____________________________________________ IN THE UNITED STATES COURT OF APPEALS FOR THE EIGHTH CIRCUIT ____________________________________________ No. 07-1559 ____________________________________________ EQUAL EMPLOYMENT OPPORTUNITY COMMISSION, Plaintiff-Appellee, v. ALLSTATE INSURANCE CO., Defendant-Appellant. ____________________________________________ On Appeal from the United States District Court for the Eastern District of Missouri ____________________________________________ BRIEF OF THE EQUAL EMPLOYMENT OPPORTUNITY COMMISSION AS APPELLEE ____________________________________________ RONALD S. COOPER General Counsel CAROLYN L. WHEELER Acting Associate General Counsel VINCENT J. BLACKWOOD Assistant General Counsel JULIE L. GANTZ Attorney EQUAL EMPLOYMENT OPPORTUNITY COMMISSION Office of General Counsel 1801 L Street, N.W. Washington, DC 20507 (202) 663-4718 SUMMARY OF THE CASE AND REQUEST FOR ORAL ARGUMENT In June 2000, Allstate Insurance Company terminated all of its employee- agents. It is undisputed that the average age of the employee-agents at the time of termination was significantly higher than the average age of Allstate's other employees. Although Allstate's normal policy is to permit terminated employees to apply for other jobs with the company unless they were fired for poor performance or misconduct, Allstate adopted a policy barring terminated agents from employment with Allstate for up to two years. The Equal Employment Opportunity Commission alleges in this action that the moratorium on rehiring terminated agents violated the Age Discrimination in Employment Act of 1967 ("ADEA"), 29 U.S.C. § 621 et seq., because it had a disparate impact based on age and was not based on a reasonable factor other than age. The district court granted the Commission's motion for partial summary judgment, ruling that the case was cognizable under a disparate impact theory and that the statistical evidence established that the policy had a disparate impact. The court denied the defendant's motion for summary judgment on the question of whether the moratorium was based on reasonable factors other than age, holding that this issue should be decided by a jury. The district court subsequently granted Allstate's motion to certify the following questions for interlocutory appeal pursuant to 28 U.S.C. § 1292(b): (1) whether a claim that a policy has a disparate impact on former employees based on age is cognizable under § 4(a)(2) of the ADEA; and (2) whether the method of determining a disparate impact used by the Commission and adopted by the district court is inadequate "as a matter of law" to establish a disparate impact. This Court granted Allstate's petition for interlocutory review of these claims. We believe that oral argument of twenty minutes per side would assist the Court in resolving these issues. TABLE OF CONTENTS SUMMARY OF THE CASE AND REQUEST FOR ORAL ARGUMENT . . . . i TABLE OF CONTENTS. . . . . . . . . . . . . . . . . . . .iii TABLE OF AUTHORITIES. . . . . . . . . . . . . . . . . . . v STATEMENT OF JURISDICTION. . . . . . . . . . . . . . . . 1 STATEMENT OF THE ISSUES AND APPOSITE CASES. . . . . . . . 1 STATEMENT OF THE CASE. . . . . . . . . . . . . . . . . . .2 A. Proceedings Below. . . . . . . . . . . . . . . . . . 2 B. Statement of Facts. . . . . . . . . . . . . . . . . . 4 C. The District Court Decision. . . . . . . . . . . . 7 STANDARD OF REVIEW. . . . . . . . . . . . . . . . . . . 10 SUMMARY OF ARGUMENT. . . . . . . . . . . . . . . . . . 11 ARGUMENT I. THE COMMISSION'S CHALLENGE TO ALLSTATE'S POLICY RESTRICTING THE REHIRE OF FORMER-EMPLOYEE AGENTS STATES A CLAIM UNDER § 4(A)(2) OF THE ADEA BECAUSE THE FORMER EMPLOYEE-AGENTS WERE CLASSIFIED AS INELIGIBLE FOR REHIRE PRECISELY BECAUSE THEY HAD BEEN ALLSTATE EMPLOYEES. . . . . . . . . . . . . . . . 14 II. EVIDENCE THAT THE EMPLOYEE-AGENTS WHO WERE CLASSIFIED AS INELIGIBLE FOR REHIRE WERE SIGNIFICANTLY OLDER THAN ALLSTATE'S OTHER EMPLOYEES WHO WERE NOT SUBJECT TO A RESTRICTION ON REHIRE AFTER TERMINATION ESTABLISHES THAT THE POLICY HAD A DISPARATE IMPACT ON THE BASIS OF AGE. . 20 CONCLUSION. . . . . . . . . . . . . . . . . . . . . . 32 CERTIFICATE OF COMPLIANCE CERTIFICATE OF SERVICE TABLE OF AUTHORITIES Abbott v. Federal Forge, 912 F.2d 867 (6th Cir. 1990). . . . . . . . . . 28, 29 Anderson Liberty Lobby, Inc., 477 U.S. 242 (1986). . . . . . . . . . . . . . 11 Banks v. East Baton Rouge Parish Sch. Bd., 320 F.3d 570 (5th Cir. 2003). . 27 Brown v. M & M/Mars, 883 F.2d 505 (7th Cir. 1989). . . . . . . . . . . . . . 27 Dothard v. Rawlinson, 433 U.S. 321 (1977). . . . . . . . . . . . . . . . 30-31 Green v. Missouri Pac. R.R. Co., 523 F.2d 1290 (8th Cir. 1975). . . . . . 27 Griggs v. Duke Power Co., 401 U.S. 424 (1971). . . . . . . . . . . . 15, 18, 31 Hameed v. International Assoc., 637 F.2d 506 (8th Cir. 1980). . . . . . . . .27 Henson v. Liggett Group, Inc., 61 F.3d 270 (4th Cir. 1995). . . . . . . . 27 Hishon v. King & Spalding, 467 U.S. 69 (1984). . . . . . . . . . . . . . . . 26 Keathley v. Ameritech Corp., 187 F.3d 915 (8th Cir. 1999). . . . . . . . . . 10 Leftwich v. Harris-Stowe State Coll., 702 F.2d 686 (8th Cir. 1983). . . .13, 23 Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574 (1986). . 11 Ortega v. Safeway Stores, Inc., 943 F.2d 1230 (10th Cir. 1991). . . . . . . .29 Robinson v. City of Dallas, 514 F.2d 1271 (5th Cir. 1975). . . . . . . . . . 27 Robinson v. Shell Oil Co., 519 U.S. 337 (1997). . . . . . . . . . 12, 15-19 Smith v. City of Jackson, 544 U.S. 228 (2005). . . . . . . . 3, 14, 18-19, 20 STATUTES 28 U.S.C. §§ 451, 1331, 1337, 1343, 1345. . . . . . . . . . 1 28 U.S.C. § 1292(b). . . . . . . . . . . . . . . . . . . 1, 3 29 U.S.C. § 621. . . . . . . . . . . . . . . . . . . . . . 1 29 U.S.C. § 623(a)(2). . . . . . . . . . . . . . . . passim 29 U.S.C. § 630(c). . . . . . . . . . . . . . . . . . . . 17 29 U.S.C. § 633a(b). . . . . . . . . . . . . . . . . . . . 17 REGULATIONS AND LEGISLATIVE HISTORY 29 C.F.R. § 1625.22(f)(3)(i)(B). . . . . . . . . . . . . . 25 Equal Employment Opportunity Act of 1972, Pub. L. No. 92-261, § 8(a), 86 Stat. 103 (1972). . . . . . 15 Conf. Rep. on H.R. 1746, reprinted in 118 Cong. Rec. 7166, 7169 § 8(a)-(b) (1972).15 STATEMENT OF JURISDICTION This is an enforcement action brought by the Equal Employment Opportunity Commission pursuant to § 7(b) of the Age Discrimination in Employment Act of 1967 ("ADEA"), 29 U.S.C. § 621 et seq. The district court had jurisdiction under 28 U.S.C. §§ 451, 1331, 1337, 1343, and 1345. The district court certified that its interlocutory order granting the plaintiff partial summary judgment presents controlling issues of law about which there exist substantial grounds for difference of opinion and resolution of the questions will materially advance the ultimate termination of the litigation. Joint Appendix Volume II ("II JA") 466. This Court granted Allstate's petition for interlocutory review on March 14, 2007. R. 80. Accordingly, this Court has jurisdiction under 28 U.S.C. § 1292(b). STATEMENT OF THE ISSUES AND APPOSITE CASES 1. Whether § 4(a)(2) of the ADEA prohibits an employer from classifying former employees in a way that adversely affects their employment opportunities based on age. Robinson v. Shell Oil Co., 519 U.S. 337 (1997) 2. Whether undisputed evidence that the employee-agents subject to the policy restricting rehire were significantly older than the other Allstate employees who were subject to no similar restriction on rehire after termination establishes that the policy has a disparate impact based on age. Leftwich v. Harris-Stowe State Coll., 702 F.2d 686 (8th Cir. 1983) STATEMENT OF THE CASE A. Proceedings Below This is an interlocutory appeal from the district court's order of October 19, 2006, granting partial summary judgment to the plaintiff and denying the defendant's motion for summary judgment. The Commission initiated this lawsuit on October 7, 2004, by filing a complaint alleging that the defendant violated the ADEA by implementing a policy barring a group of terminated employee-agents from rehire for a period of one to two years. The complaint alleged, inter alia, that the policy had a disparate impact based on age and was not based on a reasonable factor other than age. I JA 3. The Commission alleged that the policy resulted in the defendant's refusal to hire applicants who were former employee-agents and deterred former employee-agents from applying for non-agent positions with the company. Id. Based on the undisputed evidence that the hiring restriction applied only to a subset of Allstate employees who were significantly older than the rest of the workforce, the Commission moved for summary judgment on the question of whether the policy had a disparate impact based on age. R. 42. The defendant also moved for summary judgment on liability. R. 53. The company argued that the Commission's claim is not cognizable under the ADEA because it amounts to a disparate impact challenge to a "hiring" policy. According to Allstate, the Supreme Court's decision in Smith v. City of Jackson, 544 U.S. 228, 238 (2005), establishes that ADEA disparate impact claims may only be brought under § 4(a)(2) of the ADEA, and, since § 4(a)(2) refers only to "employees," not "applicants," it may not be used to challenge discrimination in hiring. Allstate also faulted the Commission's statistical analysis because it does not comport with the analysis used in other disparate impact hiring cases. The district court granted the Commission's motion for partial summary judgment, ruling that the case was cognizable under a disparate impact theory and that the statistical evidence established that the policy had a disparate impact. The court denied the defendant's motion for summary judgment on the question of whether the hiring moratorium was based on reasonable factors other than age, holding that this issue should be decided by a jury. The district court subsequently granted Allstate's motion to certify the following questions for interlocutory appeal pursuant to 28 U.S.C. § 1292(b): (1) whether a claim that a policy has a disparate impact on former employees based on age is cognizable under § 4(a)(2) of the ADEA; and (2) whether the method of determining a disparate impact used by the Commission and adopted by the district court is inadequate as a matter of law to establish a disparate impact. This Court granted Allstate's petition for permission to pursue interlocutory review of these questions. B. Statement of Facts On November 10, 1999, Allstate announced that it was implementing the Preparing for the Future Group Reorganization Program. II JA 213 (Jossund Decl. 10). Pursuant to the program, all of Allstate's 6,329 employee-agents would be terminated on June 30, 2000. II JA 213-14 (Jossund Decl. 12). The terminated employee-agents were given the option of converting to independent contractor status and continuing to work as Allstate agents. II JA 214; I JA 50-53; 64-69. Those who did not wish to become independent contractors were offered severance benefits. Id. The employee-agents were given until June 1, 2000, to select an option. I JA 27 (Guthrie Decl. 4(1)). More than 90% of the terminated employee-agents were age 40 or older. I JA 20. The average age of the employee-agents terminated under the program was 51.1 years, while the average age of the rest of the Allstate workforce was 39.4 years. I JA 21. Furthermore, 23 percent of Allstate's age 40 or older employees were terminated pursuant to the program, whereas only 2.7 percent of employees under 40 were terminated. I JA 20. In addition to terminating the employee-agents, Allstate adopted a policy prohibiting their rehire in other jobs with the company for up to two years. Ordinarily, Allstate does not limit the rehiring of former employees unless they were fired for poor performance or misconduct, or are retired from Allstate. II JA 315-17; I JA 21, 134, 248). There is conflicting evidence in the record as to when the restriction on rehire of terminated employee-agents was first adopted. Karleen Zuzich, Allstate's Assistant Vice President for Human Resources, testified that she devised the rehire moratorium in reaction to a human resources manager's question about reemployment with Allstate. I JA 124. Zuzich testified that she first provided a copy of the policy to human resources personnel on September 26, 2000, after the employee-agents were terminated. I JA 137. However, the document she produced is described as a "clarification" of the re-hire policy, implying that the policy already existed.<1> Furthermore, there is evidence that former employee-agents were informed of the moratorium while they were still employed by Allstate. Carmon Green testified that he was told in January 2000 that he would not be eligible for hire with Allstate in a non-agent capacity. II JA 339, 341. Former employee-agent Larry Finley testified that at the first meeting he attended on the Reorganization Program in November or December 1999, he was informed by an Allstate official that he would not be allowed to transfer to a claims adjustment position, a move he contemplated to save his pension. II JA 366-70, 372. The written re-hire policy states that the policy was adopted "[t]o avoid any potential violations of the non-compete provision." See Rehire policy (filed under seal). Zuzich, the author of the policy, did not explain how an employee-agent who has taken a non-agent job with Allstate could violate the non-compete provision of his former employment contract. I JA 141. In fact, Zuzich testified that she did not review the non-compete provisions of the employee-agents' contracts. Id. At her deposition, Zuzich articulated several other reasons for the policy: (1) to avoid a situation in which customers assigned to new agents encountered former agents working in another capacity for the company; (2) to encourage the former employee-agents to elect to become independent contractor agents with Allstate; (3) to prevent the former employee-agents from receiving severance benefits at the same time they were receiving salaries; (4) to avoid placing former employee-agents who had rejected the offer to become independent contractors in close proximity to other Allstate employees; and (5) to use a policy consistent with the rehiring policy implemented in a particular retirement program known as the Special Retirement Opportunity (SRO) which the company had used in the past. I JA 125-26, 131. C. The District Court's Decision On October 19, 2006, the district court granted the EEOC's motion for partial summary judgment and denied Allstate's motion for summary judgment. The court accepted Allstate's premise that the disparate impact theory may not be applied to a hiring policy under the ADEA. II JA 447 ("[A] disparate impact hiring case . . . is no longer cognizable [under the ADEA] after City of Jackson."). However, the court concluded that, based on the evidence in this case, "the most logical approach is to treat the rehire policy as part of [Allstate's] Reorganization Plan which affected all employee-agents." II JA 448. The court reasoned that "[t]he legality of the Reorganization Plan is not before this court, but the effect of that reorganization was to require the former agents to either become independent contractors, or to seek other employment. The Defendants then implemented a rehire policy which barred this same group from reapplying for positions within the company." II JA 449. The court agreed with the Commission that the action was properly litigated as a challenge to a policy that affects the status of members of the protected class as employees. II JA 449-50. "When an employee is terminated, the court treats this as a term of employment, analyzed under § 623(a)(2). The current case is no different, the employees were terminated, and then unable to apply for an alternate position." II JA 453. Consequently, the district court rejected Allstate's argument that this case constitutes a "hiring" case that cannot be brought as a disparate impact claim under § 4(a)(2). Id. Because the court did not consider the action a challenge to a hiring policy, it also rejected Allstate's argument that the method of establishing a disparate impact on older workers necessarily had to assess the effect of the rehire moratorium on the general applicant population. According to the court, "[a]pplicants from the general population would not be subject to the rehire policy at issue, because they were not former employees." II JA 447; see also II JA 452. Instead, because "the Reorganization Plan [is] inextricable from the rehire policy, the appropriate analysis is that suggested by the Plaintiff, comparing the total number of Allstate employees within the protected group that were affected by the policy, with the number of Allstate employees not in the protected group that were affected by the policy." II JA 453. The court noted that the average age of employees subject to the rehire moratorium was 51.1 years while the average age of the rest of the Allstate workforce was 39.4 years. Id. Accordingly, the district court granted the Commission's motion for partial summary judgment, holding that the Commission had established as a matter of law that the rehire moratorium had a disparate impact on the former employee-agents aged 40 and older who had been terminated under the Reorganization Program. Id. The court denied Allstate's motion for summary judgment on the reasonableness of the policy, holding that the issue should be determined by a jury. The court recounted Zuzich's testimony stating that she did not inquire as to whether any of the concerns she articulated with rehiring former employee-agents had occurred with those agents who had been rehired prior to the implementation of the policy. II JA 454-55. "Zuzich further testified that she did not research whether the rehire policy would be beneficial to the Defendant, she did not recall reviewing any documents associated with the Reorganization Plan, nor did she make any effort to determine the age demographics of the former employee-agents who would be subject to the rehire policy." Id. at 455. The court pointed out that the Commission refuted several of Allstate's justifications for the re-hire policy. One of Allstate's rationalizations for the re- hire policy was to encourage the employee-agents to become independent contractor agents. The court noted, however, that the employer-agents were required to decide whether to become independent contractors by June 1, 2000, before Zuzich says she devised the re-hire policy. Accordingly, the court stated, "the Defendant could not have intended the rehire policy to encourage enrollment in the independent contractor program, because the date for making the decision had passed." II JA 455. The court also concluded that the Commission refuted Allstate's stated concern about "double-dipping" "by pointing to Ms. Zuzich's testimony that the Defendant had a standard policy in place that would prevent double-dipping. This policy required any individual who was rehired by the defendant, and who was also receiving severance benefits, to repay the unused portion of the severance benefit." Id. Because of these inconsistencies in Allstate's asserted justification for the rehire policy, the court held, "[a] reasonable jury could find for the Plaintiff that the Plaintiff has provided sufficient evidence that the proffered reasons are unreasonable." II JA 455. Accordingly, the court denied Allstate's motion for summary judgment. STANDARD OF REVIEW A district court's grant of summary judgment is reviewed de novo applying the same standard as the district court. Keathley v. Ameritech Corp., 187 F.3d 915, 919 (8th Cir. 1999). In determining whether summary judgment was appropriate, this Court assesses the facts in the light most favorable to the non- moving party, drawing all reasonable inferences in its favor. Anderson Liberty Lobby, Inc., 477 U.S. 242, 255 (1986); Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 587 (1986). The district court's decision granting summary judgment to the Commission should be affirmed upon a finding that the Commission established as a matter of law that the specific policy challenged had a disparate impact because of age. SUMMARY OF ARGUMENT In 2000, in conjunction with its Preparing for the Future Reorganization Program, Allstate decided to terminate all of its employee-agents and to classify the terminated agents as temporarily ineligible for rehire into other positions with the company. It is undisputed that, at the time the policies were implemented, the employee-agents were, on average, significantly older than Allstate's other employees who generally remained eligible for rehire in the event of termination. The district court correctly ruled that Allstate's classification of the terminated employee-agents as ineligible for rehire was subject to challenge under § 4(a)(2) of the ADEA, and had a disparate impact based on age. Section 4(a)(2) makes it unlawful for an employer "to limit, segregate, or classify his employees in any way which would deprive or tend to deprive any individual of employment opportunities or otherwise adversely affect his status as an employee, because of such individual's age." The district court correctly recognized that Allstate's decision to classify employee-agents as ineligible for rehire deprived them of employment opportunities because of their status as terminated employee-agents, and, therefore, may violate § 4(a)(2) if status as an employee-agent is correlated with age. Allstate asserts that the rehiring policy is a separate "hiring" policy unrelated to the Reorganization Program, and, as a "hiring" policy, it may not be challenged under § 4(a)(2). In light of the fact that the written rehire policy is characterized as part of the Reorganization Program, the district court properly rejected this assertion. Allstate also argues that the policy may not be challenged under § 4(a)(2) because the agents were no longer employees when the policy was implemented. However, the company's premise that § 4(a)(2) applies only to discrimination against current employees cannot be reconciled with the Supreme Court's decision in Robinson v. Shell Oil Co., 519 U.S. 337, 346 (1997). Robinson, which held that the term "employee" in Title VII's retaliation provision includes former employees, is dispositive because, as in Title VII, the definition of "employee" in the ADEA "likewise lacks any temporal qualifier and is consistent with either current or past employment." Id. at 341-42. Accordingly, the statutory analysis on which the Supreme Court based its decision applies with equal force to the question before the Court in this case. Contrary to Allstate's argument, the policy justifications for the broader reading of "employees" in Robinson justify the same reading of the term in the ADEA. If the term "employee" were interpreted to mean "current employee," it would create a statutory void for those like the former employee-agents who were denied an employment opportunity not directly because of their age, but because of a characteristic related to age-here, status as a terminated agent. The district court correctly construed the statute to avoid a situation in which an employer could avoid liability by waiting until the day after employees were terminated to implement a policy that adversely affected their employment opportunities based on age. The district court's ruling that the rehire policy had a disparate impact based on age was correct. It is undisputed that the employee-agents were significantly older than Allstate's other employees. The method used by the Commission to show that the policy had a disparate impact-comparing the average age of employees subject to the policy with the average age of other employees not subject to the policy-is the method approved by this Court in Leftwich v. Harris- Stowe State Coll., 702 F.2d 686, 690 (8th Cir. 1983). The district court properly rejected Allstate's argument that this is a "hiring" case requiring evidence of a disparity between the average age of the employee-agents and the average age of either the labor market or actual applicants for jobs at Allstate. Because the rehire policy in this case by definition applies only to Allstate employees, it makes no sense to analyze the age of potential or actual outside applicants. ARGUMENT I. THE COMMISSION'S CHALLENGE TO ALLSTATE'S POLICY RESTRICTING THE REHIRE OF FORMER-EMPLOYEE AGENTS STATES A CLAIM UNDER § 4(A)(2) OF THE ADEA BECAUSE THE FORMER EMPLOYEE-AGENTS WERE CLASSIFIED AS INELIGIBLE FOR REHIRE PRECISELY BECAUSE THEY HAD BEEN ALLSTATE EMPLOYEES. The district court correctly held that the Commission's challenge to Allstate's policy restricting the rehire of terminated employee-agents states a claim under § 4(a)(2) of the ADEA. Section 4(a)(2) makes it unlawful for an employer "to limit, segregate, or classify his employees in any way which would deprive or tend to deprive any individual of employment opportunities or otherwise adversely affect his status as an employee, because of such individual's age." 29 U.S.C. § 623(a)(2). The Supreme Court held in Smith v. City of Jackson, 544 U.S. 228, 240 (2005), that this provision outlaws neutral practices with a disparate impact based on age. Accordingly, as the district court held, the Commission's allegation that Allstate's decision to classify all employee-agents as ineligible for rehire deprived them of employment opportunities because of their age states a claim under § 4(a)(2). On appeal, Allstate argues that the court erred because § 4(a)(2) prohibits only discrimination against "employees," and the former employee-agents affected by the challenged policy "are by definition not 'employees.'" Def. Br. at 19. Even accepting Allstate's premise that § 4(a)(2) does not prohibit discrimination against applicants for employment who have not previously worked for the company,<2> the argument that it does not apply to employees once their employment ends is manifestly wrong. The ADEA, like Title VII, defines "employee" as "an individual employed by any employer." 29 U.S.C. § 630(f); cf. 42 U.S.C. § 2000e(f) (for purposes of Title VII, "employee" means "an individual employed by an employer"). In Robinson v. Shell Oil Co., 519 U.S. 337, 345 (1997), the Supreme Court held that the term "employee" in Title VII's retaliation provision includes former employees. The holding in Robinson applies with equal force to the question before the Court in this case. In Robinson, the court of appeals had held that the plaintiff's allegation that he was retaliated against after he left his employment with Shell Oil did not state a claim under § 704(a) of Title VII because, as a former employee, Robinson was not an "employee" when the company retaliated against him. The Supreme Court reversed, noting that the term "employees" as used in Title VII's retaliation provision is "ambiguous as to whether it excludes former employees" because "there is no temporal qualifier in the statute such as would make plain that § 704(a) protects only persons still employed at the time of the retaliation." 519 U.S. at 341. Because Title VII's definition of "employee" "likewise lacks any temporal qualifier and is consistent with either current or past employment," the Court noted that the term is not limited to those having an existing employment relationship with the employer in question. Id. at 342. The Court stated that the word "employed"-in the definition of an "employee" as "an individual employed by an employer"-"is not so limited in its possible meanings, and could just as easily be read to mean 'was employed.'" Id. (emphasis in the original). The Court noted that a number of other Title VII provisions use the term "employee" to mean something other than "current employee." Id. Accordingly, the Court concluded that "the term standing alone is necessarily ambiguous and each section must be analyzed to determine whether the context gives the term a further meaning that would resolve the issue in dispute." Id. at 343-44. The same statutory analysis would apply in this case. As noted, supra at 15, the ADEA's definition of employee is the same as Title VII's, and, therefore, similarly lacks a temporal qualifier. Furthermore, as in Title VII, the term "employees" appears elsewhere in the ADEA where the term clearly means something broader than current employees. For example, 29 U.S.C. § 633a(b) authorizes enforcement efforts through appropriate remedies, such as "reinstatement or hiring of employees with or without backpay." Current employees would not be reinstated or hired. And in 29 U.S.C. § 630(c), the definition of "employment agency" is "any person regularly undertaking with or without compensation to procure employees for an employer." The Supreme Court stated in Robinson that the same language in Title VII "most naturally is read to mean 'prospective employees.'" 519 U.S. at 343 n.3. Notwithstanding the Supreme Court's conclusion that the same statutory language in Title VII could encompass former employees, Allstate asserts that § 4(a)(2) is "clear and unambiguous." The company argues that Robinson's holding that the term "employees" in § 704 of Title VII includes former employees was merely a policy decision by the Court designed "solely to confer standing to sue on former employees claiming retaliation." Def. Br. at 25. In the defendant's view, the Supreme Court's holding was based on the need to compensate for a statutory "gap" which deprived individuals who filed charges of protection from retaliation once their employment ended. According to Allstate, no such gap is present here because, even if former employees are foreclosed from bringing disparate impact challenges under § 4(a)(2), they may still bring disparate treatment claims under § 4(a)(1). Def. Br. at 26. This argument misses the point. Whether or not former employees are protected from disparate treatment, i.e., adverse actions based directly on their age, Allstate's argument would strip them of protection from disparate impact, i.e., adverse actions based on a neutral factor correlated to age. Accordingly, if Allstate's argument were accepted it would also create a statutory "gap" for those like the former employee-agents who were denied an employment opportunity not directly because of their age, but because of a characteristic related to age-here, status as a terminated agent. Allstate's argument that protection from disparate treatment is sufficient to satisfy the purposes of the ADEA with respect to former employees is merely a variation of the argument rejected by the Supreme Court in Smith. As the Court noted in Smith, 544 U.S. at 235 n.5, the considerations that led the Court to hold that Title VII prohibits neutral practices with a disparate impact in Griggs v. Duke Power Co., 401 U.S. 424 (1971), also played a role in congressional deliberations on enactment of the ADEA. Accordingly, the Court rejected the argument that construing the ADEA as limited to intentional discrimination based directly on age would satisfy the purposes of the statute. Smith, 544 U.S. at 235-37. For the same reason, this Court should reject Allstate's argument that construing § 4(a)(2) to exclude former employees from protection from neutral practices with a disparate impact would not create a "gap" in coverage comparable to that created by the court of appeals' similar construction of § 704(a) in Robinson. The Robinson Court concluded that the need for "maintaining unfettered access to statutory remedial mechanisms" supported an inclusive interpretation of "employees" in § 704(a). 519 U.S. at 346. The same policy concern applies with equal force in this case. In Robinson, the plaintiff alleged that his former employer violated Title VII when it gave him a negative job reference in retaliation for his filing an EEOC charge following his termination. The Court noted that the alleged retaliation grew directly out of the plaintiff's employment with the defendant, and construing Title VII to put this conduct beyond the reach of the courts would undermine the purpose of the statute. Similarly, the alleged discriminatory policy in this case was directly connected to the fact that the former employee-agents were employed by Allstate. Despite Allstate's attempts to characterize the rehiring policy as a separate policy unrelated to the Reorganization Program, it is clear that it was a part of that program. The written policy is entitled: "Preparing for the Future Reorganization Program (Clarification of Re-Hire Policy)." Accordingly, the district court correctly considered the rehire policy as simply another aspect of the company's treatment of its employee-agents. In effect the company decided to do two things to the employee-agents: first, it terminated their employment; then, it restricted their ability to be rehired into other positions. Allstate is now arguing that the second aspect of the program is immune from challenge under the ADEA even if it had a disparate impact because the company waited until after the agents had been terminated to implement it. Under Allstate's construction of the statute, an employer could evade liability by waiting until the day after employees were terminated to implement a policy that adversely affected their employment opportunities based on age. The district court correctly construed the statute to avoid this result. II. EVIDENCE THAT THE EMPLOYEE-AGENTS WHO WERE CLASSIFIED AS INELIGIBLE FOR REHIRE WERE SIGNIFICANTLY OLDER THAN ALLSTATE'S OTHER EMPLOYEES WHO WERE NOT SUBJECT TO A RESTRICTION ON REHIRE AFTER TERMINATION ESTABLISHES THAT THE POLICY HAD A DISPARATE IMPACT ON THE BASIS OF AGE. In a disparate impact case under the ADEA, the plaintiff must establish that a specific employment practice or policy caused a significant disparate impact on a group of older individuals. See Smith, 544 U.S. at 241. In this case, the Commission identified as causing an impact Allstate's policy of classifying employee-agents terminated pursuant to the Reorganization Program as temporarily ineligible for rehire, while generally permitting other terminated employees to be immediately rehired. To show that this policy had a disparate impact, the Commission offered evidence that the average age of the employee- agents classified as ineligible for rehire was 51.1 years, while the average age of the rest of the Allstate workforce was 39.4 years. I JA 21. Furthermore, 23 percent of Allstate's workforce aged 40 or older were classified as ineligible for rehire under the policy, whereas only 2.7 percent of employees under 40 were so classified. I JA 20. Allstate does not dispute the accuracy of these statistics, as the district court noted. II JA 453. Instead, the company argues that this is a "hiring" case, and, therefore, by analogy to other cases alleging disparate impact in hiring, the Commission must show a disparity between the average age of the employee- agents and the average age of either the labor market or actual applicants for jobs at Allstate. Def. Br. at 19-20, 30-32, 35-38. The district court correctly rejected this argument. As the court recognized, the rehire policy is not an independent hiring policy adopted by Allstate, but rather is "part of the Reorganization Plan which [a]ffected all employee-agents." II JA 448. Pursuant to that plan, Allstate identified a category of employees-employee-agents-and took two related actions with respect to that group: first the company terminated their employment, and then it restricted their eligibility for rehire. Remarkably, Allstate argues that the district court ignored the record evidence in characterizing the two actions as "inextricably linked." Def. Br. at 34. It makes this argument notwithstanding the fact that its own document setting forth the rehire policy is entitled "Preparing for the Future Reorganization Program (Clarification of Re-Hire Policy)." Rehire Policy (filed under seal). Furthermore, almost all of the justifications advanced by the company for restricting the employment opportunities of former employee-agents relate to concerns about having these former employees back at Allstate after their jobs were eliminated by the Reorganization Program: the purported worry about customer confusion, the desire to encourage the former employee-agents to sign on as independent contractors, the desire to prevent double-dipping, and the desire to keep terminated agents away from Allstate's other employees. Logically, as the district court recognized, the disparate impact of these two related actions should be measured in the same way, i.e., by comparing the age of the group of employees selected for adverse treatment with the age of the defendant's other employees who were treated more favorably. See II JA 450 (citing cases "related to a reduction-in-force (RIF), which is the same as the case at hand," in support of holding that the rehire policy is subject to disparate impact challenge). There is no reason why the disparate impact of Allstate's decision to terminate employee-agents while permitting other employees to keep their jobs should be measured differently from the decision to limit the terminated agents' opportunities for rehire while permitting other employees to be rehired immediately if terminated. The method used by the Commission to show that the rehire policy had a disparate impact-comparing the average age of employees subject to the policy with the average age of other employees not subject to the policy-is the method approved by this Court in Leftwich, 702 F.2d at 690. In that case the employer, in implementing a RIF, disqualified tenured faculty whose jobs were eliminated from competing for some available positions. The plaintiff proved that this practice had a disparate impact by showing that the tenured faculty barred from competing for these slots were on average more than ten years older than the defendant's non- tenured faculty who were not similarly restricted.<3> This Court stated that this was "clearly an appropriate method to establish disparate impact." Id. Accordingly, the district court correctly concluded that the evidence in this case establishes that Allstate's decisions to classify its employee-agents as subject to termination and ineligible for immediate rehire had a disparate impact based on age. In this action, the Commission has challenged only the second of Allstate's adverse actions toward employee-agents-the restriction on rehire. Had the Commission challenged the termination of the agents as well, the question of whether that action was based on a reasonable factor other than age would be quite different. As the district court recognized, Allstate's justification for the decision to terminate the employee-agents-the need to convert to independent contractor agents to remain competitive-on its face offers no basis for the decision to impose a moratorium on the rehire of the terminated agents into other positions. See II JA 449 n. 6 (noting that, although the issue is not presented, "it seems clear" that Allstate's reasons for terminating the agents "would constitute reasonable factors other than age"); II JA 455 (denying Allstate's motion for summary judgment on RFOA because a reasonable jury could find that Allstate's "proffered reasons [for the rehire moratorium] are unreasonable"). However, if the evidence presented by the Commission establishes that Allstate's decision to designate the employee- agents from among all of its employees for termination had a disparate impact based on age, then logically the decision to designate this same group for a restriction on rehire not applicable to its other employees also had a disparate impact. Relying in part on the Commission's OWBPA regulations, Allstate argues on appeal that the evidence in this case would not be sufficient to establish that the termination of the employee-agents had a disparate impact. Def. Br. at 40-42. This argument is fundamentally flawed. The discussion in the Commission's regulations regarding "decisional units" is addressed to the more common RIF situation where an employer chooses some employees from among all the employees in a particular facility or job category for layoff, and an allegation is made that the selection of the individual employees was based on age or on age- related criteria. See 29 C.F.R. § 1625.22(f)(3)(i)(B) ("The term 'decisional unit' has been developed to reflect the process by which an employer chose certain employees for a program and ruled out others for that program."). In this case, Allstate eliminated an entire job category, employee-agents, and restricted the opportunity of employee-agents to be rehired into all other jobs at Allstate. The Commission asserts that Allstate's decision to deny employee-agents consideration for employment in all other job classifications had a disparate impact based on age. Because jobs throughout Allstate's entire workforce were off limits to employee- agents, the appropriate decisional unit is Allstate's entire workforce. Allstate also argues that it is inappropriate to consider the ages of Allstate's incumbent non-agent employees because the restriction on rehire, by its nature, is only relevant to terminated employees. Def. Br. at 42 n. 7. This misses the point. The opportunity to be rehired if terminated is an important condition of employment of all Allstate employees. Allstate chose to strip a group of older employees of this right at the time they needed it most, while leaving it intact for the rest of its workforce. Where, as here, an employer classifies some employees in such a way that deprives them of a benefit that is part of the employment relationship because of a prohibited basis, such a practice violates the statute. Cf. Hishon v. King & Spalding, 467 U.S. 69, 78 (1984) (because the opportunity to become a partner was part of an associate's status as an employee at the defendant's firm, the allegation that the plaintiff was denied partnership because of gender states a claim cognizable under Title VII). Although most of Allstate's incumbent employees probably did not have occasion to avail themselves of the opportunity to be rehired, undoubtedly some of them did. Assuming Allstate does not generally terminate older employees more frequently than younger employees, the average age of those employees who had the opportunity to be rehired after termination would be no older than the average age of Allstate's workforce as a whole. According to Allstate, because this is a "hiring" case, the only way to establish disparate impact is by comparing the age of the terminated employee- agents and the age of the labor market for jobs at Allstate or actual applicants for those jobs. In arguing that the Commission's case fails without an assessment of the age of the labor pool, Allstate relies on cases where, unlike in this case, the challenged selection criterion applied to all those being considered for employment, including outside applicants. See, e.g., Hameed v. International Assoc., 637 F.2d 506, 509 (8th Cir. 1980) (high school diploma required for admission to ironworkers union's apprentice program); Green v. Missouri Pac. R.R. Co., 523 F.2d 1290, 1295 (8th Cir. 1975) (railroad's policy refused consideration of employment to any person with a criminal conviction); Banks v. East Baton Rouge Parish Sch. Bd., 320 F.3d 570, 574 (5th Cir. 2003) (reading test required for all applicants for janitorial position). But the challenged rehire policy in this case by its nature applies only to Allstate employees. Accordingly, it makes no sense to analyze the age of potential or actual outside applicants.<4> See, e.,g., Robinson v. City of Dallas, 514 F.2d 1271, 1273-74 (5th Cir. 1975) (in case alleging disparate impact based on race challenging the policy of disciplining employees who fail to pay debts, since the policy was imposed only on city employees, and not the population generally, "the question is whether black employees of the city of Dallas fail to pay their just debts more frequently than white employees of the city of Dallas"). Allstate's reliance on Abbott v. Federal Forge, 912 F.2d 867 (6th Cir. 1990), is similarly misplaced. Federal Forge involved a one-year hiring preference for laid-off plant workers, then a temporary hiring moratorium while a union grievance was arbitrated. Id. at 870. As the district court noted (II JA 452), the plaintiffs in Federal Forge compared the proportion of former employees 40 or older who were subject to the defendant's hiring moratorium to the proportion of persons 40 and over actually hired during the hiring moratorium. Id. at 873-74. The plaintiffs did not compare the proportion of laid-off employees 40 or older in the defendant's workforce disqualified for rehire to the proportion of laid-off employees under 40 disqualified for rehire, as was done in this case. Accordingly, the court of appeals did not consider whether the approach taken in this case was appropriate. Furthermore, the adverse effect of the employer's policies was much less apparent in Abbott, where the former employees were first given a hiring preference and then later were subject to a hiring moratorium, "unlike the present case where the rehire policy immediately [a]ffected all former employee-agents."<5> II JA 452. Allstate also argues that Ortega v. Safeway Stores, Inc., 943 F.2d 1230 (10th Cir. 1991), "confirms that the impact of a rehire rule should be assessed by determining how it affected selections for hire among those actually in competition for the positions in question." Def. Br. at 39. However, the plaintiffs failed in Ortega because they failed to show that the fact that laid-off female packers were much less likely to be reinstated than laid-off male packers was caused by Safeway's rehire policies rather than the plaintiffs' lack of qualifications for available jobs that required heavy lifting. 943 F.2d at 1246. Here, the Commission did not ask the court to infer the disparate impact of Allstate's rehire policy from the lack of success of terminated employee-agents in obtaining rehire. Instead, Allstate's policy speaks for itself: it explicitly bars the rehire of the terminated agents and no one else.<6> Allstate seems to acknowledge that its decision to take an adverse action against all of its employee-agents would affect "a significantly higher percentage of older employees" than a decision to take a similar action against a "cross-section of all Allstate employees would have," but implies that that has no significance because the employee-agents were older than the rest of the company's employees only because the company had not hired any new agents since 1990. Def. Br. at 33. The explanation for the fact that the employee-agents as a group are significantly older than the rest of Allstate's workforce is of no relevance. A challenge based on a disparate impact theory of proof does not require evidence of a discriminatory motive. For example, in Dothard v. Rawlinson, 433 U.S. 321, 332 (1977), the Court held that an Alabama law specifying that prison guards must be at least five feet two inches tall and weigh at least 120 pounds had an unlawful disparate impact on women, notwithstanding the fact that the state was not responsible for the fact that women are on average smaller than men. Finally, Allstate argues that, because it terminated 100% of the employee- agents under 40 as well as 100% of the employee-agents over 40, "the termination decisions did not have an adverse age impact." Def. Br. at 33 n.4, 43. This observation reflects a basic misunderstanding of the concept of disparate impact. Disparate-impact challenges by nature confront facially neutral policies that are applied to individuals both inside and outside the protected class but that disadvantage a protected group because of a characteristic linked to age, gender, or race. In Griggs, for example, the requirement that applicants or employees wishing to transfer to other jobs have a high school degree or pass intelligence tests was applied to 100% of the African American employees, as well as 100% of the whites. But because these requirements screened out African Americans and could not be justified as a necessary selection criterion, the policy violated Title VII. In Dothard, the minimum height and weight requirements were applied to 100% of the women seeking prison guard jobs, as well as 100% of the men. The Supreme Court nonetheless held that the standards were unlawful because they screened out a much larger percentage of women than men and were not justified by business necessity. Accordingly, the fact that Allstate restricted the rehire of all former employee-agents, whether they were over or under forty years of age, does not mean that the policy did not have a disparate impact. CONCLUSION For the foregoing reasons, the district court's decision should be affirmed, and the case should proceed to trial. Respectfully submitted, RONALD S. COOPER General Counsel CAROLYN L. WHEELER Acting Associate General Counsel VINCENT J. BLACKWOOD Assistant General Counsel ______________________________ JULIE L. GANTZ U.S. EQUAL EMPLOYMENT OPPORTUNITY COMMISSION 1801 L Street, N.W. Washington, DC 20507 (202) 663-4718 CERTIFICATE OF COMPLIANCE Pursuant to FRAP 32(a)(7)(C), I certify that this brief has been prepared in proportionally spaced typeface using Microsoft Word 2003 in 14 point Times New Roman Font, and the textual portion contains 7,725 words. _______________________________ Julie L. Gantz CERTIFICATE OF SERVICE I hereby certify that on July 18, 2007, two paper copies and one digital copy on diskette of the foregoing brief have been served by first class mail, postage prepaid, to the following counsel: Donald R. Livingston Nathan J. Oleson AKIN GUMP STRAUSS HAUER & FELD LLP 1333 New Hampshire Avenue, NW Washington, DC 20036 Rae T. Vann Ann Elizabeth Reesman MCGUINESS NORRIS & WILLIAMS, LLP 1015 15th Street, NW, Suite 1200 Washington, DC 20005 _____________________________ Julie L. Gantz *********************************************************************** <> <1> The document is entitled "Preparing for the Future Reorganization Program (Clarification of Re-Hire Policy)." It states in its entirety: To avoid any potential violations of the non-compete provision of the agent contract or the enhanced severance plan and for other business reasons, the company has decided not to offer reemployment to former employee agents (departed agents) who left as a result of the Preparing for the Future Reorganization Program. Departed agents will be eligible for reemployment when they have reached the one year anniversary of their termination or they are no longer receiving enhanced separation payments, whichever is longer. After the duration of this time, the normal re-hire policy will apply. Rehire policy (filed under seal); see also II JA 237-38, 241-42. <2> Although the issue need not be resolved in this case, Allstate's premise that disparate impact hiring claims are not cognizable under the ADEA stands on shaky ground. Allstate assumes that, because § 4(a)(2), unlike § 703(a)(2) of Title VII, does not include the word "applicants," Congress made a deliberate decision to exclude hiring discrimination from the scope of § 4(a)(2). However, Allstate fails to mention that, at the time of the Supreme Court's decision in Griggs v. Duke Power Co., 401 U.S. 424 (1971), holding that Title VII prohibits, inter alia, disparate impact in hiring, § 703(a)(2) did not include the word "applicants" either. It was not until 1972 that Congress amended the provision to add "applicants for employment." See Equal Employment Opportunity Act of 1972, Pub. L. No. 92- 261, § 8(a), 86 Stat. 103 (1972). The amendment was intended merely to express Congress's agreement with court decisions already applying 42 U.S.C. § 2000e- 2(a)(2) to applicants. See Conf. Rep. on H.R. 1746, reprinted in 118 Cong. Rec. 7166, 7169 § 8(a)-(b) (1972) (amendment was "merely declaratory of present laws as contained in [court] decisions"). When Congress drafted the ADEA in 1967, it imported the version of § 703(a)(2) then in effect. Since the lack of an explicit statutory reference to "applicants" did not prevent the Griggs Court from holding that hiring policies with a disparate impact were unlawful under Title VII, there is no reason to assume that Congress intended a different result under the ADEA. <3> Allstate argues that Leftwich supports its approach because the plaintiffs compared the age of the tenured faculty subject to the restriction only with the age of nontenured faculty. According to Allstate, the Commission's approach would be tantamount to comparing the age of the tenured faculty in Leftwich with the age of all of the college's employees, including "janitors." Def. Br. at 40, 42. But the comparison in Leftwich was limited to other faculty because the restriction at issue applied only to hire for faculty positions. Here, the hiring moratorium applied to all positions at Allstate. Accordingly, it is appropriate to consider the age of all other Allstate employees. <4> Although we believe this evidence is not relevant, it would be very surprising if the labor market or the applicant pool for jobs at Allstate were on average older than the incumbents in those jobs. As courts have recognized, common sense dictates that persons seeking a particular job will tend to be younger than the persons holding the job. See Henson v. Liggett Group, Inc., 61 F.3d 270, 276 (4th Cir. 1995) ("the hiring of young employees into entry level positions is reasonable since younger people are more apt to apply for such positions"); Brown v. M & M/Mars, 883 F.2d 505, 511 n.3 (7th Cir. 1989)(older employees tend to be replaced by younger employees; "[w]hen older employees leave the work force, for whatever reasons, they will often be replaced by younger employees"). <5> The Federal Forge court suggested that the plaintiffs should have compared "the percentage of those 40 or over in the available labor pool, including the former Federal Forge employees, with the percentage of those 40 or over in the remainder of the qualified labor pool." Id. at 874. That rationale, if applied to an employer's rehire policy, would make it all but impossible to establish a disparate impact. The number of individuals in the local labor market will normally dwarf the number of employees of any employer and, accordingly, the number of individuals subject to an employer's rehire policy. Even where an employer's rehire policy has a severe adverse impact on a sector of its workforce in a protected category, measuring the effect of the policy on the entire labor market dilutes the effect of the policy to such an extent that it would nearly ever be possible to show a disparate impact. The facts in Federal Forge illustrate this problem. There, 107 former employees were disqualified from rehire. Presumably, they made up a small percentage of the local labor market in Lansing, Michigan. The court determined there could be a disparate impact "only if the former Federal Division employees made up a significant portion of the total qualified labor pool." Id. The court noted, "The removal of 107 skilled and unskilled workers in the Lansing area could hardly have significantly altered the percentage of those 40 and over in the latter group." Id. Thus under this analysis, any time an employer's workforce comprises only a small proportion of the labor market, the adverse impact theory would be unavailable. Because such a result would subvert the purpose of the ADEA, the court's suggestion should not be followed even in a case where comparison to the labor market is appropriate. <6> The question of whether individual employee-agents were qualified for particular vacancies from which they were barred remains to be determined at the damages phase of proceedings, should a jury determine that the rehire moratorium was not based on a reasonable factor other than age.