IN THE UNITED STATES COURT OF APPEALS

FOR THE SECOND CIRCUIT

___________________________

No. 14-1782

___________________________

EQUAL EMPLOYMENT OPPORTUNITY COMMISSION,

Plaintiff-Appellant,

STERLING JEWELERS, INC.,

Defendant-Appellee.

___________________________________________

On Appeal from the United States District Court

for the Western District of New York

The Honorable Richard J. Arcara, District Judge

________________________________________________________

REPLY BRIEF OF

THE EQUAL EMPLOYMENT OPPORTUNITY COMMISSION

________________________________________________________

INTRODUCTION

This Title VII enforcement action grew out of nineteen charges from across the country, most alleging class and/or company-wide discrimination, that EEOC investigated. The parties also engaged in extensive professional mediation, in which EEOC also participated, before EEOC found cause and filed suit, alleging that Sterling engaged in a pattern or practice of sex-based discrimination in pay and promotion nationwide.

Despite the pre-suit activity, the district court granted summary judgment, holding that EEOC failed to prove that it conducted a nationwide administrative investigation. EEOC’s opening brief argued mainly that this ruling was erroneous because the investigation fulfilled the purpose of Title VII’s investigation requirement — to determine whether there was reasonable cause to believe the allegations in the charges were true. In EEOC’s view, the court improperly questioned the sufficiency of EEOC’s investigation, which courts uniformly hold is not a proper inquiry. See, e.g., EEOC v. Keco Indus., 748 F.2d 1097, 1100 (6th Cir. 1984).

Sterling does not deny that EEOC had enough information to make a reasonable cause determination. At one point, the company asserted that EEOC had “conducted a nationwide administrative investigation.” Sterling also admits that EEOC satisfied the other pre-suit requirements including notice and conciliation.

The theme of Sterling’s brief is form over substance. Sterling takes the position that unless EEOC can document that it obtained the information through formal channels — specifically, an “investigator” requested and received it — it does not count. Similarly, although EEOC’s investigative files — totaling over 2600 pages — contain copies of an expert statistical analysis that was prepared for and discussed during the mediations, that analysis does not count because EEOC did not prove that it recreated or, at least, independently verified the analysis, before finding cause. Sterling cites no authority for its articulation of EEOC’s investigation duty.

These arguments, like the court’s ruling, boil down to challenges to the sufficiency of EEOC’s investigation. They should therefore be rejected and the case remanded to resolve the real issue in the case — whether Sterling is liable for the alleged Title VII violations.

ARGUMENT

1. EEOC’s Opening Brief argued that the district court erred when it held that although EEOC investigated the nineteen charges, most alleging class or company-wide discrimination, EEOC did not prove it made a “nationwide investigation.” EEOC argued that Title VII simply requires EEOC to “make an investigation” to determine whether there is reasonable cause to believe that the allegations in the charges are (or are not) true. Once EEOC has enough information to make a reasonable cause finding, the investigatory phase is over, and if cause is found, the process moves to the conciliation phase — not at issue here. EEOC-Br.32-34 (citing, e.g., 42 U.S.C. §2000e-5(b); EEOC v. Shell Oil Co., 466 U.S. 54, 71 (1984)).

Thus, EEOC argued, the court’s inquiry should have ended when the court found that EEOC investigated the charges. Courts, including this district court, agree that it is “error” to inquire into the “sufficiency” of the investigation; the “nature and extent” of the investigation are “matters within [EEOC’s] discretion.” By scrutinizing the minutiae of the investigation under the guise of assessing whether EEOC did a “nationwide investigation,” the court effectively end-ran this ban on sufficiency reviews. EEOC-Br.34-47 (citing, e.g., Keco, 748 F.2d at 1100; EEOC v. Caterpillar, 409 F.3d 831, 832-33 (7th Cir. 2005); Newsome v. EEOC, 301 F.3d 227, 231 (5th Cir. 2002)).

Rather than address directly what EEOC actually argued, Sterling’s brief makes three related arguments, none of which has merit.

First, Sterling repeatedly asserts that EEOC takes the position that it need not do any investigation and its actions are immune from judicial review. Sterling-Br.39-40, 41, 42, 62. This position, Sterling stresses, “has been authoritatively and repeatedly rejected.” Sterling-Br.62.

In making these assertions, Sterling quotes nothing from EEOC’s brief. Nor could it, since EEOC did not take that position. On the contrary, EEOC acknowledged that under Title VII, it must “make an investigation” in order to determine whether there is reasonable cause to believe the allegations in the charge are true. EEOC-Br.32-34. EEOC also acknowledged that courts may check whether EEOC did any investigation at all. EEOC-Br.41-43. EEOC argued that the courts do not review the sufficiency of the investigation. EEOC-Br.34-40 (citing, e.g., Keco, 748 F.2d at 1100). Sterling admits that this “unbroken line of authority” “makes sense.” Sterling-Br.36.

Second, Sterling argues that while sufficiency is off-limits, the court may determine whether EEOC did “any investigation” at all. Sterling-Br.36. As noted above, EEOC agrees. EEOC’s Opening Brief identified EEOC v. Pierce Packing Co., 669 F.2d 605, 606-08 (9th Cir. 1982), as such a case. There, the Court dismissed EEOC’s suit after the agency “leap-frogged” the entire administrative process — investigation, LOD, and conciliation — going directly into court. EEOC-Br.41-43.

This case is nothing like Pierce Packing. Sterling concedes that EEOC satisfied all the other pre-suit requirements; the magistrate found that EEOC had investigated the nineteen charges (App-74-76) and EEOC participated in lengthy mediations where Sterling and the charging parties exchanged and discussed expert evidence. In addition, shortly before moving for summary judgment, Sterling stated that EEOC “conducted a nationwide administrative investigation.” App-132-33(v2). Dismissal of this case for failure to investigate at all would therefore be improper.

Third, Sterling argues that the court may review the “scope” of the investigation. As support, Sterling cites a string of cases where the charge, the investigation, the LOD, and conciliation were narrow, but EEOC’s resulting lawsuit alleged a broader claim. See Sterling-Br.37-39 & Addendum (citing, e.g., EEOC v. CRST Van Exp’d, 679 F.3d 657, 675-76 (8th Cir. 2012); EEOC v. Outback Steak House, 520 F.Supp.2d 1250, 1263-69 (D.Colo. 2007); EEOC v. Jillian’s of Indianapolis, 279 F.Supp.2d 974, 980 (S.D.Ind. 2003)). According to Sterling, these cases hold that “the contours of EEOC’s investigation define the permissible contours of the lawsuit.” Sterling-Br.37.

Because the district court also cited these “scope-notice” cases, EEOC’s Opening Brief explained why they are inapposite. EEOC-Br.43-44. They do not stand for Sterling’s proposed reading, which would swallow the rule that because the nature and extent of an investigation are matters within EEOC’s discretion, courts will not review the sufficiency of the investigation. In fact, CRST — one of Sterling’s main cases — cites Keco and the “general rule.” 679 F.3d at 674.

Instead, the cases create a narrow exception for situations where, in the courts’ view, the charge, LOD, conciliation, and investigation involved only one or a few individuals, so the employer did not have notice of the full extent of its potential exposure or an opportunity to conciliate EEOC’s resulting claim. See EEOC-Br.43-44 (discussing CRST, 679 F.3d at 676-78; Jillian’s, 279 F.Supp.2d at 979-82); see also id. (discussing EEOC v. Bloomberg, 967 F.Supp.2d 802, 813-14 (S.D.N.Y. 2013)). In such cases, courts have limited the scope of EEOC’s suit to what the employer had notice of below. See, e.g., Outback, 520 F.Supp.2d at 1269 (dismissing nationwide claims because administrative process “failed to put Defendants on notice of the nationwide scope of the potential claims against them”).

In contrast, here, Sterling admits that it had ample notice of its potential exposure and an opportunity to resolve the nationwide claim. The cases therefore do not support Sterling’s argument any more than they supported the decision below. Sterling has no answer to this argument.

2. By repeatedly referring to a “nationwide investigation,” Sterling may be suggesting that EEOC had a duty to do a special kind of investigation. At intervals throughout its brief, Sterling lists types of investigatory techniques such as an on-site investigation that, it asserts, EEOC — or Investigator David Ging — admits were never done. See Sterling-Br.3-4, 48. Thus, Sterling appears to suggest that a “nationwide investigation” requires that EEOC exhaust all possible avenues of inquiry. This argument is flawed both factually and legally.

While asserting that deposition testimony and documents “confirm” that no “nationwide investigation” was done, Sterling does not cite any record support for these assertions. For the most part, there is none. The two deponents, Ging and Rule 30(b)(6) witness Jennifer Carlo — who were deposed 5-6 years after their last involvement with the case (during which time they were busy with other investigations (App-731) — generally “admitted” that they did not know or did not recall what happened back then. They did not “admit” that nothing was done. For example, Sterling asserts that Ging admitted never speaking to a charging party (Sterling-Br.48), but Ging actually testified that he did not recall whether he had done so (App-464). Compare, e.g., Sterling-Br.3(EEOC never spoke to any current or former Sterling employee) with App-1505(Boyle file: noting “CP contact/interview”); see also Sterling-Br.47-49(arguing “evidence” reveals EEOC did not do nationwide investigation, but citing only Sterling’s “Statement of Facts” below (Supp.App.63)). Because Sterling’s brief is premised on the unsupported factual assumption that EEOC admitted doing no investigation, its argument falls of its own weight.[1]

Similarly, Sterling consistently characterizes the investigatory files as “slim,” “thin,” or “meager,” suggesting that they contain only charges, Sterling’s responses, and a little correspondence. Sterling-Br.5, 14, 26; see also Sterling-Br.3(“paucity of documents”). In fact, even after some duplicate pages were removed, the non-privileged portions of the files as a whole are over 2600 pages long. See App-970-3541. In addition to the charges, Sterling’s responses, and some correspondence, they include, for example, Sterling’s diversity, EEO, anti-harassment, open-door, and other policies; charging parties’ personnel documents; witness statements; job descriptions; EEO-1 reports; the mediation documents; and Dr.Lanier’s statistical analysis. Lead Charging Party Laryssa Jock’s file is the most complete (App-970-1381); EEOC did not copy every document for every file. What Sterling did not provide, EEOC obtained from the charging parties before, during, and after the mediations. App-3680; App-1028.[2] Thus, rather than “confirm” that EEOC did no investigation, the files strongly suggest that EEOC satisfied its duty to “make an investigation” before filing suit.

Furthermore, despite Sterling’s suggestions to the contrary, EEOC was not required to exhaust all possible investigative techniques in order to satisfy its duty to make an investigation in this case. Because the purpose of an investigation is to determine whether there is reasonable cause to believe discrimination occurred (Shell Oil, 466 U.S. at 71), EEOC was free to stop investigating once it concluded that it had enough information to make that determination. Sterling has not challenged the evidentiary sufficiency of the cause determination. At best, Sterling’s arguments, including its multiple lists, go to the sufficiency, not the existence, of the investigation. And, as Sterling acknowledged, sufficiency is something courts do not review. See Sterling-Br.36 (limitation “makes sense”).

3. EEOC’s Opening Brief discussed some reasons why courts adopted the ban on reviewing the sufficiency of the investigation. EEOC noted, for example, that courts and Congress place greater emphasis on conciliation and notice to the employer than on the investigation. EEOC-Br.36-38 (citing, e.g., Caterpillar, 409 F.3d at 832 (7th Cir. 2005); Serrano/EEOC v. Cintas Corp., 699 F.3d 884, 904 (6th Cir. 2012), cert. denied, 134 S.Ct. 92 (2013)). In addition, courts recognize that — as happened here — allowing scrutiny of the investigation diverts the parties and the courts from the real issue in the litigation: whether the employer violated Title VII. EEOC-Br.38 (citing, e.g., Keco, 748 F.2d at 1100). And because courts, rather than EEOC, ultimately enforce Title VII, the investigation has no “determinative consequences” for the employer or charging party. EEOC-Br.39-40 (citing Georator v. EEOC, 592 F.2d 765, 768-69 (4th Cir. 1979); Gibson v. Mo. Pac. R.R., 579 F.2d 890, 891 (5th Cir. 1978)).

Perhaps because the focus on the investigation here did distract the court and parties from the question of whether Sterling violated Title VII, Sterling ignores that reason for the ban. The company does, however, purport to address the other two reasons.

Recharacterizing the first reason, Sterling argues that EEOC provides no “legal support” for its “contentions” that Sterling waived its right to challenge the investigation by conceding that EEOC conciliated the claims, and that the wording of the LOD substitutes for an investigation. Sterling-Br.40 (citing EEOC-Br.37-38).

To the contrary, there is in fact relevant “legal support” for these contentions. Cintas, for example, held that since the employer had notice that EEOC was investigating “class-wide instances of discrimination,” the court on remand should simply determine whether EEOC attempted to conciliate the claims asserted in its lawsuit, instead of considering the sufficiency of the investigation. 699 F.3d at 904 (adding that LOD also provided notice of lawsuit’s potential scope). Since Sterling conceded the adequacy of the LOD and conciliation, Cintas indicates that the court here erred in allowing Sterling to refocus the court’s attention on the investigation rather than the merits of the case.

Sterling also recharacterizes the third reason in purporting to respond to it. According to the company, “EEOC asserts that, because its investigation has no ‘determinative consequences,’ it should not be subject to any judicial review.” Sterling stresses that the term “determinative consequences” comes from Georator, but Georator is “not relevant” because it concerned the Administrative Procedures Act (APA), which is not at issue here. Sterling-Br.42.

Of course, EEOC actually “asserted” only that the sufficiency of EEOC’s investigation is normally unreviewable. As for Sterling’s “determinative consequences” point, EEOC cited two cases, Georator and Gibson. Gibson is not an APA case, but its holding is essentially the same as Georator’s. That is, nothing EEOC does or omits during the administrative process affects parties’ substantial rights since only courts can compel employers to amend their practices or compensate victims of discrimination. EEOC-Br.39 (citing Gibson, 579 F.2d at 891). Sterling’s attempted distinction of Georator does not go to the substance of the holding.

Sterling also contends that the “argument” — presumably, that errors during the administrative process have no determinative consequences — “makes no sense” and is “disingenuous” since the investigation here led to the LOD from which “EEOC launched its nationwide pattern-or-practice claim.” Sterling-Br.42-43. But this so-called “non-sensical, disingenuous” argument flows directly from cases such as Georator and Gibson holding that nothing that happens during the administrative process — including the investigation — affects a party’s substantial rights. And if Sterling means to suggest that the filing of EEOC’s lawsuit is a determinative consequence, that argument likewise goes nowhere. Cf. FTC v. Standard Oil Co., 449 U.S. 232, 243-44 (1980) (company is not “aggrieved” within the meaning of the APA by an agency’s filing of a administrative complaint).

4. EEOC’s Opening Brief also challenged the district court’s definition of “investigation” as a “‘thorough’ or ‘searching’ inquiry” that must be “independent” and “genuine.” EEOC argued that in devising its definition, the court mistakenly failed to consider the purpose and context of Title VII’s requirement. The definition is not only impractical, given EEOC’s administrative workload, but also just another way of circumventing the ban on reviewing the sufficiency of an investigation. EEOC further argued that the correct standard is one adopted by EEOC in an exercise of its discretion. Specifically, an investigation should be “appropriate to the particular charge” and should continue until EEOC has “sufficient information” to make a reasonable cause determination or to conclude that “further investigation is not likely to result in [such a] finding.” EEOC noted that this standard is consistent with the purpose of the investigation requirement and has Congress’s implicit endorsement. See EEOC-Br.47-54(citing, e.g., Shell Oil, 466 U.S. at 71; App-3691-96(Priority Charge Handling Procedures), and listing later-enacted statutes incorporating, without comment, Title VII’s administrative requirements).

Rather than defend the court’s standard or offer one of its own, Sterling simply argues that EEOC’s standard should be rejected. Sterling opines that EEOC’s standard is not entitled to “Chevron deference” (EEOC did not argue Chevron). And, according to Sterling, it is “entirely circular” since the investigation is “appropriate” if EEOC can make a reasonable cause determination. Sterling-Br.43-44 (citing, e.g.¸ McMemeny v. City of Rochester, 241 F.3d 279, 284 (2d Cir. 2001); Univ. of Texas Sw. Med. Ctr. v. Nassar, 133 S.Ct. 2517, 2533 (2013)(circular)).

Sterling’s objections do not account for the fact that EEOC’s approach accords with the purpose of the provision and received Congress’s implicit stamp of approval. See McMemeny, 241 F.3d at 284 (EEOC’s interpretation is “persuasive” because it accords with plain language and purpose of provision). Furthermore, carried to its logical conclusion, rejecting EEOC’s standard means that EEOC must continue investigating even after it has enough information to make a cause finding and, so, has achieved the purpose of the requirement. That would simply waste EEOC’s limited resources.

Sterling also takes exception to what it describes as EEOC’s argument that “its investigation should be immune from review because it cannot be expected to engage in a ‘searching inquiry’ of each of the 90,000+ charges filed annually.” Sterling stresses that EEOC cannot avoid a “statutory obligation” merely because it is burdensome, adding that since EEOC files only a few suits, it should investigate those claims more carefully. Sterling-Br.41 (workload argument is “pure hyperbole”).

This argument misunderstands EEOC’s point. The phrase “immune from review” is Sterling’s; EEOC made no such argument. Rather, EEOC simply challenged the magistrate’s “searching-inquiry” standard as unsuited to Title VII, rather than the sources from which it was lifted. Nor does Sterling cite any authority for its assumption that EEOC has a statutory obligation to make a “searching inquiry” into charges. Consistent with the purpose of the investigation requirement, EEOC must investigate until it has enough information to make a cause finding. No more, no less.

As for investigating more when EEOC plans to litigate, that misunderstands the charge process. When a charge comes in, EEOC does not know that it will someday litigate those allegations: it probably will not. Most often, the investigation does not lead to a finding of reasonable cause. If cause is found (3.6% of charges in FY2013), about half the time (42% in FY2013), the claim will be resolved in conciliation. For the rest, EEOC typically issues a right-to-sue notice, opting against litigation. See www.eeoc.gov/eeoc/statistics/enforcement/all. Even here, part-way into the investigation, Sterling and the charging parties hired outside mediators to assist in settlement/conciliation, so EEOC did not anticipate litigating the claims.

Finally, Sterling argues that EEOC had a “mandatory” duty to do a “genuine,” “independent” investigation. Sterling-Br.32, 59 (citing Martini v. Fed. Nat’l Mortg. Ass’n, 178 F.3d 1336, 1346 (D.C.Cir. 1999); Pierce Packing, 669 F.2d at 609; EEOC v. Michael Constr., 706 F.2d 244, 252-53 (8th Cir. 1983)). EEOC’s Opening Brief discussed all three cases, which were cited by the court below. EEOC-Br.55-56 & n.6. Sterling does not acknowledge that discussion.

In any event, EEOC agrees that it may not litigate without first “mak[ing] an investigation” (42 U.S.C. §2000e-5(b)). The investigation must be genuine — unlike in Pierce Packing, discussed, supra, at pp.5-6, EEOC cannot “leapfrog” that step. And contrary to the employer’s position in Michael, EEOC need not accept whatever a party chooses to provide during an investigation but must decide for itself what and how much information it needs to make a cause determination. Contrary to Sterling’s arguments, EEOC did that here.

5. EEOC’s Opening Brief argued that the magistrate erred in dismissing EEOC’s suit on the ground that, because it refused to waive the deliberative process privilege and disclose confidential information that intruded on its decisionmaking processes, EEOC could not prove that it conducted a “nationwide investigation.” See App-87 (EEOC cannot use privilege as both a sword and a shield). EEOC argued that even if information about the investigation were relevant — it was not — before EEOC could be required to waive the privilege, Sterling had to make a sufficient showing of particularized need to outweigh the public interest in nondisclosure. Sterling neither did nor could make that showing. EEOC-Br.64-69 (citing cases). The company was, therefore, never entitled to the information in the first place.

Sterling initially states that EEOC “misrepresents the issue” in arguing that EEOC could prove it did a sufficient investigation only by waiving the deliberative process privilege and disclosing its pre-suit deliberations. “To the contrary,” Sterling argues, the magistrate “repeatedly and properly warned EEOC that it would not be able to assert privilege” — that is, it would have to waive the privilege — and then attempt to use that privileged information to create a genuine issue of fact as to the sufficiency of its investigation. Sterling-Br.54. Exactly EEOC’s point. Since the magistrate’s “searching-inquiry” standard required EEOC to produce information protected by the deliberative process privilege, EEOC could avoid summary judgment only by waiving the privilege and disclosing the information in discovery.

Sterling then argues that because, in Sterling’s view, EEOC bore the burden of proving that it conducted a nationwide investigation, Sterling did not have to make any showing of need for the privileged information. In light of the sword/shield doctrine, Sterling argues, the burden was all EEOC’s. Sterling-Br.54.

Sterling, however, never mentions, let alone distinguishes, any of the cases EEOC cited for the principle that before Sterling could compel discovery of privileged information, the company was required to show need. Instead, Sterling quotes Allstate Insurance Co. v. Levesque, 263 F.R.D.663, 667 (M.D. Fla. 2010): “[U]nder the sword and shield doctrine, a party who raises a claim that will necessarily require proof by way of a privileged communication cannot insist that the communication is privileged.” Sterling-Br.54-55.

This confirms Sterling’s mistake. EEOC did not “raise a claim that require[d] proof by way of a privileged communication.” Although Sterling tries hard to obscure this fact, EEOC’s claim is that Sterling engaged in a nationwide pattern or practice of sex discrimination in pay and promotion. It was Sterling that insisted on exploring the sufficiency of the investigation.

Finally, Sterling asserts, incorrectly, that the magistrate properly “held” that EEOC could not rely on Dr.Lanier’s statistical analysis because it was prepared during the mediation and EEOC agreed that it would not “lose [its] mediation privilege” after being placed in EEOC’s investigative file. EEOC, the company states, “ignore[d]” but “cannot overcome” this point. Sterling-Br.55(citing App-75-76)(“Background”).

EEOC “ignored” this point because the court made no such holding. The court “question[ed]” whether EEOC could rely on the analysis to prove that it conducted a “nationwide investigation.” App-90 n.10. But it granted summary judgment based on EEOC’s failure to prove that the analysis was the one mentioned in the LOD and was verified by someone at EEOC.

Furthermore, Sterling proffered no evidence that this so-called “privilege” extends to situations such as this one where, to attempt to satisfy the court and Sterling that it did a sufficient investigation, EEOC was required to disclose the investigative files, which contain the analysis. EEOC acknowledged that it could not rely on the analysis as proof of discrimination at trial. App-90. The term “mediation privilege” is found in a paragraph in the mediation documents entitled “confidentiality of process relative to EEOC’s records.” That paragraph states that information disclosed during the mediation and placed in EEOC’s investigative files is not subject to “disclosure under the Freedom of Information Act (FOIA)” and “shall not be made available to any non-Charging Party.” App-3663. EEOC therefore could not provide the analysis to a FOIA applicant or non-charging party, but nothing prevented EEOC from relying on it in making a reasonable cause determination. In any event, it is ironic, given Sterling’s insistence that EEOC should have waived its deliberative process privilege in order to satisfy Sterling’s curiosity about what was said and done during the administrative process, that the company is here attempting to hide behind its own claim of privilege for a defense that it has raised.

6. EEOC’s Opening Brief argued that EEOC satisfied its pre-suit duty to investigate the nationwide allegations in the charges. Along with information specific to each charge, EEOC requested EEO-1 reports, job descriptions, company policies, and information about the personnel-related company databases. Much of what EEOC obtained, from charging parties as well as Sterling, is in the investigative files. App-970-3541.

Moreover, and significantly, EEOC’s then-regional attorney participated in the mediations — a potential gold mine of information — where EEOC had access, inter alia, to the parties’ respective expert analyses and to discussions of the pros and cons of each. Dr.Lanier’s statistical analysis, prepared for the charging parties, later became part of EEOC’s investigative files. And, EEOC argued, because the regional attorney had permission to share her impressions with “other EEOC personnel,” it is fair to assume that she communicated them to the office director — the person who actually signed the LOD. EEOC-Br. 59-60.

In responding, despite the references to EEOC’s “slim,” “thin,” or “meager” investigative files, Sterling does not deny that EEOC had enough information, as a whole, to support its reasonable cause finding. Rather, the company simply argues that nothing EEOC did or mentioned, individually, demonstrates that EEOC conducted a nationwide investigation. The arguments lack merit.

Sterling disputes EEOC’s “suggestion” that the class-related language in the charges and LOD established that EEOC’s investigation was sufficient. Sterling-Br.61-62. EEOC never “suggested” that it did. What the language would establish — had Sterling not already admitted this — is that the company was on notice that EEOC might bring a lawsuit containing nationwide claims. To support its argument, Sterling cites, for example, EEOC v. Reichhold Chemicals, 700 F.Supp. 524, 526-58 (N.D.Fla. 1988). Sterling-Br.61. That case provides that where a charge is narrow and an investigation uncovers other potential violations, a subsequent lawsuit — by EEOC or the individual — may encompass the claims uncovered in the investigation. While a correct statement of law, the principle is inapplicable here because the charges already alleged company-wide discrimination and, so, were as broad as EEOC’s investigation.

Sterling also argues that merely requesting “generic company policies” does not prove EEOC’s investigation was sufficient, citing, as authority, Outback, 520 F.Supp.2d at 1265-66. Sterling-Br.50, 51-52 (citing case twice for the same proposition). As noted above, however, Outback is a “scope-notice” case where the charges alleged discrimination in one region but EEOC’s suit alleged discrimination nationwide. The court held that “in [that] case,” because many company policies apply nationwide, “EEOC’s request for general company policies during an investigation that otherwise was focused on a single region did not provide Defendants with adequate notice that EEOC’s investigation was national in scope.” 520 F.Supp.2d at 1266. Since notice is not an issue in this case, and the charges alleged company-wide discrimination, Outback is inapposite.

Sterling then takes issue with EEOC’s argument that because other EEOC personnel were involved in the investigation, the district court erred in considering only what “investigators” — specifically, Ging — did during the investigation. Describing this argument as “frivolous,” Sterling asserts that EEOC cannot create an issue of fact by suggesting that “some unidentified individual at the EEOC might have conducted a nationwide investigation.” Sterling-Br.60.

That misses the point. EEOC did not have to “identify” a specific “individual” who “conducted a nationwide investigation,” nor does Sterling cite any authority holding that it did. EEOC’s investigation encompassed everything learned by any EEOC personnel associated with the case that shed light on whether there was reasonable cause to believe the allegations in the charges were true — including the substantial information EEOC obtained pursuant to the mediations. Ging was only part of that process. At a minimum, at least three other investigators were also involved (App-74), as were the office director (who signed the LOD) and the regional attorney.

Sterling then argues that EEOC cannot cite Dr.Lanier’s statistical analysis (which Sterling refers to as “mediation tables”) as evidence that its investigation was sufficient. Sterling-Br.52-55. The LOD mentions a “statistical analysis” which “reveals” that Sterling “promoted” and “compensated” “male employees at a statistically significant, higher rate than similarly situated female employees.” App-966. Dr.Lanier’s analysis matches that description, and EEOC had permission to put it in the investigative files. EEOC argued that the fact that Dr.Lanier’s analysis is the only statistical analysis in the files raises a strong inference that it and the one referred to in the LOD are the same.

Sterling, however, argues that because EEOC invoked the deliberative process privilege to prevent Ging, for example, from attesting that Dr.Lanier’s analysis “played any role in its LOD finding,” EEOC’s present argument is now barred. According to Sterling, under the sword/shield doctrine, “‘[a party] may not use [a] privilege to prejudice his opponent’s case or to disclose some selected communications for self-serving purposes.’” Sterling-Br.53(citing, e.g., U.S. v. Bilzerian, 926 F.2d 1285, 1292 (2d Cir. 1991)). In Sterling’s view, EEOC “seeks to do precisely what this Court forbids” because it is asking the court to “draw inferences of a nationwide investigation from the LOD but at the same time shielding Sterling and the District Court from evidence that purports to support those inferences.” Id.

This proposed convoluted extension of the sword/shield doctrine should be rejected. The sword/shield doctrine would apply if, having prevented Ging from answering Sterling’s questions, EEOC attempted to offer his affidavit attesting to the previously-undisclosed facts. See Bilzerian, 926 F.2d at 1292-93 (defendant could not testify that he relied on legal advice while asserting the attorney-client privilege to prevent exploration of his conversations with counsel).

That is not what happened here, nor did EEOC ask the Court “to draw inferences of a nationwide investigation.” EEOC simply noted that two undisputed facts — that Dr.Lanier’s statistical analysis was the only one in the investigative files and that the LOD cited a “statistical analysis” that makes findings consistent with those made by Dr.Lanier — coupled with Carlo’s testimony that cause determinations are generally based on “whatever we have in the files” (App-888)(v3), strongly suggest that the LOD referenced Dr.Lanier’s analysis. If this argument “prejudiced” Sterling, it was not because EEOC was relying on previously undisclosed information but because the inference is so strong. That is not the kind of prejudice Bilzerian addressed.

Along the same lines, Sterling then challenges EEOC’s argument that because the regional attorney participated in the mediations as EEOC’s representative, and the parties had agreed that she could share her impressions with other EEOC personnel, it was reasonable to infer that she did so with the office director, who signed the LOD. EEOC-Br.60. Sterling argues that because EEOC asserted privilege to prevent Ging and Carlo from answering questions regarding the “factual bases” for the LOD, EEOC cannot now rely on that “evidence” — presumably, the inference that the regional attorney might have communicated her impressions to the office director — to defeat summary judgment. Moreover, Sterling adds, because the regional attorney’s “purported communications” are not in the summary judgment record, EEOC cannot rely on them on appeal. Sterling-Br.55-57.

This argument fails for the same reason the previous argument fails. EEOC is not proffering previously undisclosed testimony of the regional attorney, Ging, or Carlo. The inference that the regional attorney discussed her impressions with the office director follows from the facts that the regional attorney attended the mediations where statistical evidence was explored, she was authorized to communicate that information to the office director, she and the director shared an interest in the case, and the director signed the LOD which referred to statistical evidence.[3] It is just that — an inference.

Sterling further argues that merely placing Dr.Lanier’s statistical analysis in the investigative files does not prove that EEOC’s investigation was sufficient. The company comments that there is no evidence EEOC “did anything” with Dr. Lanier’s statistical analysis besides placing it in the files. The company also comments that EEOC “played no role in [the analysis’s] creation” and did not hire Dr.Lanier as an expert itself until after issuing the LOD. In addition, though charging parties might have done so, Sterling never supplied EEOC with the underlying data, and there is no evidence EEOC independently verified the analysis or could have conducted the statistical analysis referenced in the LOD. Sterling-Br.51, 57-59.

Even assuming all of this is true, Sterling cites no authority holding that it matters, at least under the facts in this case. As EEOC explained, the mediation documents contemplate that, during the lengthy mediations, which EEOC attended, the parties would exchange and defend their competing expert analyses. EEOC-Br.13-14. Exercising its discretion over the “nature and extent” of the investigation, EEOC could reasonably find this give-and-take was a suitable substitute for any vetting by EEOC’s in-house expert at the investigative phase of the administrative process. By second-guessing EEOC’s choices, Sterling is simply challenging the sufficiency of the investigation and attempting to substitute its opinion of what EEOC should have done for what EEOC chose to do.

7. EEOC pointed out that shortly before moving for summary judgment on the ground that EEOC failed to prove it conducted a nationwide investigation, Sterling’s counsel asserted in open court that EEOC had “conducted a nationwide administrative investigation.” EEOC-Br.18 (citing App-132-33(v2)). Describing this as “frivolous,” Sterling opines that EEOC unearthed this nugget only after “desperately scouring” the record for “some form of admission” about the “purported investigation.” Sterling-Br.45. But, Sterling stresses, beginning with its Answer, it always challenged the “existence and scope of the investigation.” The depositions of Ging and Carlo “finally” “confirmed the lack of a nationwide investigation.” Sterling-Br-45-46 (citing counsel’s oral argument below).

Pointing out the statement was not “frivolous.” Sterling’s Answer contains an exhaustive list of 22 possible “affirmative defenses.” App-63-67. Paragraph 6, cited by Sterling, broadly challenges the administrative process, including the charge, conciliation, and cause-finding as well as the investigation even though Sterling admits that all of the other pre-suit requirements were satisfied. This broad language does not unequivocally establish that Sterling always doubted the sufficiency of the investigation.

Furthermore, Sterling knew little more after than before deposing Ging and Carlo. Although the company regularly asserts Ging and Carlo “admitted” that a nationwide investigation was not done, they actually almost always answered that they did not know or did not recall. See, e.g., Sterling-Br.15-17 (acknowledging that Ging “could not recall” any aspect of EEOC’s investigation). As for “desperately scouring” the record, EEOC’s appellate attorney reviewed the transcripts as well as pleadings and rulings to become familiar with the case. The statement was not hidden.

8. EEOC’s Opening Brief did not rely extensively on EEOC v. Mach Mining, 738 F.3d 171 (7th Cir. 2013), cert. granted, 134 S.Ct. 2872 (2014), which concerns EEOC’s pre-suit duty to conciliate and is thus relevant to but not dispositive of the issue in this case. After analyzing the “language of the statute, the lack of any meaningful standard for courts to apply, and the overall statutory scheme,” the Seventh Circuit held that any alleged failure to conciliate is not an affirmative defense to the merits of a discrimination suit. Id. at 173.

Sterling suggests that because Mach involves the duty to conciliate, not investigate, the Supreme Court’s decision “would have no impact on this case.” Sterling-Br.34 n.9. That position — based largely on differences between the conciliation and investigation duties and provisions — is in some tension with Sterling’s reliance on cases, such as Outback, Jillian’s, and CRST, that turn on notice and conciliation. If the statutorily-based distinctions are significant enough to make an affirmance in Mach irrelevant, those same cases are clearly not “authoritative” on the question in this case: whether EEOC’s alleged failure to sufficiently investigate alleged discrimination — where other pre-suit requirements, including conciliation, are satisfied — provides a basis for dismissing EEOC’s potentially meritorious Title VII suit.

Sterling also distinguishes Mach on the ground that allowing review of conciliation could “incentivize” employers to “sabotage” EEOC’s efforts to settle whereas EEOC’s subpoena power minimizes that concern in the context of investigations. Sterling-Br.34 n.9. This supposed distinction is illusory. As EEOC’s Opening Brief explained, subpoenas can add years to an investigation.[4] EEOC-Br.57-58. Stone-walling and foot-dragging are as attractive to employers during an investigation as during conciliation.

This case is Exhibit One. Despite asserting in open court that EEOC “conducted a nationwide administrative investigation” and admitting that EEOC satisfied its other pre-suit requirements, Sterling nevertheless peppered EEOC with discovery requests and, over objection, deposed two EEOC investigators. Learning little useful information about the investigation, Sterling nevertheless moved for summary judgment, alleging failure to investigate. And the court agreed, so EEOC’s potentially meritorious Title VII claims were never addressed. See, e.g., Keco, 748 F.2d 1100 (concluding that inquiries into the sufficiency of EEOC’s investigation are off-limits to the courts for just that reason). Thus, if, as Sterling contends, Mach’s relevance depends upon the extent to which the availability of judicial review would “incentivize” employers to game the system, the case would be squarely on point.

9. Finally, Sterling defends the district court’s dismissal order, suggesting that the ruling was a sanction for the “uniquely egregious circumstances of the case.” Sterling-Br.63. Accordingly, Sterling argues, this Court’s review is for abuse of discretion. Sterling-Br.62-63.

This argument is meritless. Sterling did not seek sanctions; it moved for summary judgment, and that is what the district court granted. Review is de novo. See, e.g., Price Trucking Corp. v. Norampac Indus., 748 F.3d 75, 79 (2d Cir. 2014).

Moreover, EEOC’s conduct here was fully consistent with Title VII. Sterling concedes that EEOC satisfied all “pre-suit” administrative requirements except the investigation — at one point, Sterling even admitted that EEOC “conducted a nationwide administrative investigation.” It is undisputed that EEOC has discretion to decide how and how much to investigate because courts will not review the sufficiency of the investigation. The district court here acknowledged that, at a minimum, EEOC investigated the nineteen charges from across the country, most alleging class, company-wide, and/or pattern-or-practice discrimination. That should be the end of the matter.

Sterling supports its argument with string-cites of the same cases cited throughout Sterling’s brief. See Sterling-Br.63-64 (citing cases). As noted above and in EEOC’s Opening Brief, those cases turn on notice and conciliation, not the investigation (unless the investigation and the LOD and conciliation failed to provide notice of the scope of EEOC’s potential/actual claim). Notice and conciliation are not at issue here.

In addition, Sterling does not deny that EEOC had sufficient information — gleaned through the mediations as well as from materials supplied by Sterling and the charging parties — to make a reasonable cause finding in this case. And that is, after all, the purpose of an investigation. Sterling’s arguments elevate form over substance.

Sterling argues that unless this Court affirms the dismissal of EEOC’s case on this “uniquely” and “extraordinarily” “egregious” record, EEOC “will have no incentive ever to perform its statutory duties in good faith, thus undermining the carefully calibrated statutory scheme Congress created when it enacted Title VII.” Sterling-Br.63. This concern is unwarranted.

As Sterling itself concedes, the administrative “scheme” is an “integrated, multistep process.” Sterling-Br.32. The main purpose of the investigation is to determine whether there is reasonable cause to believe the allegations in the charge are true. Once it gathers enough information to make that determination, EEOC can decide that the investigative step is over; the process moves on to other steps. “Scorched earth” may be a tactic some practitioners use in litigation but it is inappropriate in an EEOC investigation.

Moreover, courts have long rejected arguments that even a shoddy EEOC investigation — not an issue here — causes any cognizable harm. See, e.g., Georator, 592 F.2d at 768-69; Gibson, 579 F.2d at 891. Sterling ignores this line of cases, but the holding in those cases has not caused EEOC to stop “perform[ing] its statutory duties in good faith.” Nor would a reversal have that effect, especially under the “circumstances” in this case.

On the other hand, an affirmance would provide a perverse incentive — to employers. Courts consistently recognize that allowing employers to delve into the sufficiency of EEOC’s investigation distracts the courts and parties from the “main purpose of the litigation: to determine whether [the employer] has violated Title VII.” Keco, 748 F.2d at 1100.

The “circumstances” here are “extraordinary” and “unique” only because no court has ever held that EEOC’s potentially meritorious Title VII suit could properly be dismissed for failure to investigate — despite a finding that EEOC investigated the nineteen charges, most alleging wide-spread discrimination; despite Sterling’s concessions that it had ample notice of the potential scope of EEOC’s claims and that EEOC satisfied all its other pre-suit requirements including conciliation; and despite Sterling’s admission that “EEOC conducted a nationwide administrative investigation.” Under these circumstances, an affirmance would practically guarantee that employers in the future would challenge EEOC’s investigation in every case.

EEOC therefore urges this Court to reject Sterling’s arguments and hold that EEOC satisfied its pre-suit administrative duty to investigate.

CONCLUSION

The judgment should be reversed and the case remanded for consideration of the merits of EEOC’s discrimination claim.

Respectfully submitted,

P. DAVID LOPEZ

General Counsel

CAROLYN L. WHEELER

Acting Associate General Counsel

JENNIFER S. GOLDSTEIN

Acting Assistant General Counsel

/s/ Barbara L. Sloan_____________

BARBARA L. SLOAN

Attorney

EQUAL EMPLOYMENT

OPPORTUNITY COMMISSION

Office of General Counsel

131 M Street N.E., 5th Floor

Washington, DC 20507

202-663-4721

barbara.sloan@eeoc.gov

[1] EEOC agrees it did not — and normally would not — do an “onsite” investigation. See Sterling-Br.3. EEOC might, for example, seek to enter an employer’s premises to visually assess the feasibility of possible accommodations under the Americans with Disabilities Act. But visiting a jewelry store was unlikely to yield useful information for these pay and promotion claims.

[2] Sterling also cites a 2010 pleading where EEOC’s then-trial attorney said the files contain “little investigative material.” Sterling-Br.4, 26; cf. EEAC amicus brief at 6 (mistakenly asserting statement was made in opposing summary judgment). Whatever that attorney meant, it does not decrease the size of the files.

[3] Here, as throughout its brief, Sterling discusses arguments EEOC supposedly made below. See, e.g., Sterling-Br.56. But only arguments in EEOC’s appellate brief are relevant to the appeal, Any other arguments, and Sterling’s responses to them, are irrelevant.

[4] Two groups of amici filed briefs in support of Sterling. Their arguments are generic, however, focusing mainly on conciliation and political issues, which are not at issue here. Nor does it appear that they are familiar with the facts of this case or EEOC’s brief. For example, they refer to “individual” charges, although most charges here are class, and do not acknowledge that Sterling conceded that EEOC satisfied its duty to conciliate. See also http://www.eeoc.gov/eeoc/legislative/report_response_final.cfm

Curiously, like Sterling, Amici also reference EEOC’s subpoena power, faulting the Commission for failing to issue a subpoena here. Yet, they previously filed a joint amicus brief in an EEOC case where the district court had enforced a subpoena seeking class-wide information. Urging reversal, Amici argued that the court of appeals should limit EEOC’s subpoena authority. See Brief of EEAC & Chamber of Commerce as Amici Curiae in EEOC v. Schwan’s Home Servs., No.10-3012 (8th Cir. filed 11/10/2010). Despite Amici’s argument, the Eighth Circuit affirmed the lower court’s decision. 644 F.3d 742 (8th Cir. 2011).