function." Bielskis, 663 F.3d at 894.
B. Dr. Bendick's Qualifications
"Whether a witness is qualified as an expert can only be determined by comparing the area in which the witness has superior knowledge, skill, experience, or education with the subject matter of the witness's testimony." Gayton, 593 F.3d at 616 (quoting Carroll v. Otis Elevator Co., 896 F.2d 210, 212 (7th Cir. 1990)). Defendants do not contest Dr. Bendick's qualification as an expert. He has served as an expert witness or consultant in more than 200 cases where he has analyzed issues including the demographics of job-seekers, processes related to recruitment, hiring, promotion, compensation and discipline of employees, as well as damages for workers denied work opportunities. Dr. Bendick has authored 138 research papers, including articles in peer-reviewed journals. Federal and state courts across the country have considered his testimony as an expert in the field of labor economics. See, e.g., Hardie v. Nat'l Collegiate Athletic Ass'n, 876 F.3d 312, 320-21 (9th Cir. 2017) (racial disparities in hiring); Houser v. Pritzker, 28 F. Supp. 3d 222, 243 (S.D.N.Y. 2014) (racial disparities in hiring). Similar to these cases, Dr. Bendick seeks to testify regarding racial disparities in MVP's referrals to GSB. He is qualified to do so.
C. Admissibility of Rebuttal Declaration
Defendants bring two motions to strike. First, they ask the Court to disregard Dr. Bendick's rebuttal declaration that Plaintiffs submitted to defend the admissibility of Dr. Bendick's expert report. Defendants initially argue that it is an unsworn hearsay statement based on 28 U.S.C. § 1746, which requires statements prepared outside of the United States to be sworn "under the laws of the United States of America." While Dr. Bendick's original rebuttal declaration did not comply with this requirement, he has since re-executed his rebuttal declaration in accordance with the statute, so this issue is moot.
Second, Defendants argue that portions of the rebuttal declaration offer "wholly new expert opinions" that are not admissible under Rule 26. Butler v. Sears Roebuck & Co. ("Butler I"), No. 06 C 7023, 2010 WL 2697601, at *1 (N.D. Ill. July 7, 2010). Rule 26(a)(2)(D) requires parties to disclose expert witness testimony "at the times and in the sequence that the court orders." A party may submit a rebuttal report "intended solely to contradict or rebut evidence on the same subject identified by another party." Fed. R. Civ. P. 26(a)(2)(D)(ii). Additionally, Rule 26(e) requires parties to supplement their disclosures if they learn that the initial disclosures were materially "incomplete or incorrect." However, "[a] rebuttal expert report 'cannot be used to advance new arguments or new evidence to support plaintiff's expert's initial opinions.'" Lowe v. CVS Pharmacy, Inc., No. 14 C 3687, 2017 WL 2152385, at *2 (N.D. Ill. May 17, 2017) (citation omitted). "The proper function of rebuttal evidence is to contradict, impeach or defuse the impact of the evidence offered by an adverse party." Peals v. Terre Haute Police Dep't, 535 F.3d 621, 630 (7th Cir. 2008) (citation omitted); see also Stanfield v. Dart, No. 10 C 6569, 2013 WL 589222, at *3 (N.D. Ill. Feb. 14, 2013) ("A party may not offer testimony under the guise of 'rebuttal' only to provide additional support for his case in chief." (citing Peals, 535 F.3d at 630)).
The contested portions of Dr. Bendick's rebuttal declaration, detailing his revised calculations according to Dr. Amidon's critiques, could arguably fall on either side of the case-in-chief/rebuttal dichotomy. As Defendants point out, Dr. Bendick opined that his estimates in the rebuttal declaration are less accurate than the estimates in the Revised Declaration. He testified that they are nonetheless useful to "demonstrate how, if Dr. Amidon's suggestions are followed, they do or do not change the findings and conclusions" in his expert report. Doc. 714-6 at 13. Thus, the Court primarily considers Dr. Bendick's rebuttal declaration as an attempt to "defuse the impact" of Dr. Amidon's critiques, Peals, 535 F.3d at 630, rather than as "wholly new opinions" to bolster class certification, Butler I, 2010 WL 2697601, at *1.
Defendants correctly note that "point[ing] out an ostensible flaw" in Bendick's expert analysis does not trigger a duty to correct the expert report under Rule 26(e). See Berkheimer v. Hewlett-Packard Co., No. 12 C 9023, 2016 WL 3030170, at *5 (N.D. Ill. May 25, 2016) ("Rule 26(e) is not a back door that can be used to 'supplement' expert reports ad infinitum[.]"). But Defendants' reliance on Rule 26(e) is misplaced because it does not limit the Court's ability to consider supplemental disclosures in its discretion. Talbert v. City of Chicago, 236 F.R.D. 415, 421 (N.D. Ill. 2006) ("The Rule, by its plain terms, imposes duties on the parties; it does not purport to limit a court's discretion under Rule 26(a)(2)(C) to direct the 'times and . . . the sequence' in which expert reports should be filed."). "So long as [the supplemental disclosure] does not entail some greater harm to the opponent of the report, sound discretion would seem to counsel allowing the supplemental report to be filed." Id. Here, not only did Defendants have an opportunity to respond to Dr. Bendick's rebuttal declaration, Dr. Amidon filed a rebuttal declaration of her own on September 6, 2018.
Additionally, one of Defendants' main contentions for striking Dr. Bendick's report, discussed further below, is that he unjustifiably relied on the List of 18,191 Referrals as part of his analysis, rendering his conclusions hopelessly flawed. Defendants have had a copy of the List of 18,191 Referrals since 2014, when Plaintiffs first asked Defendants to verify its accuracy. Defendants refused, even though they could have addressed many of Dr. Amidon's concerns at that time. Their newfound critiques are the primary reason Dr. Bendick submitted a rebuttal declaration in the first place. If Defendants want to hide the ball for several years on such a central issue, the Court will not punish Plaintiffs for submitting a belated rebuttal report.
D. Reliability of Estimated Expected Representation
Defendants argue that Dr. Bendick's expected representation statistics are unreliable because Dr. Bendick 1) failed to properly vet the accuracy and completeness of data upon which he relied, 2) improperly relied on an assistant to make calculations that cannot be replicated, 3) excluded almost half of the relevant geographic area in his calculations, and 4) failed to report margins of error. The Court considers these arguments in turn.
Defendants' first contention is that Dr. Bendick failed to verify the completeness or the significance of the data upon which he relied to calculate the expected representation figure. Defendants argue that the first misstep was arbitrarily excluding 3,498 alternate addresses for some of the laborers in the List of 18,191 Referrals. Plaintiffs respond that this was proper because including several addresses for a single employee places more weight on some individuals than others, making them "twice as important." Doc. 714-6 ¶ 8. Plaintiffs argue this is particularly appropriate for some addresses that are obviously outside of MVP's reasonable recruitment area, such as addresses from California and Florida.
Dr. Amidon contends there are 3,552 alternate addresses that Dr. Bendick excluded from the List of 18,191 Referrals, but it is unclear from the parties' briefs which number is accurate.
The Court understands that Dr. Bendick disregarded addresses from other states because they fall outside the reasonable recruitment area of an employer offering minimum-wage, temporary jobs. But, with respect to addresses in the Chicagoland area, Plaintiffs offer little more than an axiomatic assurance that double-counting is wrong. As Dr. Amidon argues, the expected representation estimate begins by determining the reasonable recruitment area according to the largest PUMAs that account for 95% of referrals. If an employee came to MVP from one address one year and returned to work for MVP using a different address in another year, it is plausible that both addresses are still relevant for determining the reasonable recruitment area.
Dr. Bendick contends that it is nonetheless standard practice to rely on the first address listed in employer records because this is the address where the employer can most likely reach the employee. Relying on the first address listed might be appropriate when interpreting the employer's original records. Dr. Amidon concedes that "[i]t is reasonable practice among labor economists and other employment analysts to draw certain inferences from business records organized by an employer in a human resources information system." Doc. 653 ¶ 7. But that is not what happened here. Instead, Plaintiffs' counsel compiled the List of 18,191 Referrals from the Master Address List of over 64,000 referrals, so Dr. Bendick did not rely on primary source material. In this situation, it is not clear that the "alternate" 3,498 addresses are any more or less relevant than the addresses Dr. Bendick included in his analysis.
The parties also dispute how Plaintiffs' counsel compiled the List of 18,191 Referrals, and Defendants critique Dr. Bendick for failing to consider whether he should have included an additional 4,862 addresses in the Master Address List as a part of his analysis. Dr. Bendick points out that these addresses likely relate to work referrals from prior to 2009 and it would "not be standard professional practice among economists and other employment analysts to rely on these [] records of unclear origin and relevance." Doc. 714-6 ¶ 11. But Plaintiffs have not satisfactorily explained how they culled the Master Address List to ensure the List of 18,191 Referrals only included relevant addresses.
In her first report Dr. Amidon contends there were an additional 5,710 addresses that Plaintiffs' counsel excluded, but it is unclear from the parties' briefs which number is accurate.
Plaintiffs' counsel disputes that the Master Address List was formatted as a CSV file and contends that it contained an "Address" and "Alternate Address" field, from which he could easily extract the primary address. See Doc. 644-1, Williams Decl. ¶ 24. Plaintiffs returned the original disc containing the Excel files to Defendants in the midst of the case. Although Defendants re-produced this disc, which contained CSV formatted files, Plaintiffs still maintain that the original files were not in CSV format. In short, the Court cannot come to any definitive conclusion on how Plaintiffs culled the Master Address List to create the List of 18,191 Referrals.
Without knowing how the List of 18,191 Referrals was compiled, Defendants make a persuasive argument that Dr. Bendick's ultimate conclusion is not reliable. "Reliability [under Daubert], however, is primarily a question of the validity of the methodology employed by an expert, not the quality of the data used in applying the methodology or the conclusions produced." Manpower, Inc. v. Ins. Co. of Pa., 732 F.3d 796, 806 (7th Cir. 2013); Smith v. Ford Motor Co., 215 F.3d 713, 718 (7th Cir. 2000) ("[W]e emphasize that the court's gatekeeping function focuses on an examination of the expert's methodology."). Some caselaw supports the notion that courts can use questionable sources of data as grounds to exclude an expert's opinion. See Macy v. Whirlpool Corp., 613 F. App'x 340, 344 (5th Cir. 2015) ("[A]n expert's testimony must be reliable at every step, including the methodology employed, the facts underlying the expert's opinion, and the link between the facts and the conclusion."); Sommerfield v. City of Chicago, 254 F.R.D. 317, 324 (N.D. Ill. 2008) ("[N]otwithstanding the wide latitude accorded experts in choosing the sources on which to base opinions, those sources must be shown to be reliable."). As Defendants argue, the subset of addresses Dr. Bendick relied on from the List of 18,191 Referrals is the cornerstone of Dr. Bendick's expected representation estimates and his failure to investigate how the list came about gives the Court some concern about the accuracy of his conclusions. See Forte v. Liquidnet Holdings, Inc., 675 F. App'x 21, 24 (2d Cir. 2017) ("A failure to validate data by itself can constitute grounds for excluding an expert report."); Munoz v. Orr, 200 F.3d 291, 301-02 (5th Cir. 2000) ("Dr. Benz relied on the plaintiffs' compilations of data, which gives rise to a 'common-sense skepticism' regarding the expert's evaluation, and did not seek to verify the information presented to him." (citations omitted)); In re Testosterone Replacement Therapy Prod. Liab. Litig., No. 14 C 1748, 2018 WL 3586182, at *12 (N.D. Ill. July 26, 2018) (expert who relied on summaries of pertinent information compiled by plaintiff's attorney did not adequately investigate evidence to render expert opinion). But the Seventh Circuit has held that an accounting expert can rely on information furnished by the plaintiff's attorney to render an expert opinion on the value of lost future earnings. Tuf Racing Prod., Inc. v. Am. Suzuki Motor Corp., 223 F.3d 585, 591 (7th Cir. 2000). As the Seventh Circuit commented in Manpower, "[t]hat the expert accountant in Tuf could opine on future earnings on the basis of information supplied by counsel should make clear that the reliability of the data itself is not the object of the Daubert inquiry. The reliability of data and assumptions used in applying a methodology is tested by the adversarial process and determined by the jury; the court's role is generally limited to assessing the reliability of the methodology—the framework—of the expert's analysis." 732 F.3d at 808. Here, Defendants do not contest the methodology, or perhaps more accurately the framework, of Dr. Bendick's analysis. Instead, Defendants contest that Dr. Bendick failed to properly vet the original data files on which he relied, rendering his conclusions faulty because the data is of questionable relevance. But the "Daubert inquiry is not designed to have the district judge take the place of the jury to decide ultimate issues of credibility and accuracy," Lapsley v. Xtek, Inc., 689 F.3d 802, 805 (7th Cir. 2012), including the "soundness of the factual underpinnings" of the expert opinion, Ford Motor Co., 215 F.3d at 718. Defendants raise valid criticisms of the data on which Dr. Bendick relied. These critiques go to the weight of Dr. Bendick's analysis, not its admissibility. Manpower, 732 F.3d at 807-08.
Defendants' remaining arguments regarding the expected representation figure similarly go to the weight of Dr. Bendick's testimony. Defendants point out that Dr. Bendick relied on a consultant, Warren Glimpse, to do coding work and calculate the estimated demographic composition of people who were likely to be qualified, available, and interested in MVP referrals within the relevant PUMAs. Part of the reason Dr. Bendick worked with Glimpse is because he is proficient in a computer code in which Dr. Bendick does not write. After producing the same code to Defendants, however, Dr. Amidon could not replicate all of Glimpse's figures, and during his deposition Dr. Bendick could not explain what caused the discrepancy. Defendants argue that Dr. Bendick's testimony is therefore inadmissible. See Zenith Elecs. Corp. v. WH-TV Broad. Corp., 395 F.3d 416, 419 (7th Cir. 2005) ("Someone else using the same data and methods must be able to replicate the result.").
Experts can rely on assistants to formulate an opinion, Dura Auto. Sys. of Ind., Inc. v. CTS Corp., 285 F.3d 609, 612 (7th Cir. 2002), and can rely on an assistant's work within a certain discipline without being an expert in that discipline themselves, Walker v. Soo Line R. R. Co., 208 F.3d 581, 589 (7th Cir. 2000) ("Nor do we believe that the leader of a clinical medical team must be qualified as an expert in every individual discipline encompassed by the team in order to testify as to the team's conclusions."). The "[a]nalysis becomes more complicated if the assistants aren't merely gofers or data gatherers but exercise professional judgment that is beyond the expert's ken." Dura, 285 F.3d at 613. In other words, an expert cannot serve as the mouthpiece for another expert. Id. at 614. Here, however, Glimpse performed geocoding work under Dr. Bendick's supervision and direction—work that other courts have found to involve straightforward calculations as opposed to "a host of discretionary expert judgments." Id. at 615; see also EEOC v. DHL Express (USA), Inc., No. 10 C 6139, 2016 WL 5796890, at *5 (N.D. Ill. Sept. 30, 2016) (finding assistant's geocoding work "d[id] not raise the concerns voiced by the court in Dura"). Further, the fact that Dr. Amidon could not replicate precisely the same estimate of African Americans expected to be interested in MVP referrals does not mean Dr. Bendick's testimony is inadmissible. Defendants rely on Zenith, which involved an expert who, as the Seventh Circuit put it, relied on his "expert intuition" to forecast a company's hypothetical projected growth. 395 F.3d at 418-19. Intuition is not a method that can be tested. Id. at 419. Here, by contrast, Defendants do not really dispute that geocoding is an acceptable methodology that can be tested, only that they reached a slightly different result after crunching the numbers. Dr. Bendick, for example, estimated there to be 1,606 NHAA from PUMA number 03408 that were interested in MVP referrals, while Dr. Amidon estimated the number to be 1,667. This could be due to any number of variables with which an expert could tinker. More importantly, Plaintiffs disclosed the code that Glimpse used, so Defendants have the tools they need to try to verify or falsify Glimpse's figures. The fact that they reached marginally different estimates—differences that, as Plaintiffs point out, make no significant impact on the final analysis—is not a reason to exclude Dr. Bendick's testimony. See Ford Motor Co., 215 F.3d at 718 ("The soundness of the factual underpinnings of the expert's analysis and the correctness of the expert's conclusions based on that analysis are factual matters to be determined by the trier of fact[.]").
Next, Defendants argue that Dr. Bendick arbitrarily excluded 869 addresses, located in thirty additional PUMAs, from his calculations. But determining a reasonable recruitment area necessarily means that Dr. Bendick must exclude some PUMAs. Economists determine an employer's reasonable recruitment area according to the area the employer "usually seeks or reasonably could seek workers to fill the position in question." Doc. 714-6 ¶ 12 (citing requirements for affirmative action programs under 41 C.F.R § 60-2.14, which requires federal contractors to separately determine the availability of qualified minorities and women available for employment "to establish a benchmark against which the demographic composition of the contractor's incumbent workforce can be compared"). Dr. Bendick included the PUMAs that accounted for over 95% of the referrals. Including the PUMAs that account for less than 5% of referrals would do little to define the area from which an employer usually draws workers.
Dr. Bendick also states that he excluded high-income areas, such as Chicago's Gold Coast, because few workers interested in low-paid jobs with MVP were likely to reside in these areas. But if there were significant referrals from these areas, that would seem to undercut Dr. Bendick's assumption that areas like the Gold Coast are exclusively high-income neighborhoods with few workers interested in low-paid jobs. But, as best as the Court can tell, there were not a substantial number of home addresses from this area.
Finally, Defendants criticize Dr. Bendick for failing to report a margin of error for each PUMA included in his analysis. Defendants argue that since the ACS is based on a sample of the population within each PUMA, there is necessarily an error rate attached to each PUMA. But the margin of error for each separate PUMA is not particularly relevant because the estimated expected representation figure is based on all of the PUMAs combined. Defendants do not explain how the margin of error for each individual PUMA would assist a "judge to evaluate how potential error . . . may have affected the obtained pattern of responses." Doc. 601 at 23 (citing Shari Seidman Diamond, Reference Manual on Scientific Evidence 362 (Nat'l Acads. Press, 3rd ed. 2011).
In summary, "[t]he critical inquiry is whether there is a connection between the data employed and the opinion offered; it is the opinion connected to existing data 'only by the ipse dixit of the expert,' that is properly excluded under Rule 702." Manpower, 732 F.3d at 806 (citation omitted). Here, there is more than a connection between the data and Dr. Bendick's estimated representation figure. Critiques regarding the quality of the data on which he relied do not render his opinion inadmissible.
E. Reliability of Estimated Actual Representation
Defendants argue that Dr. Bendick's estimated actual representation figures are unreliable because Dr. Bendick 1) could not identify the source for the data in the List of 11,255 Referrals, 2) underestimated the proportion of NHAA among GSB referrals, 3) failed to estimate the race for more than a quarter of GSB referrals, and 4) relied on two different sources of data from the Census Bureau to estimate the expected representation figure and the actual representation figures, respectively.
First, Defendants point out that Dr. Bendick is unable to explain where he obtained the addresses that he used to compile the List of 11,255 Referrals and therefore Dr. Amidon cannot recreate a crucial component of his calculations. Similar to Defendants' arguments regarding the List of 18,191 Referrals, these concerns go to the weight, rather than the admissibility of Dr. Bendick's testimony. Manpower, 732 F.3d at 807-08. Significantly, Defendants do not allege that Dr. Bendick paired names to the wrong addresses. Whatever the source of the addresses, they must have originated in Defendants' materials. Defendants ask for too much when they demand step-by-step instructions on how to replicate Dr. Bendick's calculations. Studiengesellschaft Kohle, m.b.H. v. Dart Indus., Inc., 862 F.2d 1564, 1580 (Fed. Cir. 1988) ("[T]he opinion testimony of expert witnesses need not contain step-by-step calculations."). The more important consideration is whether Defendants have the tools necessary to test Dr. Bendick's results, based on the data and methods disclosed. Zenith, 395 F.3d at 419 ("An expert must offer good reason to think that his approach produces an accurate estimate using professional methods, and this estimate must be testable."). With respect to the List of 11,255 Referrals, they do.
Second, Defendants argue that Dr. Bendick underestimated the proportion of NHAA referrals because he incorrectly assigned referrals without an address a 0% probability of being NHAA. During his deposition, Dr. Bendick testified that he had intended to exclude referrals without an address from his calculation, rather than including them. Then, in his rebuttal declaration, Dr. Bendick testified that it is "unclear whether including or excluding zeros when computing the representation of [NHAA] increases or decreases the accuracy and reliability of the estimate." Doc. 714-6. His reasoning was that discriminatory hiring preferences might have caused MVP to record less addresses for NHAA, in which case excluding these referrals would overestimate NHAA representation. The problem with this reasoning is that it places the cart before the horse. Dr. Bendick's analysis is offered to help demonstrate that MVP and GSB colluded in a racially-discriminatory referral practice. But assuming at the outset that the Defendants discriminated against Plaintiffs and controlling the variables as if that were true would undermine the usefulness of the statistics. EEOC v. Sears, Roebuck & Co., 628 F. Supp. 1264, 1286 (N.D. Ill. 1986) ("[T]he assumptions made by a statistician in formulating a model can be far more important than the numerical complexities and results of the analysis. Without a sound theoretical basis, which is carefully reasoned and closely tailored to the factual circumstances of the case, the statistical results can be meaningless."), aff'd, 839 F.2d 302 (7th Cir. 1988). Ultimately, Dr. Bendick's reason for including referrals without an address is not "too complex for the jury" to understand. Stollings v. Ryobi Techs., Inc., 725 F.3d 753, 766 (7th Cir. 2013). Thus, this too goes to the weight, rather than the admissibility of his analysis. Id. ("The judge should permit the jury to weigh the strength of the expert's conclusions, provided [the] shortcomings are within the realm of a lay juror's understanding."). Dr. Amidon, however, ran the numbers excluding referrals without an address and Dr. Bendick incorporated these figures into his rebuttal declaration. For the reasons stated above, the Court finds these estimates of the actual representation rate to be more informative than the figures in Dr. Bendick's Revised Declaration. Consistent with Dr. Bendick's preference to opt for the more conservative of two plausible estimates, the Court places more weight on the estimated actual representation figures in Dr. Bendick's rebuttal declaration in considering the motion for certification.
Dr. Amidon does not endorse these figures as accurate estimates of actual representation. She calculated the figures to demonstrate that excluding referrals without an address (and therefore ones that did not have an assigned probability of being NHAA) increases the actual representation estimates.
Third, Defendants argue that Dr. Bendick failed to assign addresses to more than a quarter of GSB referrals, rendering his calculations inherently unreliable because he fails to consider that referrals without an address had a greater chance of being NHAA than the referrals with addresses. In other words, the referrals with an address might not be representative of the entire population of people referred to GSB during any single year. Further, Defendants argue the 72.4% figure is misleading because Dr. Bendick assigned a probability to roughly half of the referrals in 2009 to 2010—two years that had the largest shortfalls. He assigned a probability to more than 80% of referrals in 2014 to 2015—two years that had, according to Defendants, likely no shortfall in NHAA representation. As Dr. Bendick explains, the primary reason he did not have addresses for some of the referrals is that Defendants provided incomplete data sets. For the remaining referrals Dr. Bendick either could not place the address within a census block or the Census Bureau did not have useful information for that particular block. Defendants essentially blame Dr. Bendick for failing to investigate why the records they provided him were of such poor quality. Again, this goes to the weight afforded Dr. Bendick's analysis, not its admissibility. Walker, 208 F.3d at 586-87.
Dr. Bendick states that he assigned a probability to 73.5% of referrals across all years to calculate the estimated actual representation figures, but Defendants argue that he only assigned a probability to 72.4% of the referrals.
Defendants also argue that Dr. Bendick's calculation of the shortfall in hours worked is unreliable because Dr. Bendick relied on incomplete data furnished by Defendants. Specifically, Defendants point out that the Yearly GSB Hours Files contained names and hours for referrals that did not correspond neatly with data in the Yearly Cicero Address Files. Some of the employees also had, as Defendants call it, "suspect hours," Doc. 601 at 28, including some that had accumulated more than 2,000 hours in a single year. For a temporary position that was supposed to last no more than ninety days, Defendants point out that hours of this magnitude tend to reflect a full-time employee instead of a temporary laborer. Defendants argue that these issues "warranted further investigation, not blind reliance." Doc. 601 at 28. For the same reasons outlined above, this goes to the weight given Dr. Bendick's testimony because it deals with the "factual underpinnings" of his analysis. Ford Motor Co., 215 F.3d at 718. Additionally, Dr. Bendick's shortfall analysis with respect to hours worked is, in the Court's view, much less significant at the class certification stage than the expected representation and actual representation figures. The shortfall in hours is primarily relevant to determining damages.
Although MVP and GSB had an agreement that it would not employ laborers as temporary assignments for more than ninety days, they may have ignored this rule. Former class representative Hunt, for instance, testified that he worked more than ninety days at GSB without being offered a permanent position. See Doc. 700-1 at 185-86. It seems this is at least one plausible explanation for why MVP kept records of these hours, so it is not obvious that Dr. Bendick should have excluded the numbers from his analysis.
Finally, Defendants argue that Dr. Bendick's testimony is unreliable because he used demographic information from the ACS to estimate his expected representation figure but relied on the decennial census to estimate the actual representation rate. Defendants argue that this renders his calculations unreliable according to "accepted scientific standards, including those described by the Bureau responsible for publishing both data sets." Doc. 601 at 29. This is what the Census Bureau states: "Use caution in comparing ACS data with data from the decennial census or other sources. Every survey uses different methods, which could affect the comparability of the numbers." U.S. Census Bureau, A Compass for Understanding and Using American Community Survey Data: What General Data Users Need to Know, p. 25, U.S. Government Printing Office, Washington, DC, 2008. This cautionary instruction does not mean that a comparison between data from one source and data from the other necessarily departs from "accepted scientific standards." Doc. 601 at 29. The Court ventures to guess that Dr. Bendick used the ACS data to estimate the expected representation figure because it provides data relevant for determining the population likely to seek work at MVP, such as those who are part of the labor force and who earn less than the median salary for the Chicago metropolitan area. Census blocks, by contrast, consist of smaller geographical units that provide a better predictor that a person with a home address located within the census block is of a certain race, which is the only Census Bureau data relevant to Dr. Bendick's actual representation estimate. Defendants offer no substantive argument for why Dr. Bendick could not rely on both the ACS and the decennial census in his analysis.
In sum, Dr. Bendick's estimates for both the expected and actual representation figures are sufficiently reliable for the Court to consider as part of Plaintiffs' motion for class certification. Defendants' critiques go to the weight that should be afforded his testimony, not its admissibility.
II. Class Certification
Plaintiffs seek to certify the following class:
African-American laborers who sought work assignments through MVP and were otherwise eligible to work at GSB, but on one or more occasion were not assigned to work at GSB when a position for which they were otherwise qualified was available during the
period of four years prior to the filing of the Plaintiffs' Original Complaint for the § 1981 claims up through and including December 31, 2014.Doc. 700 at 38.
A. Legal Standard
Class certification is appropriate where a plaintiff can meet the four requirements of Rule 23(a)—numerosity, commonality, typicality, and adequacy of representation. Fed. R. Civ. P. 23(a). Additionally, a plaintiff must also satisfy one of the three subsections of Rule 23(b). Fed. R. Civ. P. 23(b); Oshana v. Coca-Cola Co., 472 F.3d 506, 513 (7th Cir. 2006). Here, Plaintiffs seek certification under Rule 23(b)(3), which requires a finding that "questions of law or fact common to class members predominate over any questions affecting only individual members, and that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy." Fed. R. Civ. P. 23(b)(3). Finally, although not an explicit requirement of Rule 23, the party seeking certification must demonstrate that the class members are identifiable. Oshana, 472 F.3d at 513.
The Court has broad discretion in determining whether to certify a proposed class. Keele v. Wexler, 149 F.3d 589, 592 (7th Cir. 1998). The party seeking certification bears the burden of demonstrating that certification is proper by a preponderance of the evidence. Messner v. Northshore Univ. HealthSystem, 669 F.3d 802, 811 (7th Cir. 2012). The Court must engage in a "rigorous analysis," resolving material factual disputes where necessary. Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 350-51 (2011); Szabo v. Bridgeport Machs., Inc., 249 F.3d 672, 676 (7th Cir. 2001). But "[i]n conducting [the Rule 23] analysis, the court should not turn the class certification proceedings into a dress rehearsal for the trial on the merits." Messner, 669 F.3d at 811; Amgen Inc. v. Conn. Ret. Plans & Tr. Funds, 568 U.S. 455, 465-66 (2013) (merits questions are to be considered only to the extent relevant to determining if Rule 23's prerequisites are met).
B. Numerosity
Defendants do not contest that Plaintiffs miss the numerosity requirement in this case. According to Dr. Bendick's analysis, there was an approximate shortfall of anywhere from 34 to 668 NHAA laborers each year from 2009 to 2014. The shortfall each year is nearly enough to constitute a class in itself. Phillips v. Waukegan Hous. Auth., 331 F.R.D. 341, 350 (N.D. Ill. 2019) ("While there is no magic number, forty or more members is generally considered to be sufficient to satisfy the numerosity requirement."). Thus, Plaintiffs have met the numerosity requirement.
These numbers come from Dr. Bendick's Revised Declaration. Unfortunately, neither party includes a chart comparing the 29.5% expected representation figure against the actual representation figures that exclude referrals without an address. This would, in the Court's opinion, produce the most informative estimates of the shortfall in representation. Here, the differences are insignificant because Defendants do not contest the numerosity requirement.
C. Commonality
Rule 23(a)(2) provides that there must be "questions of law or fact common to the class." "Commonality requires the plaintiff to demonstrate that the class members 'have suffered the same injury.'" Wal-Mart, 564 U.S. at 349-50 (citing Gen. Tel. Co. of Sw. v. Falcon, 457 U.S. 147, 157 (1982)). There must be a "common contention" that 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. at 350. "[S]uperficial common questions—like . . . whether each class member 'suffered a violation of the same provision of law'—are not enough." Jamie S. v. Milwaukee Pub. Sch., 668 F.3d 481, 497 (7th Cir. 2012) (quoting Wal-Mart, 564 U.S. at 350).
Whether Plaintiffs can proceed on their claims on a classwide basis depends on whether "examination of all the class members' claims for relief will produce a common answer to the crucial question why was I disfavored." Wal-Mart, 564 U.S. at 352. The existence of an illegal policy may provide the "glue" to hold together class members' claims in answering this question. See Jamie S., 668 F.3d at 498 (quoting Wal-Mart, 564 U.S. at 352). Here, Plaintiffs contend that Defendants instituted such illegal policies by systematically instructing on-site supervisors and dispatchers not to refer African Americans to GSB.
In Falcon, the Supreme Court suggested that parties can show commonality in a pattern or practice case in one of two ways: 1) through proof that the employer used a biased company-wide testing procedure to evaluate job applicants or current employees, or 2) through significant proof of a general policy of discrimination that manifests itself in the same general fashion throughout the company. Falcon, 457 U.S. at 159 n.15. Subsequent cases, including Wal-Mart, have found commonality lacking where the allegedly discriminatory policy is highly discretionary and the plaintiffs do not identify a common way in which the defendants exercise that discretion. See Wal-Mart, 564 U.S. at 355-56; Bolden v. Walsh Constr. Co., 688 F.3d 893, 896-98 (7th Cir. 2012) (reversing class certification where alleged discrimination resulted from acts of individual supervisors exercising independent discretion). But commonality will exist where allegedly discriminatory general policies are enforced at the corporate level rather than by individual supervisors, even where there is some discretion in the policies' execution. See McReynolds v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 672 F.3d 482, 488-91 (7th Cir. 2012) (allowing class certification for disparate impact claim challenging company-wide practices that local managers had to follow), abrogated on other grounds by Phillips v. Sheriff of Cook Cty., 828 F.3d 541 (7th Cir. 2016); Scott v. Family Dollar Stores, Inc., 733 F.3d 105, 114 (4th Cir. 2013) ("[E]ven in cases where the complaint alleges discretion, if there is also an allegation of a company-wide policy of discrimination, the putative class may still satisfy the commonality requirement for certification."). As the Seventh Circuit has stated, "a company-wide practice is appropriate for class challenge even where some decisions in the chain of acts challenged as discriminatory can be exercised by local managers with discretion—at least where the class at issue is affected in a common manner, such as where there is a uniform policy or process applied to all." Chicago Teachers Union, Local No. 1 v. Bd. of Educ. of City of Chicago, 797 F.3d 426, 437 (7th Cir. 2015).
Defendants argue that Plaintiffs have not provided significant proof of a general policy of discrimination. They argue that MVP supervisors and dispatchers had individual discretion to hire and refer workers based on a number of factors, including their prior experience, willingness to abide by GSB's GMPs, or availability to work shifts that were in demand. See Jones v. GES Exposition Servs., Inc., No. 02 C 6243, 2004 WL 2011396, at *9 (N.D. Ill. Sept. 7, 2004) ("[A] decentralized hiring procedure, which allows decisionmakers to consider subjective factors, may support individual claims of discrimination but cuts against the assertion that an employer engages in a pattern or practice of discriminatory hiring as a standard operating procedure." (internal quotation marks omitted) (citation omitted)). Defendants contend that the statistical and anecdotal evidence does nothing to overcome the inherently individualized nature of Plaintiffs' claims.
The Court disagrees. First, Dr. Bendick's analysis provides persuasive evidence that disparities in hiring can be explained by a general policy of discrimination, rather than individualized processes. See Int'l Bhd. of Teamsters v. United States, 431 U.S. 324, 339 n.20 (1977) ("Statistics showing racial or ethnic imbalance are probative . . . because such imbalance is often a telltale sign of purposeful discrimination; absent explanation, it is ordinarily to be expected that nondiscriminatory hiring practices will in time result in a work force more or less representative of the racial and ethnic composition of the population in the community from which employees are hired."). Neither party includes a chart comparing the 29.5% expected representation figure against the actual representation figures from Dr. Bendick's rebuttal declaration, which the Court would find more probative than the actual representation figure in Dr. Bendick's Revised Declaration. But even if the Court were to simply refer to the rebuttal declaration, which incorporates an even lower expected representation figure, there are significant disparities in the representation of NHAA from 2009 to 2013, the same year this lawsuit was initiated. According to Dr. Bendick, the shortfalls are all statistically significant at nine standard deviations or more. Statisticians generally consider discrepancies of two or more standard deviations statistically significant. Perez v. City of Batavia, No. 98 C 8226, 2004 WL 2967153, at *10 (N.D. Ill. Nov. 23, 2004) ("Generally, statisticians believe that two standard deviations is enough to show that the result is unlikely (less than a 5% probability) to be the result of chance[.]"). These are stark figures that suggest a common answer to the central question: why were African American laborers disfavored? Wal-Mart, 564 U.S. at 352.
The cases Defendants point to demonstrate that Dr. Bendick's analysis actually helps substantiate Plaintiffs' common claim. Bennett v. Roberts involved a minority job applicant who claimed that a school district discriminated against her and similarly situated applicants because of their race. No. 96 C 6917, 2000 WL 781868, at *3 (N.D. Ill. June 15, 2000). Her claim relied in part on an expert's statistical analysis that concluded "race had to be a factor in hiring because so few minorities were employed by the District." Id. at *5. The parallel here would be if Dr. Bendick only estimated the number of African Americans referred to GSB with nothing else and then concluded that an arbitrarily small number of referrals was proof of discrimination. But Dr. Bendick calculated the expected representation figure to provide context for the actual representation figures, making Bennett inapposite to the facts in this case. See also 41 C.F.R § 60-2.14(b) (requiring federal contractors establishing affirmative action programs to "separately determine the availability of [qualified] minorities or women for each job group").
Defendants also point to Betts v. Sundstrand Corp., where the plaintiffs offered statistics showing that the population in Rockford, Illinois, was 15.1% African American and claimed discrimination because the defendant corporation's workforce was only 3.5% African American. No. 97 C 50188, 1999 WL 436579, at *3 (N.D. Ill. June 21, 1999). Although one step further along than the statistics offered in Bennett, the court similarly found that "these statistics do not paint the entire picture." Id. at *4. Because the plaintiffs' claims involved hiring practices, "the applicant pool, including their qualifications, race and interest in the jobs at issue, [wa]s key in determining the relevant statistics." Id. The plaintiffs in Betts had not controlled for any of these factors. They had, for example, assumed that Rockford was the appropriate reasonable recruitment area for determining the expected representation of African Americans in the workforce. Id. But the court noted that the "defendant's recruitment efforts for professionals and managers are nationwide," and that "[e]ven with regards to entry-level positions, plaintiffs have not shown that it is proper to exclude areas close to Rockford, within Winnebago and Boone counties[.]" Id. Dr. Bendick, by contrast, first determined the reasonable recruitment area for GSB based on the home addresses of those MVP referred to GSB. EEOC v. O & G Spring & Wire Forms Specialty Co., 38 F.3d 872, 876-77 (7th Cir. 1994) ("Determining the relevant labor market is an essential step in determining whether there are any statistically significant deviations between the market and the employer's hiring patterns."). The resulting cluster of PUMAs around the Chicagoland area is borne out by the anecdotal evidence. Many positions at GSB were minimum wage jobs that would make a long commute too expensive to justify the cost and, as Defendants acknowledge, MVP recruited in areas close to the Cicero office and to the bakery because individuals in these areas would have shorter commutes to work. Dr. Bendick then calculated the expected representation based on the population that was qualified, available, and interested in MVP work assignments, something that the plaintiffs in Bennett and Betts failed to do. For the reasons already recounted, Dr. Bendick's analysis provides significant proof of a discriminatory practice.
Plaintiffs have also offered persuasive anecdotal evidence establishing a discriminatory policy. Green and Zollicoffer testified that dispatchers gave Latino laborers assignments the moment they walked in the door, even if they arrived after African Americans workers who were there early and waiting for assignments. Former MVP employees testified that this did not happen randomly. Carretero, Ceja, Corral, and Sanchez all testified that dispatchers and supervisors regularly steered African Americans away from assignments by telling them to come back the next day, falsely telling them GSB was on a hiring freeze, or crossing out their names on assignment lists based off of their surnames. For Latino laborers, by contrast, dispatchers immediately gave them an assignment or began calling around to see if clients had open positions. Carretero testified that she heard managers calling African Americans niggers, and "lazy, good-for-nothing 'negros.'" Doc. 700-1 at 136. Ceja testified that she was "yelled at by representatives of client companies" for assigning them African Americans and that MVP's president warned her to comply because he did not want to lose the account. Id. at 153. Sanchez testified that Salgado, GSB's plant manager, told her he only wanted Latino workers.
The fact that MVP utilized a decentralized decision-making process and MVP supervisors and dispatchers had a degree of autonomy to make referral decisions does not preclude Plaintiffs' claim because they challenge Defendants' overarching policy against hiring African Americans. Similar to McReynolds, the maintenance of a mandatory, top-down policy that causes racial disparities is sufficient to establish a common question. 672 F.3d at 489-91 (explaining that if company-wide policies increased the amount of discrimination by local decisionmakers, "[t]he incremental causal effect . . . of those company-wide policies—which is the alleged disparate impact—could be most efficiently determined on a class-wide basis"); see also Beley v. City of Chicago, No. 12 C 9714, 2015 WL 8153377, at *4 (N.D. Ill. Dec. 7, 2015) (finding commonality requirement met where the plaintiffs challenged the Chicago Police Department's alleged policy of refusing to register homeless sex offenders, finding that variances in interactions between individual police officers and homeless sex offenders did not defeat commonality because the "purported policy nonetheless shaped those interactions"). As the Seventh Circuit found in Chicago Teachers Union, it "is more efficient to answer the question 'did these early discriminatory processes have a disparate impact on race' just one time rather than over and over again in multiple separate lawsuits." 797 F.3d at 436 & n.5 (finding that if objective criteria in first steps of turnaround process "narrowed the pool in such a way as to have a disparate impact on African-American teachers," the plaintiffs had identified "the glue that binds the claims together without regard to the later, subjective step," meaning that a "discriminatory step in a chain of events" could provide a common question and affect the "ultimate outcome").
The cases Defendants rely on are inapplicable for these same reasons. In Jones, the plaintiffs did not assert a discriminatory policy of general application, only that various foreman were predominantly white and gave preference to other white carpenters in training and hiring opportunities. 2004 WL 2011396, at *9. That is as if the Plaintiffs in this case had only alleged that the supervisors and dispatchers were all Latino and therefore favored Latino job applicants. In Stubbs v. McDonald's Corp., the plaintiffs sought to certify a class that would have included all African American employees of McDonald's restaurants, that were owned or operated by McDonald's Corporation, since 1996. 224 F.R.D. 668, 675 (D. Kan. 2004). The plaintiffs made no allegations of a company-wide policy against promoting or hiring African Americans, and they conceded that all decisions were decentralized and supervisors at different locations were given discretion to make promotions. Id. The court found that such allegations involved "myriad individual considerations" that would not be appropriate for classwide adjudication. Id. Here, by contrast, Plaintiffs have alleged that managers, supervisors, and dispatchers all adhered to one policy: no African Americans.
Defendants also make much of the fact that several of the former MVP employees only provide testimony regarding events in 2014 and 2015—years for which "Plaintiffs' own expert found no statistically significant shortfall." Doc. 711 at 30, 37-38. To begin, Defendants misrepresent Dr. Bendick's findings. Dr. Bendick found a statistically significant shortfall in 2014. The alternative analysis in his rebuttal declaration did not find a shortfall in 2014, but Dr. Bendick never reneged on his initial estimate. Secondly, critiquing Dr. Bendick's analysis while simultaneously arguing that it disproves the anecdotal evidence strikes the Court as disingenuous. But at this stage, the Court does not need to determine whether Plaintiffs have presented credible testimony that will allow them to prevail at trial. "Rule 23 grants courts no license to engage in free-ranging merits inquiries at the certification stage. Merits questions may be considered to the extent—but only to the extent—that they are relevant to determining whether the Rule 23 prerequisites for class certification are satisfied." Amgen, 568 U.S. at 466. How many times Plaintiffs visited MVP, or whether they eventually received an assignment at GSB, are peripheral issues that do not diminish their core contention.
Defendants also argue that proposed class members who received an assignment from MVP do not have common claims against GSB because GSB did not interfere with these individuals' ability to form an at-will contract with MVP. This misconstrues the class definition. The class Plaintiffs seek to certify is composed of African Americans who sought work and "on one or more occasion were not assigned to work at GSB." Doc. 700 at 38. If a plaintiff can show that he was denied a referral to GSB, then he has a claim, regardless of whether he received an assignment to a different client. If a plaintiff cannot show that he was denied a referral to GSB, then he would fall outside of the proposed class definition.
In short, the statistical evidence and anecdotes together provide "significant proof" of a "uniform employment practice" that caused a common injury to the class. Wal-Mart, 564 U.S. at 353-55. Plaintiffs have met the commonality requirement.
D. Typicality
A plaintiff's claim is typical if it "arises from the same event or practice or course of conduct that gives rise to the claims of other class members and . . . her claims are based on the same legal theory." Oshana, 472 F.3d at 514 (citing Rosario v. Livaditis, 963 F.2d 1013, 1018 (7th Cir. 1992)). Some factual variations will not destroy typicality as long as the plaintiffs "have the same essential characteristics as the claims of the class at large." Id. (citation omitted). Typicality is determined with reference to a defendant's actions, not with respect to specific defenses a defendant may have against certain class members. Wagner v. NutraSweet Co., 95 F.3d 527, 534 (7th Cir. 1996).
Plaintiffs contend that they meet the typicality requirement because they each suffered the same injury as a result of Defendants' company-wide policies under the same circumstances. Namely, they each arrived at the Cicero office early, signed in, and waited unsuccessfully for work even though other laborers who arrived after them were put on work tickets. They allege that all class members were denied work on one or more occasion as a result of the allegedly discriminatory hiring policy.
Defendants argue that Plaintiffs' claims are not typical because of each Plaintiff's unique circumstances. They argue that Green stopped visiting the Cicero office after he was stranded during a ride home one night and Zollicoffer was unavailable to work for significant segments of the class period, either due to being incarcerated or due to working elsewhere. But just because each Plaintiff has unique circumstances—as the Court would expect of every putative class member—that does not make his claims atypical or inadequately aligned with those of the class. See Rosario, 963 F.2d at 1017 ("The fact that there is some factual variation among the class grievances will not defeat a class action."); De La Fuente v. Stokely-Van Camp, Inc., 713 F.2d 225, 232 (7th Cir. 1983) ("The typicality requirement may be satisfied even if there are factual distinctions between the claims of the named plaintiffs and those of other class members."); In re Am. Med. Sys., Inc., 75 F.3d 1069, 1082 (6th Cir. 1996) (typicality ensures that "the representative's interests will be aligned with those of the represented group, and in pursuing his own claims, the named plaintiff will also advance the interests of the class members"). As previously stated, Plaintiffs testified to substantially the same experiences and allege that Defendants discriminated against them according to the company-wide hiring policies. Moreover, Defendants' arguments address the merits of Plaintiffs' case and are best left for summary judgment. Messner, 669 F.3d at 823 ("All of this is at best an argument that some class member' claims will fail on the merits if and when damages are decided, a fact generally irrelevant to the district court's decision on class certification.").
Defendants also argue that typicality is not met "when it appears that a major focus of the litigation will be on a defense unique to a named plaintiff." Gorence v. Eagle Food Ctrs., Inc., No. 93 C 4862, 1994 WL 445149, at *10 (N.D. Ill. Aug. 16, 1994). Defendants argue that Zollicoffer has insurmountable credibility issues that will overshadow his common claim and that Green's bias against MVP will be a major issue in the case. These arguments overlap substantially with Defendants' arguments regarding Plaintiffs' adequacy as class representatives, and the Court addresses them below. See William B. Rubenstein, 1 Newberg on Class Actions § 3:28 (5th ed. 2019) (typicality and adequacy of class representatives are closely related inquiries). Because the named Plaintiffs' claims are typical of those of the class at large, the Court finds Plaintiffs have met this requirement.
E. Adequacy of Class Representatives
To show that they are adequate representatives of the class, Plaintiffs must show that: 1) they do not have interests that conflict with the class as a whole, 2) they are "sufficiently interested in the case outcome to ensure vigorous advocacy," and 3) class counsel is competent and willing to vigorously litigate the case. Cavin v. Home Loan Ctr., Inc., 236 F.R.D. 387, 393 (N.D. Ill. 2006). Defendants argue that Plaintiffs have met none of these prongs.
1. The Named Plaintiffs
First, Defendants argue that Plaintiffs' interests are not aligned with those of the class because they have not shown that they are part of the class and suffered the same injury. See Amchem Prod., Inc. v. Windsor, 521 U.S. 591, 625-26 (1997) ("A class representative must be part of the class and 'possess the same interest and suffer the same injury' as the class members." (citation omitted)). They contend that Green and Zollicoffer are not in the class because they stopped seeking work at MVP.
Caselaw suggests that plaintiffs in failure-to-hire lawsuits must show that they actually sought the employment in question so that they can represent the class of job applicants. See Culver v. City of Milwaukee, 277 F.3d 908, 912 (7th Cir. 2002) (plaintiff who sought work with police department was inadequate because he "made only perfunctory efforts . . . to obtain a job application," had no interest in injunctive relief or money damages, and conceded that his claim was moot); Smith v. Merchants & Farmers Bank of W. Helena, Ark., 574 F.2d 982, 984 (8th Cir. 1978) (plaintiff, who had limited contact with employer bank and was "not certain that she even completed the application process," was inadequate representative for class that included current employees), abrogated on other grounds by Gardner v. Westinghouse Broad. Co., 437 U.S. 478 (1978). This does not mean, as Defendants suggest, that Plaintiffs must demonstrate a certain level of tenacity beyond what they have already shown. Plaintiffs have all alleged that they visited the Cicero office more than once and were denied work on several occasions. Defendants have not explained why these facts would take the named Plaintiffs outside of the proposed class because all they have to show is that they were denied work on one or more occasions because of Defendants' discriminatory policy. Doc. 700 at 38.
Second, Defendants argue that Zollicoffer cannot act as a fiduciary of the class because of his criminal history. The Court disagrees. "Rule 23 contemplates, and the district court should insist on, a conscientious representative plaintiff." Rand v. Monsanto Co., 926 F.2d 596, 599 (7th Cir. 1991), overruled on other grounds by Chapman v. First Index, Inc., 796 F.3d 783 (7th Cir. 2015). The concern is that "the representative and counsel may be tempted to sell out the class for benefits to themselves." Id. Here, Defendants seek to impeach Zollicoffer's credibility as if this were a trial on the merits. "Most courts have rejected the contention that a proposed representative is inadequate because of prior unrelated unsavory, unethical, or even illegal conduct." Rubenstein, supra, § 3:68 (collecting cases). To the degree courts have allowed challenges on this basis, they are often relevant to the litigation. Id. Defendants, for example, cite Davidson v. Citizens Gas & Coke Utility, where the plaintiffs alleged the defendant employer administered a discriminatory test that had a disparate impact on African Americans in promotions and hires. 238 F.R.D. 225, 226 (S.D. Ind. 2006). The court found two of the plaintiffs with multiple criminal convictions were inadequate representatives because "the convictions themselves would have barred their consideration for employment with Citizens, regardless of their test scores." Id. at 229-30. Here, by contrast, MVP did not require criminal history checks as part of the hiring process. Additionally, Zollicoffer's past crimes are nearly all more than a decade old. His armed robbery conviction dates back to 1986, when Zollicoffer was approximately twenty-two years old, and his most recent conviction from 2011 was for a misdemeanor domestic battery. It is unlikely that any of these convictions are admissible at trial. See United States v. Byrd, 771 F.2d 215, 219 (7th Cir. 1985) ("Theft is arguably not a crime involving dishonesty or false statement."); United States v. Smith, 181 F. Supp. 2d 904, 909 (N.D. Ill. 2002) (defendant's convictions for robbery, burglary, and theft were not admissible because "the government ha[d] not shown that any of them involved false statements or acts of deceit beyond the basic crime itself").
Third, Defendants argue that Plaintiffs' deposition testimony demonstrates they are inadequately knowledgeable about the case to be representatives. See Rand, 926 F.2d at 598-99 ("Although a representative plaintiff need not immerse himself in the case . . . the named plaintiff must have some commitment to the case, so that the 'representative' in a class action is not a fictive concept."). Green testified that he had never heard of GSB and that he did not know he was suing them. Zollicoffer testified that he had no basis for suing GSB and that he did not know what it meant to be a class representative. Additionally, Defendants contend that evidence coming out of Hunt's December 20, 2020 and February 16, 2020 deposition demonstrates that the named Plaintiffs are "mere figureheads" and inserted into the case to litigate claims driven by counsel. Doc. 775 at 12.
Defendants again exaggerate the import of Plaintiffs' statements. It is not clear from Zollicoffer's deposition, for example, whether his statements were merely a concession that he had no personal knowledge about GSB's conduct in the allegedly discriminatory hiring policy, or whether he was completely unaware about Plaintiffs' grounds for suing the bakery. While Green and Zollicoffer should have known—and hopefully know by now—why they are suing GSB, their lack of knowledge in this respect is at least mitigated by the fact that they never worked at GSB, and so have no personal knowledge about GSB's conduct. The Court only expects them to testify to their personal interactions with MVP. They seek to substantiate their allegations of a company-wide discriminatory hiring policy through other witnesses. The Court also does not find it appropriate to import any concerns about Hunt and his interactions with counsel to the two remaining named Plaintiffs.
Although the Court is troubled by Plaintiffs' lack of knowledge regarding the legal process, their knowledge in this respect is mostly relevant as a means to demonstrate their interest in the case and ensuring that they will vigorously pursue their claims. See Cavin, 236 F.R.D. at 393. This showing is "not difficult." Ocampo v. GC Servs. Ltd. P'ship, No. 16-CV-9388, 2018 WL 6198464, at *9 (N.D. Ill. Nov. 28, 2018). Named plaintiffs have a "modest" burden to demonstrate "[a]n understanding of the basic facts underlying the claims, some general knowledge, and a willingness and ability to participate in discovery." Cavin, 236 F.R.D. at 394. Even if Plaintiffs in this case would ideally have a better understanding of their role, their actions reflect that they are adequate class representatives. Plaintiffs have complied with what has been required of them for several years. They have completed discovery requests, submitted to multiple depositions, completed interrogatories, and stayed in contact with their counsel. This indicates that they are intent on pursuing their claims, and that is enough in this case. See Ries v. Humana Health Plan, Inc., No. 94 C 6180, 1997 WL 158337, at *9 (N.D. Ill. Mar. 31, 1997) (finding plaintiff who "failed to convey a grasp of the class action process in her deposition" was nonetheless an adequate representative because "the court [wa]s persuaded that Ms. Ries possesse[d] a willingness to participate, when necessary, in the lawsuit").
Additionally, what constitutes adequate representation will depend on the circumstances of the specific case. In Eggleston v. Chicago Journeymen Plumbers' Local Union No. 130, U.A., the Seventh Circuit pointed to the Supreme Court's decision in Surowitz v. Hilton Hotels Corp., 383 U.S. 363, 370 (1966), where "the named plaintiff did not understand her complaint at all, could not explain the statements in it, had little knowledge of what the lawsuit was about, did not know the defendants by name, nor even the nature of the misconduct of the defendants." 657 F.2d 890, 896 (7th Cir. 1981). The plaintiff in Surowitz, an immigrant from Poland, had relied on her son-in-law to help her file a class action complaint against a corporation for securities fraud. The Supreme Court explained that her limited knowledge was not a basis to summarily dismiss the case:
The basic purpose of the Federal Rules is to administer justice through fair trials, not through summary dismissals as necessary as they may be on occasion. These rules were designed in large part to get away from some of the old procedural booby traps which common-law pleaders could set to prevent unsophisticated litigants from ever having their day in court. If rules of procedure work as
they should in an honest and fair judicial system, they not only permit, but should as nearly as possible guarantee that bona fide complaints be carried to an adjudication on the merits. Rule 23(b), like the other civil rules, was written to further, not defeat the ends of justice.Surowitz, 383 U.S. at 373. The Seventh Circuit noted that Surowitz was an extreme case but that it illustrated "the flexibility and broad area for the exercise of the trial court's common sense and good judgment" in assessing the adequacy of class representatives. 657 F.2d at 896. Similarly, the Court is loath to deny class certification in this case because one of the Plaintiffs could not describe his role as class representative or a now-withdrawn named Plaintiff appeared uninformed as to case proceedings, particularly when there has already been a complete substitution of the class representatives and the litigation is now nearing its seventh year. See Smith v. Adventist Midwest Health, No. 16 C 7606, 2018 WL 948600, at *4 (N.D. Ill. Feb. 20, 2018) ("[A] putative class representative can be adequate without being perfect or the best representative."). For these reasons, the Court finds that Green and Zollicoffer are adequate class representatives.
2. Class Counsel
Under Rule 23(a)(4) and 23(g), the Court must also scrutinize the adequacy of class counsel. Similar to the class representatives, counsel must show that "they would prosecute the case in the interest of the class . . . rather than just in their interests as lawyers who if successful will obtain a share of any judgment or settlement as compensation for their efforts." Creative Montessori Learning Ctrs. v. Ashford Gear LLC, 662 F.3d 913, 918 (7th Cir. 2011) ("We and other courts have often remarked the incentive of class counsel, in complicity with the defendant's counsel, to sell out the class by agreeing with the defendant to recommend that the judge approve a settlement involving a meager recovery for the class but generous compensation for the lawyers[.]"). The Court must also scrutinize counsel's competence. Greisz v. Household Bank (Illinois), N.A., 176 F.3d 1012, 1013 (7th Cir. 1999) (court must "assess the class lawyer's competence").
Plaintiffs are currently represented by lead counsel Christopher Williams of the National Legal Advocacy Network and co-counsel Joseph Sellers, Shaylyn Cochran, and Harini Srinivasan, of Cohen Milstein Sellers & Toll P.L.L.C. ("Cohen Milstein"). Defendants argue that class counsel is inadequate, mostly pointing to the conduct of lead counsel Christopher Williams, including: 1) prior discovery violations; 2) missed deadlines following the Court's recommendation of sanctions; 3) alleged lack of candor; and 4) missteps during litigation, extending to potential ethical violations.
Defendants also argue that Plaintiffs improperly rely on the notebooks that the Court previously barred from further use in this litigation. Defendants misconstrue the Court's previous Order. The Court excluded excerpts from the notebooks that Sanchez disclosed, and specifically advised that, "given the overlap in documents MVP produced in discovery and documents Sanchez provided to Plaintiffs, the evidence subject to exclusion should be limited to information and documents not otherwise produced in discovery." Doc. 680 at 32. Sanchez disclosed the dispatchers' practice of keeping notebooks with a list of laborers to call in her declaration and in her deposition testimony. Defendants also disclosed the practice in their own disclosures. So, Plaintiffs' reference to this practice was not improper.
Defendants level serious accusations against Williams, which the Court agrees call into question Williams' ability to represent the class. Indeed, the Court has already sanctioned Williams for discovery lapses. See Docs. 680, 692. Although Judge Kim declined Defendants' request to disqualify Williams in connection with that sanctions motion and determined that monetary sanctions should suffice to penalize Williams and deter future discovery failures or litigation misconduct, he noted that the Court could "weigh Williams's conduct in discovery as a factor in evaluating his fitness to serve as class counsel." Doc. 680 at 29. Despite this warning, Williams has continued to make missteps in this and related litigation. For example, similar to his treatment of Sanchez's phone, one of the issues underlying the imposition of monetary sanctions, it recently came to light that Williams did not fully review Hunt's cell phone for discoverable evidence and instead only produced one video from that phone. Plaintiffs respond that Defendants did not take any steps to obtain the cell phone, but Plaintiffs had the affirmative obligation to produce its contents given that, at the time they produced the video, Hunt remained a named Plaintiff. The Court also notes the potential ethical concerns arising from the loan agreement Williams entered with Hunt to cover the monetary sanctions Judge Kim imposed on Hunt personally and Williams' admitted failure to communicate this arrangement to Zollicoffer and Green. See Ill. R. Prof'l Conduct 1.8(e) ("A lawyer shall not provide financial assistance to a client in connection with pending or contemplated litigation, except that:.(1) a lawyer may advance court costs and expenses of litigation, the repayment of which may be contingent on the outcome of the matter; and (2) a lawyer representing an indigent client may pay court costs and expenses of litigation on behalf of the client."). The Court recognizes that the agreement between Hunt and Williams specifies that Hunt remains personally liable for the entirety of the sanction amount, but, given the discrepancy between the ledger presented to the Court, which reflects that Hunt made two payments toward the sanction, and Hunt's testimony that he has only ever made one payment, the Court questions the arrangement and Williams' forthrightness with the Court concerning the issue. The Court also does not condone Williams' behavior in allowing Hunt to leave his in-progress deposition in the Vee Pak matter and his apparent lack of communication with Hunt about the status of the case while Hunt remained a named Plaintiff. Although "[n]ot every ethical breach justifies the grave option of denying class certification," Reliable Money Order, Inc. v. McKnight Sales Co., 704 F.3d 489, 498 (7th Cir. 2013), the Court finds that the ethical concerns surrounding Williams' arrangement with Hunt and his failure to disclose this arrangement to the remaining named Plaintiffs, coupled with Williams' other discovery lapses and litigation misconduct over the seven years this case has been pending, create "serious doubt that counsel will represent the class loyally," Creative Montessori, 662 F.3d at 918; see also Reliable Money Order, 704 F.3d at 499 ("[U]nethical conduct, not necessarily prejudicial to the class, nevertheless raises a 'serious doubt' about the adequacy of class counsel when the misconduct jeopardizes the court's ability to reach a just and proper outcome in the case."). Therefore, the Court declines to appoint Williams as class counsel.
But the Court's concerns about Williams' adequacy as class counsel do not extend to his co-counsel from Cohen Milstein. Although Defendants contend that the Court should consider counsel collectively, they do not specifically raise concerns about the adequacy of counsel from Cohen Milstein or explain why the Court should impute Williams' conduct to his co-counsel. Given the Court's experience with this case, the Court does not find it appropriate to do so and does not have the same concerns about the adequacy of Cohen Milstein to loyally represent the class as it does with Williams. The Court therefore finds Plaintiffs have met the adequacy of representation requirement with Cohen Milstein serving as class counsel. The Court expects counsel from Cohen Milstein to faithfully represent the named Plaintiffs and class members in accordance with their obligations as class counsel.
In supplemental briefing, Plaintiffs also note that their counsel will likely seek to add additional local counsel in preparation for trial.
F. Rule 23(b)(3) Requirements
Defendants argue that Plaintiffs have not met the requirements for certification under Rule 23(b)(3), which requires that the "the court find[] that the questions of law or fact common to class members predominate over any questions affecting only individual members, and that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy." Some of the relevant topics include: 1) whether individual class members have an interest in individual control over the litigation, 2) whether there is other litigation concerning the same controversy, 3) the advantages and disadvantages of assembling claims in one judicial forum, and 4) problems that might arise in managing the case as a class action. Fed. R. Civ. P. 23(b)(3).
Defendants argue common questions will not predominate in this case because the trier of fact will have to consider a litany of individual questions before even reaching questions regarding the discriminatory hiring policy. Defendants argue these individual questions include when the laborer sought work at MVP, what shifts they were willing to work, whether GSB needed employees on those particular days, and whether the laborers even qualified for positions at GSB. But Defendants fundamentally misunderstand the predominance inquiry. The predominance requirement demands that the proposed class is "sufficiently cohesive to warrant adjudication by representation." Amchem, 521 U.S. at 623. Individual questions need not be absent. The inquiry simply asks "whether the common, aggregation-enabling, issues in the case are more prevalent or important than the non-common, aggregation-defeating, individual issues." Tyson Foods, Inc. v. Bouaphakeo, 577 U.S. ----, 136 S. Ct. 1036, 1045 (2016) (citation omitted). Here, Defendants' list of individualized questions are actually secondary to Plaintiffs' common claims. If the trier of fact determines that Defendants had a discriminatory policy against hiring African Americans—something that Plaintiffs mainly seek to prove through the testimony of former MVP employees and Dr. Bendick's analysis—the individualized questions related to when the laborer sought work at MVP, whether they were qualified, and whether GSB needed laborers on that particular day, will largely bear on damages. Similar to the defendant in Butler v. Sears, Roebuck & Co. ("Butler II"), Defendants "think[] that predominance is determined simply by counting noses: that is, determining whether there are more common issues or more individual issues, regardless of relative importance." 727 F.3d 796, 801 (7th Cir. 2013). "But predominance requires a qualitative assessment too; it is not bean counting." Id. None of the questions that Defendants cite are central to assessing Plaintiffs' main contention. As the Seventh Circuit noted in Butler II: "the fact that damages are not identical across all class members should not preclude class certification. Otherwise defendants would be able to escape liability for tortious harms of enormous aggregate magnitude but so widely distributed as not to be remediable in individual suits." Id. This is one of the central justifications for asserting a class action for damages: to aggregate small individual claims that would not be justified if each class member sought to litigate the case by himself. Mace v. Van Ru Credit Corp., 109 F.3d 338, 344 (7th Cir. 1997) ("The policy at the very core of the class action mechanism is to overcome the problem that small recoveries do not provide the incentive for any individual to bring a solo action prosecuting his or her rights. A class action solves this problem by aggregating the relatively paltry potential recoveries into something worth someone's (usually an attorney's) labor.").
Defendants next contend that class members likely have an individual interest in controlling the litigation to avoid the distractions that the named Plaintiffs bring to the case. As already discussed, the named Plaintiffs are adequate representatives. It is also questionable whether there are other class members with an appetite to initiate their own lawsuit at this stage given the cost/benefit analysis and the passage of time. Defendants offer no argument, and the Court does not think it likely, that a class member's individual recovery would justify the costs of litigation. See Amchem, 521 U.S. at 616 ("Th[e] interest [in individual control] can be high where the stake of each member bulks large and his will and ability to take care of himself are strong; the interest may be no more than theoretic where the individual stake is so small as to make a separate action impracticable." (citation omitted)).
Defendants also contend that the class action is unmanageable because it will degenerate into several individualized trials and because damages are not susceptible to proof on a classwide basis under Comcast Corp. v. Behrend, 569 U.S. 27 (2013). As already explained, Plaintiffs' central contentions will predominate in the liability phase of the case, and the Court can conduct individualized inquiries to determine damages if necessary. Wal-Mart, 564 U.S. at 366-67 (in cases alleging discriminatory employment practices, courts "must usually conduct additional proceedings . . . to determine the scope of individual relief" where the burden of proof shifts to the employer who "will have the right to raise any individual affirmative defenses it may have" (citing Teamsters, 431 U.S. at 361)). Defendants reliance on Comcast is misplaced. Comcast dealt with an antitrust suit where damages flowed from the reduced competition in the marketplace. 569 U.S. at 35. The Supreme Court held that the plaintiffs needed to show that their methodology for calculating damages on a classwide basis was tied to their theory of liability. "If the model does not even attempt to do that, it cannot possibly establish that damages are susceptible of measurement across the entire class for purposes of Rule 23(b)(3)." Id. Defendants attempt to interpret this to mean that every class action must be susceptible to proof of damages on a classwide basis—an argument that the Seventh Circuit, and other circuits, have rejected. See Butler II, 727 F.3d at 801 ("It would drive a stake through the heart of the class action device, in cases in which damages were sought rather than an injunction or a declaratory judgment, to require that every member of the class have identical damages."); In re Deepwater Horizon, 739 F.3d 790, 815 (5th Cir. 2014) ("[N]othing in Comcast mandates a formula for classwide measurement of damages in all cases.").
Defendants contend that Plaintiffs' proposal for a bifurcated trial will violate the Seventh Amendment reexamination clause. Plaintiffs propose that the first phase of a trial be dedicated to the question of liability, specifically:
1) Whether Defendants have engaged in a pattern or practice of discrimination;Doc. 700 at 57. Plaintiffs propose that damages and any other relief can be determined in a separate proceeding according to any number of different procedures, including individualized hearings or the use of a formulaic calculation for damages. Defendants argue that in the case of a second jury, sworn solely for the purpose of determining damages, the second jury might find that a class member is not entitled to damages based on his work availability, willingness to do certain types of work, or a number of other factors. They argue that this would nullify the first jury's finding of liability, since these questions are relevant to both determinations. Contrary to what Defendants contend, the individual questions regarding the date each class member sought work and whether there were jobs at GSB available on those particular days will largely bear on the question of damages. Defendants cite to Matter of Rhone-Poulenc Rorer, Inc., where the Seventh Circuit invalidated the district court's plan to certify a class action where the first jury would solely consider the question of negligence for the class and could go on to decide additional issues only with respect to the named plaintiffs. 51 F.3d 1293, 1297 (7th Cir. 1995). If it found for the plaintiffs on the question of negligence, "individual members of the class would then file individual tort suits in state and federal district courts around the nation and would use the special verdict, in conjunction with the doctrine of collateral estoppel, to block relitigation of the issue of negligence." Id. The subsequent cases would each have needed to consider questions of comparative negligence and proximate causation—issues that necessarily overlapped with the first jury's determination and might have involved findings that contradicted the first jury's verdicts for the named plaintiffs. Id. at 1303. The procedure in Rhone-Poulenc is entirely different from the procedure Plaintiffs propose. As the Rhone-Poulenc court noted, a reexamination problem "is not inherent in bifurcation." Id. The caselaw endorsing bifurcated proceedings for individualized damage assessments should make it clear that Plaintiffs' proposal for a separate damages proceeding will not violate the Seventh Amendment because the second trier of fact would not need to reconsider the central question of liability. See Wal-Mart, 564 U.S. at 366-67; Butler II, 727 F.3d at 801. For these reasons, the Court finds that Plaintiffs have met the requirements under Rule 23(b)(3).
2) Whether Defendants are joint employers and/or MVP is an agent of GSB within the meaning of Section 1981;
3) Whether Defendants' conduct meets the standard for an award of punitive damages; and
4) Whether a four-year limitations period governs Plaintiffs' Section 1981 claim.
G. Ascertainability
Lastly, Defendants argue that the class is not ascertainable because it is vague, includes people who received work assignments to clients other than GSB, and because the proposed class extends beyond the statute of limitations. A class must be clearly defined according to objective criteria. Mullins v. Direct Digital, LLC, 795 F.3d 654, 657 (7th Cir. 2015). It cannot be vague or subjective. Id.
The court in Vee Pak considered a motion to dismiss involving essentially identical allegations of racially discriminatory hiring practices through staffing agencies. No. 12-CV-09672, 2017 WL 6733688, at *4 (N.D. Ill. Dec. 20, 2017). The proposed class in Vee Pak was defined as "all African Americans who sought a work assignment through [the temporary staffing agency] and were otherwise eligible to work at Vee Pak but, on one or more occasion, were not assigned or hired to work [during the class period]." Id. at *2. The court found that the term "otherwise eligible" was too vague and eliminated the phrase from the class definition. Similar to the jobs at GSB, the Vee Pak court reasoned that the term was "essentially meaningless" because "there were no qualifications for the Vee Pak jobs at issue," and that "even if some minimal eligibility requirements did exist, a class member's eligibility to work at Vee Pak could be determined at the remedial stage of the litigation." Id. at *4. The same reasoning applies to this case because there were essentially no prerequisites necessary to work at GSB, and minimal differences in qualifications, such as prior experience, can be handled in the second phase of a trial. Thus, the Court finds that Defendants' vagueness concerns are remedied by eliminating the phrases "otherwise eligible" and "otherwise qualified" from the class definition.
Zollicoffer is a named plaintiff in the case and MVP was one of the defendants. --------
Defendants' argument regarding laborers who were assigned to different placements has already been addressed. Defendants' remaining argument regarding the statute of limitations is a closer question. Plaintiffs previously asked the Court to amend the class definition in their fifth amended complaint by expanding the class period from two years to four years, which the Court allowed while noting that the appropriate limitations period would depend on how MVP conducted its sign-in procedures. Doc. 320 at 7-8. Plaintiffs' theory is that signing in at the Cicero office constituted a contract of employment because it placed laborers within MVP's pool of laborers available for assignments. By steering Plaintiffs away from assignments at GSB, Plaintiffs contend that Defendants interfered with an established contractual relationship and not just the potential formation of a contract with GSB. Failure-to-hire claims, which implicate the formation of a contract, borrow the applicable statute of limitations from state law, which in this case is two years. See Rainey v. United Parcel Serv., Inc., 543 F. App'x 606, 608 (7th Cir. 2013) ("Rainey alleges that unlawful discrimination prevented the making of an employment contract, and for claims of discrimination in hiring arising in Illinois, a two-year statute of limitations governs."). If Plaintiffs' theory is correct, however, their claims would carry a four-year statute of limitations under 28 U.S.C. § 1658(a). Id. (four-year statute of limitations for § 1981 claims "applies only to claims based on conduct occurring after the formation of an employment contract").
Plaintiffs and Defendants offer little argument on this issue, and there is no controlling caselaw on the topic. Cf. Hunt v. Pers. Staffing Grp., LLC, No. 16-CV-11086, 2018 WL 1014513, at *7 (N.D. Ill. Feb. 22, 2018) (determining whether temporary laborers entered into a contract with MVP by signing in and requesting work assignments is a fact-intensive inquiry that cannot be resolved on a motion to dismiss); Pruitt v. Pers. Staffing Grp., LLC, No. 16 C 5079, 2017 WL 1128457, at *2 (N.D. Ill. Mar. 23, 2017) (same). Plaintiffs assure the Court that this issue can be decided by the trier of fact during the liability phase of the trial. But deciding which statute of limitations applies is a question of law central to class certification. See Phillips v. Asset Acceptance, LLC, 736 F.3d 1076, 1081 (7th Cir. 2013) (statute of limitations is a question of law that implicates composition of class and whether it is appropriate to maintain the suit as a class action). The Court directs the parties to submit additional briefing, limited to no more than ten pages each, on the question of whether MVP formed a contract with job seekers who signed in and filled out an application for work assignments at the Cicero office. In particular, it would be helpful for the Court to review a copy of the application form that job seekers were required to complete. Because this question will determine the scope of the class, the Court reserves adopting a final class definition pending resolution of this issue.
CONCLUSION
For the foregoing reasons, the Court grants Plaintiffs' motion for class certification [699]. The Court appoints James Zollicoffer and Norman Green as class representatives and Joseph M. Sellers, Shaylyn Cochran, and Harini Srinivasan of Cohen Milstein Sellers & Toll P.L.L.C. as class counsel. The Court reserves defining the class pending additional briefing by the parties. The parties should follow this briefing schedule: Plaintiff's brief is due by May 4, 2020 and Defendants' brief is due by June 1, 2020. The Court sets a status date of July 8, 2020, where the Court will provide a ruling defining the class. Dated: March 31, 2020
/s/_________
SARA L. ELLIS
United States District Judge