Full opinion text
PER CURIAM: This age discrimination appeal requires us to decide several issues concerning application of the “single-filing,” or “piggybacking,” rule to opt-in collective actions under 29 U.S.C. § 216(b). We address these important issues in Part I of this opinion, and we provide a brief summary of our holdings here. We first clarify the meaning of the “similarly situated” requirement under § 216(b). We conclude the similarly situated requirement is not particularly stringent, and we suggest an approach district courts can use to better manage these cases. We next consider the proper temporal scope of § 216(b) collective actions, an issue this Court has not yet directly addressed. As to the proper rearward cutoff date, we conclude in order to properly opt into a § 216(b) action, a plaintiff must allege discriminatory treatment within 180 or 300 days before the representative plaintiffs charge is filed with the Equal Employment Opportunity Commission (EEOC). We conclude the proper forward cutoff date is the date of filing of the representative charge. Plaintiffs who do not allege discriminatory treatment occurring during this period may not opt into a § 216(b) action. In other words, a plaintiff must have been able to file his or her charge of discrimination on the date the representative plaintiff filed the representative charge. In Part II of this opinion, we address the sufficiency of the evidence presented in this case. BACKGROUND & PROCEDURAL HISTORY This case arises under the Age Discrimination in Employment Act of 1967 (ADEA), as amended, 29 U.S.C. §§ 621-34, and the Florida Civil Rights Act of 1992 (FCRA), Fla. Stat. Ann. §§ 760.01-760.11. Plaintiffs David Hipp, Harry W. McKown, Jr,, and Brad Stein filed their original complaint in Florida state court, alleging Appellant Liberty National Life Insurance Company (Liberty National) engaged in a pattern and practice of age discrimination resulting in Plaintiffs’ constructive discharges. These three named Plaintiffs claimed to bring the case on behalf of themselves and others similarly-situated. Liberty National removed the case to the United States District Court for the Middle District of Florida. Plaintiffs amended the complaint to add another named Plaintiff, Mike Stell. Plaintiffs then informed the district court they intended to pursue a collective action under 29 U.S.C. § 216(b). Plaintiffs sought to distribute notice of an opt-in class under § 216(b). Liberty National opposed this motion, maintaining Plaintiffs were not “similarly situated.” Liberty National further argued that even if Plaintiffs were similarly situated such that a collective action was proper, Plaintiffs’ proposed notice was defective because it would allow opt-in by individuals who could not properly piggyback into the case. On February 6, 1996, the district court approved Plaintiffs’ proposed notice. Hipp v. Liberty Nat’l Life Ins. Co., 164 F.R.D. 574, 576 (M.D.Fla. Feb. 6, 1996). The district court adopted the following class definition, proposed by Plaintiffs: All persons who are, or were, employed by Liberty National Life Insurance Company on or after August 25, 1998, who are, or were, managerial employees, district managers or above, residing in the United States, who were over 40 years of age. Id. Over twenty individuals eventually filed consents to opt in, although some were untimely. After the close of extended discovery, Liberty National sought to sever the cases. Liberty National also filed motions for summary judgment as to the claims of several Plaintiffs. The court substantially denied Liberty National’s motions for summary judgment. The court also denied Liberty National’s request to certify an interlocutory appeal and its motion to sever the cases. The claims of ten Plaintiffs were tried before a jury over the course of five weeks beginning June 1, 1998. The jury returned verdicts on July 9, 1998, finding that Liberty National had engaged in a pattern and practice of age discrimination, but returning defense verdicts as to three of the ten Plaintiffs, whose claims are not at issue in this appeal. The jury’s verdicts awarded back pay and ADEA liquidated damages to the seven prevailing Plaintiffs. Hipp v. Liberty Nat’l Life Ins. Co., 65 F.Supp.2d 1314, 1334-35 (M.D.Fla.1999); Hipp v. Liberty Nat’l Life Ins. Co., 29 F.Supp.2d 1314, 1318-19 (M.D.Fla.1998). Plaintiffs Hipp and Stein received punitive damages and pain and suffering damages under the FCRA, and the jury returned advisory front pay verdicts as to the five prevailing Plaintiffs other than Hipp and Stein. The district court ordered that no Plaintiff should receive front pay because the Plaintiffs’ front pay award estimates were too speculative. Hipp, 65 F.Supp.2d at 1336; Hipp v. Liberty Nat’l Life Ins. Co., 39 F.Supp.2d 1359, 1361-64 (M.D.Fla.1999). Furthermore, the district court remitted some of the punitive damages and liquidated damages awarded to Stein and Hipp. Hipp, 39 F.Supp.2d at 1364-65; Hipp, 65 F.Supp.2d at 1335. In all other respects, the court entered judgment in accordance with the jury’s verdicts. Liberty National moved for judgment as a matter of law (JMOL) both during and after trial, and filed motions for remittitur and new trial after entry of judgment. The district court denied Liberty National’s motions, except to the extent that it remitted some of the damages. Hipp, 65 F.Supp.2d at 1345. Liberty National raises the following issues on appeal: (1) the district court erred in permitting a collective action in this case; (2) the district court erred in denying Liberty National’s motions for judgment as a matter of law on Plaintiffs’ pattern and practice claims; (3) the district court erred in denying Liberty National’s motions for judgment as a matter of law on the Plaintiffs’ individual claims; (4) the district court erred in denying Liberty National’s motions for judgment as a matter of law as to liquidated damages; and (5) the district court erred in denying Liberty National’s motion to remit the jury’s verdicts. For the reasons stated in Part I of this opinion, we affirm the district court’s judgment on the propriety of a collective action, but we reverse as to the temporal scope of the action. For the reasons stated in Part II, we reverse the pattern and practice finding, and we reverse the verdicts in favor of the individual Plaintiffs. In light of our disposition regarding the verdicts, we need not address Liberty National’s arguments pertaining to the damages awarded in this case. DISCUSSION I. OPT-IN COLLECTIVE ACTIONS UNDER 29 U.S.C. § 216(b) A. Introduction and Background Plaintiffs wishing to sue as a class under ADEA must utilize the opt-in class mechanism provided in 29 U.S.C. § 216(b) instead of the opt-out class procedure provided in Fed.R.Civ.P. 23. See Grayson, 79 F.3d at 1102 (citing Price v. Maryland Cas. Co., 561 F.2d 609, 610-11 (5th Cir.1977)). In a Rule 23 class action, each person who falls within the class definition is considered to be a class member and is bound by the judgment, favorable or unfavorable, unless he has opted out. Fed.R.Civ.P. 23(c)(3); Grayson, 79 F.3d at 1106. By contrast, a putative plaintiff must affirmatively opt into a § 216(b) action by filing his written consent with the court in order to be considered a class member and be bound by the outcome of the action. § 216(b); Grayson, 79 F.3d at 1106; see also LaChapelle v. Owens-Illinois, Inc., 513 F.2d 286, 288-89 (5th Cir.1975) (recognizing this “fundamental” difference between Rule 23 class actions and § 216(b) class actions). To maintain an opt-in class action under § 216(b), plaintiffs must demonstrate that they are “similarly situated.” § 216(b); Grayson, 79 F.3d at 1093. In subpart B, we will clarify the meaning of § 216(b)’s “similarly situated” requirement in this circuit. We will also suggest an approach for district courts to use in considering certification of ADEA opt-in classes. This case also requires us to consider application of the piggybacking rule to ADEA opt-in classes under § 216(b). As a general rule, an employee who wishes to sue his employer for age discrimination must first file an administrative charge of discrimination with the EEOC. Under the piggybacking rule, however, a putative plaintiff who has not filed his own EEOC charge may “piggyback” his claim onto the claim of a plaintiff who has filed a timely charge. Grayson, 79 F.3d at 1101; Calloway v. Partners Nat’l Health Plans, 986 F.2d 446, 450 (11th Cir.1993). We have specifically held that the piggybacking rule is applicable to ADEA cases. Grayson, 79 F.3d at 1101. In so holding, we adopted the two requirements used in Title VII piggybacking cases: A plaintiff may piggyback on another plaintiffs charge provided “ ‘(1) the relied upon charge [to which he is piggybacking] is not invalid, and (2) the individual claims of the filing and non-filing plaintiff [the named filing plaintiff and the piggybacking plaintiff] arise out of similar discriminatory treatment in the same time frame.’ ” Id. at 1101-02 (quoting Calloway, 986 F.2d at 450). The parties agree Plaintiff Stein filed the representative charge in this case, and they agree his charge was valid. They disagree, however, as to the precise meaning of the second requirement, and we must therefore now determine the scope of the statement “in the same time frame.” In subpart C, we address the appropriate temporal scope of an ADEA opt-in collective action. B. “Similarly situated” requirement We first must determine whether Plaintiffs were “similarly situated” as required to create an opt-in class under § 216(b). We review the district court’s decision that a collective action was proper because Plaintiffs were similarly situated for abuse of discretion. See Grayson, 79 F.3d at 1097; see also Armstrong v. Martin Marietta Corp., 138 F.3d 1374, 1388 (11th Cir.1998) (en banc). For an opt-in class to be created under § 216(b), a named plaintiff must be suing on behalf of himself and other “similarly situated” employees. “ ‘[Plaintiffs need show only that their positions are similar, not identical, to the positions held by the putative class members.’ ” Grayson, 79 F.3d at 1096 (quoting Sperling v. Hoffman-La Roche, Inc., 118 F.R.D. 392, 407 (D.N.J.1988), aff'd in part and appeal dismissed in part, 862 F.2d 439 (3d Cir.1988), aff'd, 493 U.S. 165, 110 S.Ct. 482, 107 L.Ed.2d 480 (1989)). Liberty National claims the district court should have used a two-tiered approach in making the similarly situated determination. Under this approach, during the early stages of litigation, the district court would have evaluated the case under a lenient standard and likely would have granted preliminary certification of an opt-in class. The court would then have re-evaluated the similarly situated question at a later stage, once discovery produced more information regarding the nature of Plaintiffs’ claims. In Mooney v. Aramco Servs. Co., 54 F.3d 1207 (5th Cir.1995), the Fifth Circuit, while finding it unnecessary to decide which methodology district courts should use in making ADEA class certification decisions, see id. at 1216, described the two-tiered approach used by the district court in that case: The first determination is made at the so-called “notice stage.” At the notice stage, the district court makes a decision — usually based only on the pleadings and any affidavits which have been submitted — whether notice of the action should be given to potential class members. Because the court has minimal evidence, this determination is made using a fairly lenient standard, and typically results in “conditional certification” of a representative class. If the district court “conditionally certifies” the class, putative class members are given notice and the opportunity to “opt-in.” The action proceeds as a representative action throughout discovery. The second determination is typically precipitated by a motion for “decertification” by the defendant usually filed after discovery is largely complete and the matter is ready for trial. At this stage, the court has much more information on which to base its decision, and makes a factual determination on the similarly situated question. If the claimants are similarly situated, the district court allows the representative action to proceed to trial. If the claimants are not similarly situated, the district court de-certifies the class, and the opt-in plaintiffs are dismissed without prejudice. The class representatives — i.e. the original plaintiffs — proceed to trial on their individual claims. Based on our review of the case law, no representative class has ever survived the second stage of review. Id. at 1213-14 (internal footnote omitted). Liberty National sought to have the district court utilize the two-tiered approach in this case, but the district court declined to do so. If the court had employed this approach, Liberty National contends, the proper result would have been decertification of the class after discovery because of the individualized nature of Plaintiffs’ claims. See, e.g., Thiessen v. Gen. Elec. Capital Corp., 996 F.Supp. 1071, 1083 (D.Kan.1998) (conditionally certifying opt-in class), and Thiessen v. Gen. Elec. Capital Corp., 13 F.Supp.2d 1131, 1141 (D.Kan.1998) (decertifying opt-in class and dismissing claims of opt-in plaintiffs); see also Lusardi v. Xerox Corp., 118 F.R.D. 351, 353-54 (D.N.J.1987) (utilizing two-tiered approach to class certification), mandamus granted in part, appeal dismissed, Lusardi v. Lechner, 855 F.2d 1062, 1080 (3d Cir.1988), on remand, Lusardi v. Xerox Corp., 122 F.R.D. 463, 464 (D.N.J.1988); Vaszlavik v. Storage Tech. Corp., 175 F.R.D. 672, 678-79 (D.Colo.1997) (same); Bayles v. Am. Med. Response of Colorado, 950 F.Supp. 1053, 1066-67 (D.Colo.1996) (same); Brooks v. Bellsouth Telecomm., Inc., 164 F.R.D. 561, 568 (N.D.Ala.1995) (endorsing the two-tiered approach), aff'd mem., 114 F.3d 1202 (11th Cir.1997). After the district court substantially denied Liberty National’s motion for summary judgment, Liberty National filed a “Motion for Certification of Interlocutory Appeal of July 28, 1997, Order and Alternate Motion for Reconsideration.” Liberty National also filed a motion to sever the cases pursuant to Fed.R.Civ.P. 42(b). The district court denied these motions. In denying the motion to sever, the district court noted that the similarly situated requirement is not very stringent in this circuit. This Court expressed its view of the similarly situated requirement in Grayson: “[T]he ‘similarly situated’ requirement of § 216(b) is more elastic and less stringent than the requirements found in Rule 20 (joinder) and Rule 42 (severance).” 79 F.3d at 1095. “[A] unified policy, plan, or scheme of discrimination may not be required to satisfy the more liberal ‘similarly situated’ requirement of § 216(b).” Id. “[Plaintiffs bear the burden of demonstrating a reasonable basis for their claim of classwide discrimination. The plaintiffs may meet this burden, which is not heavy, by making substantial allegations of class-wide discrimination, that is, detailed allegations supported by affidavits which successfully engage defendants’ affidavits to the contrary.” Id. at 1097 (internal citations and quotation marks omitted). The two-tiered approach to certification of § 216(b) opt-in classes described above appears to be an effective tool for district courts to use in managing these often complex cases, and we suggest that district courts in this circuit adopt it in future cases. Nothing in our circuit precedent, however, requires district courts to utilize this approach. The decision to ■ create an opt-in class under § 216(b), like the decision on class certification under Rule 23, remains soundly within the discretion of the district court. See Grayson, 79 F.3d at 1097 (§ 216(b) class); Rutstein v. Avis Rent-A-Car Sys., Inc., 211 F.3d 1228, 1233 (11th Cir.2000) (Rule 23 class), cert. denied, - U.S. -, 121 S.Ct. 1354, 149 L.Ed.2d 285 (2001). It may have been prudent for the district court in this case to have decertified the class upon defendant’s motion, but we cannot reverse the decision unless we find the district court abused its discretion. Given the flexibility of the similarly situated requirement under Grayson, we cannot find the district court abused its discretion in allowing the opt-in class in this case. Liberty National emphasizes that Plaintiffs in this case worked in different geographical locations. This factor is not conclusive. The plaintiffs in Grayson worked in several states, and the court still held that they met the similarly situated requirement. 79 F.3d at 1091. Liberty National also argues that each Plaintiffs case was unique and required an individual analysis of his or her working conditions. Like the plaintiffs in Gmyson, however, Plaintiffs in this case all held the same job title, and they all alleged similar, though not identical, discriminatory treatment. See id. at 1098-99. C. Temporal Scope We next must determine whether the opt-in Plaintiffs have met the “in the same time frame” requirement such that they can properly piggyback their claims onto Stein’s EEOC charge. Specifically, Liberty National claims (1) Plaintiff Lee should have been excluded because he did not fall within the appropriate rearward temporal scope of the action, and (2) Plaintiffs Carter, Tuggle, and Agee should have been excluded because they did not fall within the appropriate forward temporal scope. At oral argument, Liberty National argued the district court’s decisions regarding the temporal scope of the collective action should be reviewed de novo. Plaintiffs argue the standard of review is abuse of discretion. We agree with Liberty National that, since the issue is analogous to a statute of limitations issue, the standard of review is de novo. See Atlantic Land & Improvement Co. v. United States, 790 F.2d 853, 857 (11th Cir.1986) (noting that district court’s application of statute of limitations, as a question of law, is reviewed de novo). 1. Appropriate Rearward Scope Liberty National claims Plaintiff Lee should not have been allowed to participate in the opt-in collective action because his claim was time-barred. Plaintiff Lee retired from his position as District Manager in Anniston, Alabama, on December 1,1993, more than a year before Plaintiff Stein filed the representative charge in this case. Since Alabama is a non-deferral state, Plaintiff Lee had only 180 days from December 1, 1993, to file a charge of discrimination with the EEOC. Plaintiff Lee did not file a charge, and he could not have filed one on December 9, 1994, the date Plaintiff Stein filed the representative charge in this case. Liberty National claims Grayson established that the rearward scope of an ADEA opt-in action is 180 or 300 days (depending on whether the putative opt-in plaintiff resides in a deferral or a non-deferral state) before the filing of the representative charge. Grayson did not, however, establish such a rule. Instead, the Grayson plaintiffs conceded such a rearward cut-off date should apply. See 79 F.3d at 1101, nn.26-27 and accompanying text. Liberty National emphasizes the Gray-son court’s statement that misapplication of the piggybacking rule would “virtually eliminate the statute of limitations for opt-in plaintiffs in ADEA actions.” 79 F.3d at 1107. We recognize that the Court made this statement in the context of determining that the filing of an opt-in plaintiffs written consent to opt into the action, rather than the filing of the original complaint in the action, serves to toll the statute of limitations on that plaintiffs ADEA claim. The same concept is, however, involved in this case. Here, the concern is that plaintiffs would be able to revive stale claims for which they failed to file EEOC charges. “The purpose of the requirement that a plaintiff file an EEOC charge within 180 days (or 300 days in a deferral state) of the allegedly illegal act or practice is (1) to give the employer prompt notice of the complaint against it, and (2) to give the EEOC sufficient time to attempt the conciliation process before a civil action is filed.” Id. at 1102-03. While the issue of the proper rearward scope of the action was not directly before the court in Gray-son, we believe the rule adopted by the parties in that case is the appropriate one to serve these purposes. We therefore hold the rearward scope of an ADEA opt-in action should be limited to those plaintiffs who allege discriminatory treatment within 180 or 300 days before the representative charge is filed. Plaintiff Lee’s claim is procedurally barred because it arose before the “piggybacking window” opened in Alabama, a non-deferral state. Other courts that have addressed this issue have come to the same conclusion. See, e.g., McDonald v. United Air Lines, Inc., 587 F.2d 357, 361 (7th Cir.1978) (holding that the rearward scope must be defined by going back 90 days (the then-existing EEOC charge-filing period) from the date of the earliest-filed charge). In McDonald, the Seventh Circuit held that the EEOC and the employer were put on notice by intervening plaintiffs’ EEOC charges, which were filed earlier than the named plaintiffs’ charges. See id. The court left room for the fact that continuing discovery in that case “might turn up a prior EEOC charge, thus resulting in a different tolling date.” Id. at 361 n. 10. The court refused, however, to hold the employer liable for discrimination that occurred before the employer and the EEOC were on notice and could therefore attempt to resolve the issue through conciliation. See id. at 360-61; see also, e.g., Thiessen, 996 F.Supp. at 1076-77 (holding that “the single-filing rule is properly applied only to those individuals who could have filed timely EEOC charges at the time [the representative plaintiff] actually filed his charge”); Brooks, 164 F.R.D. at 570 (same); Church v. Consol. Freightways, Inc., 137 F.R.D. 294, 309 (N.D.Cal.1991) (same); Levine v. Lane Bryant, 700 F.Supp. 949, 957 (N.D.Ill.1988) (same); Walker v. Mountain States Tel. & Tel. Co., 112 F.R.D. 44, 47 (D.Colo.1986) (same); cf. Larkin v. Pullman-Standard Div., Pullman, Inc., 854 F.2d 1549, 1562-65 (11th Cir.1988) (suggesting liability dates from 180 days prior to filing of earliest EEOC charge), vacated on other grounds sub nom., Pullman-Standard, Inc. v. Swint, 493 U.S. 929, 110 S.Ct. 316, 107 L.Ed.2d 307 (1989). Plaintiffs argue even if Plaintiff Lee’s claim otherwise falls outside the rearward scope of the class, the continuing violation doctrine revives it. As noted above, it is undisputed that no action was taken against Lee during the charge-filing period. Some cases hold that the continued enforcement of a discriminatory policy against an individual plaintiff constitutes a continuing violation such that the individual plaintiff may sue on otherwise time-barred claims as long as one act of discrimination has occurred against that individual during the statutory period. See Jenson v. Eveleth Taconite Co., 130 F.3d 1287, 1303 (8th Cir.1997); Roberts v. North Am. Rockwell Corp., 650 F.2d 823, 826-28 (6th Cir.1981) (in failure-to-hire context, violation was continuing because plaintiff, after being rejected because of her sex, made several more unsuccessful attempts to apply for job). These courts held that claims of discrimination were not time-barred because some acts of discrimination against the individual plaintiffs had occurred within the statutory period, even though prior acts did not. The earlier acts of discrimination were actionable because they were part of a continuing violation. We can find no authority, however, for allowing one plaintiff to revive a stale claim simply because the allegedly discriminatory policy still exists and is being enforced against others. See, e.g., Woodburn v. LTV Aerospace Corp., 531 F.2d 750, 751 (5th Cir.1976); Miller v. Int’l Tel. & Tel. Corp., 755 F.2d 20, 25 (2d Cir.1985) (“[Sjeveral courts have held that an employee who has been discharged pursuant to a discriminatory policy may not take advantage of later discriminatory acts against other employees for the purpose of postponing the running of the statutory period as to him on a continuing violation theory.”). Even if the continuing violation doctrine could be read broadly enough to preserve otherwise stale claims of all class members as long as the illegal policy is being enforced against some members of the class, allowing it to save Plaintiff Lee’s claim in this case would contravene the doctrine’s purpose. The continuing violation doctrine is premised on “the equitable notion that the statute of limitations ought not to begin to run until facts supportive of the cause of action are or should be apparent to a reasonably prudent person similarly situated.” Alldread v. City of Grenada, 988 F.2d 1425, 1432 (5th Cir.1993) (internal quotation marks omitted). This circuit and others have recognized this underlying premise in holding that “[a] claim arising out of an injury which is ‘continuing’ only because a putative plaintiff knowingly fails to seek relief is exactly the sort of claim that Congress intended to bar by the 180-day limitation period.” Roberts v. Gadsden Mem. Hosp., 850 F.2d 1549, 1550 (11th Cir.1988); see also Carter v. West Publ’g Co., 225 F.3d 1258, 1264 (11th Cir.2000); Doe v. R.R. Donnelley & Sons Co., 42 F.3d 439, 446 (7th Cir.1994) (“[T]he purpose of permitting a plaintiff to maintain a cause of action on the continuing violation theory is to permit the inclusion of acts whose character as discriminatory acts was not apparent at the time they occurred.”) (citing Moskowitz v. Trustees of Purdue Univ., 5 F.3d 279, 282 (7th Cir.1993)); Martin v. Nannie and the Newborns, Inc., 3 F.3d 1410, 1415 n. 6 (10th Cir.1993) (“[I]f an event or series of events should have alerted a reasonable person to act to assert his or her rights at the time of the violation, the victim cannot later rely on the continuing violation doctrine to overcome the statutory requirement of filing a charge with the EEOC with respect to that event or series of events.”); Berry v. Bd. of Supervisors, 715 F.2d 971, 981 (5th Cir.1983) (In determining whether a continuing violation exists, courts should consider whether “the [alleged discriminatory] act [has] the degree of permanence which should trigger an employee’s awareness of and duty to assert his or her rights, or which should indicate to the employee that the continued existence of the adverse consequences of the act is to be expected without being dependent on a continuing intent to discriminate.”). Plaintiff Lee testified he suspected age discrimination when he resigned, in 1993. He claimed Regional Vice President Andy King was harassing him and making age-based comments, and Lee felt he was being forced to retire. During his direct examination at trial, Lee testified that, at the time he left the company, he felt King’s harassment was based on Lee’s age. The Gadsden Memorial rationale therefore applies here, and Lee may not rely on the continuing violation doctrine to revive his stale claim of age discrimination. Plaintiff Lee should not have been permitted to opt into this action, and the jury verdict in his favor must be reversed. 2. Appropriate Fonvard Scope Liberty National argues Plaintiffs Carter, Tuggle, and Agee should not have been allowed to opt into the action because their claims arose after the forward scope of the action should have been cut off. Plaintiff Carter was District Manager in Roanoke, Alabama, until July 1994, when he took medical leave. At that point, he was replaced by Plaintiff Agee, who was older than Carter. One year later, Carter expressed interest in returning to his position, but Agee was still District Manager. Liberty National reassigned Carter to a sales manager position in Opelika, Alabama, and Carter resigned. Carter did not file a charge of age discrimination with the EEOC, and he could not have filed one on December 9, 1994, the date Plaintiff Stein filed the representative charge in this case. Plaintiff Agee was fired from his position as District Manager in Roanoke, Alabama, in June 1996. Agee did not file a charge of discrimination with the EEOC, and he could not have filed one on December 9,1994. Plaintiff Tuggle resigned from his position as District Manager in Tullahoma, Tennessee, in December 1995, at age 43. Tuggle was replaced by Donnie Ventress, who was older than Tuggle. Tuggle did not file a charge of age discrimination with the EEOC, and he could not have filed one on December 9,1994. Liberty National claims the forward scope of this action should have been cut off on the day Stein (the representative plaintiff) left Liberty National. In the alternative, Liberty National contends the forward scope should have been cut off on the day Stein filed the representative charge, or, at the latest, on the day the complaint was filed in this case. Plaintiff Stein resigned from Liberty National on July 29, 1994, and he filed his EEOC charge on December 9, 1994. The original complaint in this case was filed on June 22, 1995, and the amended complaint was filed on October 5, 1995. Plaintiff Carter resigned from Liberty National in July 1995; Plaintiff Tuggle resigned in December 1995; and Plaintiff Agee’s employment was terminated in June 1996. Under Liberty National’s theory, then, all three of these Plaintiffs should have been barred from joining the collective action. Liberty National relies on Jones v. Firestone Tire & Rubber Co., 977 F.2d 527, 532 (11th Cir.1992), a Title VII case in which this Court recited the district court’s decision that the representative charge could be relied on by non-filing plaintiffs whose claims arose between the date 180 days before the representative charge was filed and the date of the representative plaintiffs departure from the company. See id. at 532. The temporal scope of that case was not, however, an issue in the appeal. This Court therefore had no occasion to pass on the correctness of the district court’s decision in that respect. See id. Liberty National argues that if the piggybacking window did not close the day Stein left the company, it closed the day he filed his EEOC charge. Liberty National finds support for this position in Grayson. In finding plaintiff Grayson’s EEOC charge could not be the representative charge in that case, the Court noted that Grayson’s charge did “not put the EEOC on notice of demotions occurring after [its] filing.” Grayson, 79 F.3d at 1104. The Court noted that if Grayson’s charge were used as the representative charge, putative plaintiffs demoted after the date it was filed could not opt into the class. See id. This language suggests the Court would have considered plaintiffs who were demoted after Grayson was demoted, but before Grayson filed his EEOC charge, able to opt in, undermining Liberty National’s argument that the cut-off date is the date the representative plaintiff leaves the company. It also, however, strongly suggests the Court agreed with Liberty National’s alternative argument that the forward scope of the opt-in class should be cut off on the date the representative charge was filed. The Court in Grayson went on to select the charge of plaintiff Kempton, which was “the first timely filed charge of one of the named plaintiffs that gives adequate notice of the scope of the class.” Id. at 1104-05. See also Walker, 112 F.R.D. at 47 (“The latest date on which potential plaintiffs may claim discrimination occurred is also circumscribed by [the representative] charge.... The EEOC could not have been expected to perform its conciliatory function for alleged acts of discrimination which could have occurred long after [the representative] charge was filed and of which no notice was given.”); Griffin v. Casey, 42 Fair Empl. Prac. Cas. (BNA) 1423, 1429 (M.D.Fla. Jan.20, 1987) (“[T]he fact that the present case involves a class action should not expand the temporal scope of the claims that would otherwise be heard if Griffin had litigated alone.”). While the issue was not squarely before this Court in Grayson, our statement as to the inadequacy of Grayson’s charge supports the view that the forward scope of an opt-in class ends on the date the representative charge is filed. We agree that this rule is the proper one, and we adopt it today. This rule is consistent with the purposes of requiring putative ADEA plaintiffs to file charges with the EEOC. As stated above, the charge-filing requirement is designed to put the EEOC and the employer on notice of the allegations. The objective of the piggybacking rule is to allow non-filing plaintiffs to rely on the charges of filing plaintiffs when it would be a “useless act” for the non-filing plaintiffs to file their own charges. Grayson, 79 F.3d at 1103 (internal quotation marks omitted). It is not, however, “useless” for plaintiffs with claims arising after the representative charge is filed to file their own charges. On the contrary, such filing is necessary to remain faithful to the overriding purpose of the charge-filing requirement: notifying employers and the EEOC of the allegations in a timely manner. Plaintiffs allege the claims of Carter, Tuggle, and Agee “related to and grew out of’ the allegations in Stein’s EEOC charge, and these Plaintiffs therefore should have been allowed to opt into this action under a continuing violation theory. Indeed, there is some authority supporting this position. See, e.g., McDonald, 587 F.2d at 361 (permitting opt-in of plaintiffs discharged after representative charges were filed but before discriminatory rule was abolished); Church, 137 F.R.D. at 309 (The rule that “the latest date upon which a potential class member must have been able to file is defined by the date of the last filed EEOC charge” does not apply “where acts of ongoing discrimination in furtherance of an earlier implemented plan are alleged.”); Levine, 700 F.Supp. at 957 (representative charge “put [the employer] on notice that someone was challenging its alleged age discrimination”). Most of the cases Plaintiffs cite, however, do not establish that one plaintiff may opt into an action because his claims of discrimination relate to allegations made in an EEOC charge filed by another plaintiff. See Turner v. Orr, 804 F.2d 1223, 1226 (11th Cir.1986) (Title VII complaint may allege incidents that occurred during the pendency of an EEOC charge if the incidents could reasonably be expected to “grow out of’ the allegations in the charge); Sanchez v. Standard Brands, Inc., 431 F.2d 455, 465-67 (5th Cir.1970) (same). It makes sense to allow an individual plaintiff to include in his complaint allegations of conduct occurring after the representative charge is filed, as long as that conduct relates to the conduct alleged in the charge. For example, in Turner, the plaintiffs EEOC charge generally alleged the company had failed to promote him, and he wanted to include in his complaint allegations that he had not received a specific promotion. 804 F.2d at 1226-27. In such a case, the EEOC and the employer can reasonably be expected to be on notice that the plaintiff will challenge the employment decision. Here, however, the actions about which Carter, Lee, and Tuggle complained occurred well after Stein filed his charge. A single charge cannot be expected to put the EEOC and employer on notice that general policies as applied to different individuals in different offices are being challenged indefinitely. Where the representative charge clearly alleges a continuing, concrete policy that is illegal, as in McDonald, it might be fair to assume the company has knowledge of later-arising claims challenging the same policy. See, e.g., Bush v. Liberty Nat’l Life Ins. Co., 12 F.Supp.2d 1251, 1258-59 n. 11 (M.D.Ala.1998), aff'd mem., 196 F.3d 1261 (11th Cir.1999) (continuing violation must be “clearly asserted both in the EEOC filing and in the complaint”) (quoting Miller v. Int’l Tel. & Tel. Corp., 755 F.2d 20, 25 (2d Cir.1985)). Since we are not faced with such a case, we need not decide that more difficult issue today. The forward cut-off date in this case should have been December 9, 1994, the date Plaintiff Stein filed his charge. Plaintiffs Carter, Tuggle, and Agee should not have been permitted to opt into this action. The jury verdicts in favor of these three Plaintiffs must therefore be reversed. II. SUFFICIENCY OF THE EV- . IDENCE AS TO PLAINTIFFS HIPP, STELL, AND STEIN Liberty National argues Plaintiffs presented insufficient evidence of a pattern and practice of age discrimination. Liberty National further argues the district court should have granted its motion for judgment as a matter of law (JMOL) on each Plaintiffs claims of age discrimination. We will first address the pattern and practice evidence, and we will then address the Plaintiffs’ individual claims. We conclude the case should not have been allowed to proceed as a pattern and practice case. We further conclude the jury ver-diets in favor of the individual Plaintiffs must be reversed because they presented insufficient evidence of constructive discharge in violation of ADEA. ■ A. Sufficiency of Pattern and Practice Evidence Plaintiffs in this case proceeded under a pattern and practice theory of intentional discrimination. See generally EEOC v. Joe’s Stone Crab, Inc., 220 F.3d 1263, 1286 (11th Cir.2000). In a pattern and practice case, plaintiffs must establish by a preponderance of the evidence not only that the employer discriminated against certain individuals, but that “discrimination was the company’s standard operating procedure — the regular rather than the unusual practice.” Int’l Bhd. of Teamsters v. United States, 431 U.S. 324, 336, 97 S.Ct. 1843, 1855, 52 L.Ed.2d 396 (1977). To meet this burden, plaintiffs must “prove more than the mere occurrence of isolated or accidental or sporadic discriminatory acts.” Id. (internal quotation marks omitted); Maddox v. Claytor, 764 F.2d 1589, 1556-57 (11th Cir.1985). When plaintiffs successfully establish a pattern and practice of employment discrimination, a rebuttable presumption that each plaintiff was a victim of discrimination obtains, and the burden shifts to the employer to prove that each individual employment decision was not made in furtherance of its illegal policy. Cox v. Am. Cast Iron Pipe Co., 784 F.2d 1546, 1559 (11th Cir.1986) (citing Teamsters, 481 U.S. at 359, 97 S.Ct. at 1866). The employer may meet this burden with “clear and convincing evidence that job decisions made when the discriminatory policy was in force were not made in pursuit of that policy.” Id. (citing Teamsters, 431 U.S. at 362, 97 S.Ct. at 1868). The jury in this case found by a preponderance of the evidence that Liberty National engaged in a pattern and practice of age discrimination. The jury further found that Liberty National did not prove it operated outside the pattern and practice with respect to each of the prevailing Plaintiffs. The jury also found that each prevailing Plaintiff had “suffered a tangible job detriment, such as discharge, constructive discharge, demotion or loss of substantial job benefits.” Plaintiffs proceeding under a pattern and practice theory often introduce statistics to bolster their claims of discrimination. See, e.g., Teamsters, 431 U.S. at 339-40, 97 S.Ct. at 1856-57; Maddox, 764 F.2d at 1556; Coates v. Johnson & Johnson, 756 F.2d 524, 532 (7th Cir.1985); In re W. Dist. Xerox Litig., 850 F.Supp. 1079, 1084-85 (W.D.N.Y.1994). Plaintiffs in this case did not introduce statistical evidence at trial. Liberty National urges us to hold Plaintiffs’ pattern and practice evidence insufficient as a matter of law because Plaintiffs failed to provide statistical evidence to support their claims. While we note that statistical evidence is often extremely useful, we decline to announce the broad rule Liberty National requests. See, e.g., Xerox, 850 F.Supp. at 1085 (the absence of statistical evidence is not always fatal in pattern and practice cases, but when either statistical evidence or anecdotal “evidence is missing altogether, the other must be correspondingly stronger”). We nevertheless do not think this case should have been allowed to proceed as a pattern and practice case. Even assuming the truth of all their allegations, Plaintiffs did not prove that age discrimination was the “standard operating procedure” of the company regarding retention of District Managers. WTiile the absence of statistical evidence is not necessarily fatal to Plaintiffs’ claims, they presented only vague allegations of a policy of forcing out older District Managers. Plaintiffs argue on appeal that the evidence at trial showed that “Liberty National’s discriminatory policies emanated from its chief executive officer and were uniformly enforced by the company’s regional vice presidents. Liberty National’s top executives employed a standardized set of harassing tactics and techniques.” Plaintiffs’ evidence at trial did not, however, show such a widespread pattern of discrimination. Rather, Plaintiffs alleged an amorphous “policy” that was mainly implemented by one man, Andy King. Plaintiff Hipp testified he noticed a pattern of older District Managers leaving the company. He did not, however, know how many District Managers over age 40 remained at Liberty National, and he did not know whether the average age of District Managers had increased or decreased since the alleged policy began. Plaintiff Stell testified he would have stayed at Liberty National if he had been transferred out of King’s division. This testimony indicates age discrimination was not Liberty National’s “standard operating procedure.” Rather, if age discrimination existed at all with respect to the treatment of District Managers, it was apparently confined to King’s division. Based on these facts, we conclude Plaintiffs presented insufficient evidence of a pattern and practice of age discrimination. B. Individual Constructive Discharge Claims In reviewing the denial of a motion for JMOL, this Court uses the following standard; When considering whether or not a ruling on a motion for directed verdict or for judgment notwithstanding the verdict should be upheld, the standard of review to be applied by this Court is the same as that applied by the district court. Thus, we consider all the evidence, and the inferences drawn therefrom, in the light most favorable to the nonmoving party. If the facts and inferences point overwhelmingly in favor of one party, such that reasonable people could not arrive at a contrary verdict, then the motion was properly granted. Conversely, if there is substantial evidence opposed to the motion such that reasonable people, in the exercise of impartial judgment, might reach differing conclusions, then such a motion was due to be denied and the case was properly submitted to the jury. It bears repeating that a mere scintilla of evidence does not create a jury question. Motions for directed verdict and for judgment notwithstanding the verdict need not be reserved for situations where there is a complete absence of facts to support a jury verdict. Rather, there must be a substantial conflict in evidence to support a jury question. Carter v. City of Miami, 870 F.2d 578, 581 (11th Cir.1989) (footnotes omitted). Plaintiffs allege they were constructively discharged in violation of ADEA. The fact that we have found Plaintiffs’ evidence insufficient as a matter of law to establish a pattern and practice of discrimination does not necessarily mean Plaintiffs are not entitled to relief on their individual claims. Cf. Cooper v. Fed. Reserve Bank of Richmond, 467 U.S. 867, 876-78, 104 S.Ct. 2794, 2799-2800, 81 L.Ed.2d 718 (1984) (judgment in class action that employer had not engaged in pattern and practice of discrimination did not preclude filing of separate individual actions by class members). We will therefore examine Plaintiffs’ individual constructive discharge claims. “[T]he plaintiff in an employment discrimination lawsuit always has the burden of demonstrating that, more probably than not, the employer took an adverse employment action against him on the basis of a protected personal characteristic.” Wright v. Southland Corp., 187 F.3d 1287, 1292 (11th Cir.1999). See also Benson v. Tocco, Inc., 113 F.3d 1203, 1207-08 (11th Cir.1997); Poole v. Country Club of Columbus, Inc., 129 F.3d 551, 553 n. 2 (11th Cir.1997). We have long recognized that constructive discharge can qualify as an adverse employment decision under ADEA. See Poole, 129 F.3d at 553 n. 2; Maddow v. Procter & Gamble, 107 F.3d 846, 852 (11th Cir.1997); Stamey v. Southern Bell Tel. & Tel. Co., 859 F.2d 855, 859-60 (11th Cir.1988). The jury in this case found that Plaintiffs were constructively discharged in violation of ADEA. The district court denied Liberty National’s motion for JMOL on this issue, and Liberty National appeals this denial. The threshold for establishing constructive discharge in violation of ADEA is quite high. In evaluating constructive discharge claims, we do not consider the plaintiffs subjective feelings. Instead, we employ an objective standard. Doe v. Dekalb County Sch. Dist., 145 F.3d 1441, 1450 (11th Cir.1998). “To successfully claim constructive discharge, a plaintiff must demonstrate that working conditions were ‘so intolerable that a reasonable person in [his] position would have been compelled to resign.’ ” Poole, 129 F.3d at 553 (quoting Thomas v. Dillard Dep’t Stores, Inc., 116 F.3d 1432, 1433-34 (11th Cir.1997)). The standard for proving constructive discharge is higher than the standard for proving a hostile work environment. Landgraf v. USI Film Prods., 968 F.2d 427, 430 (5th Cir.1992) (“To prove constructive discharge, the plaintiff must demonstrate a greater severity or pervasiveness of harassment than the minimum required to prove a hostile working environment.”), aff'd, 511 U.S. 244, 114 S.Ct. 1483, 128 L.Ed.2d 229 (1994); see also Steele v. Offshore Shipbuilding, Inc., 867 F.2d 1311, 1316-18 (11th Cir.1989) (affirming district court’s conclusion that Title YII plaintiffs were subjected to hostile work environment, but were not constructively discharged); Huddleston v. Roger Dean Chevrolet, Inc., 845 F.2d 900, 905-06 (11th Cir.1988) (same). This circuit has required pervasive conduct by employers before finding that a hostile work environment existed or a constructive discharge occurred. See, e.g., Mendoza v. Borden, Inc., 195 F.3d 1238, 1247 (11th Cir.1999) (en banc) (no sex-based hostile work environment where male supervisor (1) told female employee he was “getting fired up;” (2) rubbed his hip against employee’s hip while smiling and touching her shoulder; (3) twice made a sniffing sound while looking at employee’s groin area; and (4) constantly followed employee and stared at her in a very obvious manner), cert. denied, 529 U.S. 1068, 120 S.Ct. 1674, 146 L.Ed.2d 483 (2000); Poole, 129 F.3d at 553 (issue of fact existed, precluding summary judgment for employer on constructive discharge claim, where plaintiff was “[sjtripped of all responsibility, given only a chair and no desk, and isolated from conversations with other workers”); EEOC v. Massey Yardley Chrysler Plymouth, 117 F.3d 1244, 1247-48 and nn. 2 and 4 (11th Cir.1997) (affirming jury finding of age-based hostile work environment and constructive discharge where supervisors (1) made offensive comments regarding effects of aging on plaintiffs body; (2) told her she was too old for the clothes she was wearing and should wear “old lady dresses” instead; (3) asked if she was “having any hot flashes;” (4) remarked that she was becoming “senile” or “losing it,” and asked “whether Alzheimer’s disease was setting in early;” (5) told her she would “just have to get used to” these comments since she “was, after all, ‘an old lady’ ”); Morgan v. Ford, 6 F.3d 750, 752, 756 (11th Cir.1993) (issue of fact as to sex-based constructive discharge where plaintiffs supervisor (1) repeatedly asked her out on dates; (2) constantly “hovered” around her after she turned him down; (3) told her she “had not had a real man until she had him;” (4) made offensive, sexually charged comments; and (5) electronically monitored her conversation); Hill v. Winn-Dixie Stores, 934 F.2d 1518, 1527 (11th Cir.1991) (affirming district court’s grant of judgment notwithstanding the verdict because evidence of reprimands, criticism, and supervisor’s withdrawal of support was insufficient to prove constructive discharge, especially in light of employer’s efforts to remedy the situation after plaintiffs resignation); Wardwell v. Sch. Bd. of Palm Beach County, 786 F.2d 1554, 1557-58 (11th Cir.1986) (reversing as clearly erroneous the district court’s finding of sex-based constructive discharge where female plaintiff was passed over for promotion in favor of less experienced male and was assigned extra duties). Plaintiff Hipp Plaintiff Hipp was District Manager in Ft. Myers, Florida, before he resigned in August 1993 at age 43. On June 23, 1994, Hipp filed an individual charge of discrimination with the Florida Commission on Human Relations (FCHR). This charge alleged he was forced to resign because of his age. At trial, he testified that Regional Vice President Andy King, and, to a lesser extent, Regional Vice President Don Horton, harassed him until he felt he had no choice but to resign. He said the criticism he received prior to 1991 was constructive and helpful, but after 1991, “it was nothing but harassment.” Hipp testified that Liberty National’s C.E.O., C.B. Hudson, said on at least one occasion that “older people are harder to get to change,” and that King and Horton made similar statements. Hipp did not allege King and Horton made these statements in reference to Hipp himself, but rather alleged they were evidence of widespread age animus within the company. Hipp’s claims mainly centered around two confrontations with King, as well as several smaller incidents. Hipp’s first confrontation with King occurred at an April 1993 meeting he had with King and Horton. Hipp complained that the meeting was supposed to be in his district office at 8 a.m., but around 8:30 or 8:45, King and Horton called and asked Hipp to meet at a coffee shop instead. Hipp claimed that at the meeting, in a public area in the restaurant, King “verbally attacked [him] in a manner that [he had] never been spoken to before.” Hipp said the attack related to every aspect of his job, but King’s main concern was the negative growth rate in Hipp’s district at the time. Hipp denied that his growth i'ate was negative and believed his own numbers were more accurate than King’s, but King got very angry when Hipp questioned his numbers. King and Horton told Hipp he should quit if he was unable to do the job. Hipp said he would not quit; they would have to fire him. King and Horton then told Hipp to summon his sales managers to the coffee shop, and, when he complied, Bang and Horton berated the sales managers in much the same manner as they had Hipp. Hipp’s second confrontation with King concerned a conflict between the two men regarding an agent in Hipp’s district who accused another agent and a sales manager of sexual harassment. Hipp disagreed with King about the punishment the accused harassers should receive and thought they were being treated unfairly because the company had not confirmed the harassment. At a weekend retreat, King approached Hipp and told him he wanted to discuss the harassment problem and the “lousy job” Hipp had done in controlling the situation. Hipp wanted to have the conversation at another time because both men had been drinking, but King refused. King berated Hipp and told him he was a “terrible district manager.” Hipp walked away, but King followed him. King apologized the next day, but Hipp was upset that the apology was private because the humiliation had been public. Hipp refused to shake King’s hand. ADEA does not guarantee employees a stress-free working environment. While we might disagree with the behavior of Horton and King, we cannot find that it rises to the level of constructive discharge in violation of ADEA. See, e.g., Clowes v. Allegheny Valley Hosp., 991 F.2d 1159, 1162 (3d Cir.1993) (“[A] constructive discharge claim based solely on evidence of close supervision of job performance must be critically examined so that the ADEA is not improperly used as a means of thwarting an employer’s nondiscriminatory efforts to insist on high standards.”); Bristow v. Daily Press, Inc., 770 F.2d 1251, 1255 (4th Cir.1985) (“Every job has its frustrations, challenges, and disappointments .... An employee is protected from a calculated effort to pressure him into resignation through the imposition of unreasonably harsh conditions, in excess of those faced by his co-workers. He is not, however, guaranteed a working environment free of stress.”). Hipp also complained it was difficult for him to abide by Liberty National’s alleged policy of not hiring agents over age 55, because his district, Fort Myers, Florida, had a large elderly population. He talked to Horton about this problem, but Horton said King would not allow Hipp to violate the over-55 policy. Hipp also testified that Horton told him to fire an agent named Wes Adams because Adams was too old. Hipp wanted to allow Adams to work two more years so Adams would receive retirement benefits, and he ultimately succeeded in doing so. Hipp also stated that'after he was promoted to District Manager, a sales manager in his district, a Mr. Johnson, was giving him trouble. He later found out Mr. Johnson was supposed to be promoted to the District Manager position Hipp received. Horton told Hipp that Johnson was too old to do the job. Horton also made negative comments about the age of Hipp’s secretary. Hipp stated that these comments “bothered” him. The testimony that Liberty National would not allow Hipp to hire agents over age 55 fails to show Hipp was constructively discharged because of his own age, 43. As noted above, this policy did not apply to Hipp. See supra note 30. Furthermore, if the policy existed, the evidence is that it applied equally to all districts, and was not implemented to burden Hipp. According to Hipp’s testimony, the policy was intended to reduce the company’s costs in paying benefits for agents over 55, not to force District Managers in their 40s to leave the company. Similarly, even if Hipp was told to fire Adams because Adams was too old, this testimony does nothing to establish that Hipp was constructively discharged because of his own age. Likewise, the comments regarding Johnson and Hipp’s secretary, although perhaps offensive, fail to establish that Hipp was subjected to a hostile environment or constructively discharged because of his own age. Hipp testified about several other alleged indignities he suffered in the latter .years of his employment with Liberty National. He complained that agents, Sales Managers, and District Managers began to receive less training after Hudson joined the company, and he began to receive more personnel transaction memos (PTMs) criticizing his performance. He complained that the company began to emphasize bank-budget or bank-draft sales over field-collection sales, but failed to provide any training in this new sales method. Hipp asked King to provide training for the agents, but King told him to train the agents himself. Hipp also complained that around 1990 or 1991 the company went to a “growth bonus” instead of a bonus based on the employee’s overall performance. According to Hipp, the new bonus system was unfair. Hipp further complained that King cut expenses. He claimed King did not allow him to have a full-time secretary during part of his employment. He sometimes had to close the office during the day because he did not have a secretary, and his business suffered as a result. While it may be true that he was not allowed to have a full-time secretary, and while this restriction may have been bad business, Hipp did not testify that younger District Managers were permitted to have secretaries or in any way connect the company’s actions to his age. Hipp also complained that he was told to cut his lawn service expenses from $73 per month to $42 per month, the company-wide limit. He claimed it was impossible to trim that much money off his lawn bill, and he complained to Horton and King. King refused to make an exception for Hipp, and King said he did not care about Hipp’s problems. Hipp also testified that his office was in disrepair and needed to be remodeled. He had been trying to get it remodeled since 1986, and when he finally got it remodeled, he was not allowed to make all of the desired repairs. It may have been bad business for the company to cut Hipp’s lawn expenses and delay remodeling his office. Perhaps customers were drawn to the office because of its appealing landscaping, or perhaps they were repelled by its state of disrepair. It was not, however, age discrimination. One’s working environment does not become objectively intolerable simply because it becomes less attractive. Hipp also complained that all districts were told to keep their expenses at the same level, even though the needs of the districts varied widely. He was charged for going over the budget on janitorial and landscaping services, even though he did not exceed the total budget for all items. Again, Hipp did not allege that the expense targets were intended to force older District Managers into retirement, or even that the expense targets had a disproportionately harsh impact on districts run by older District Managers. Rather, it appears from the testimony that the expense targets applied to all District Managers. In fact, Hipp admitted on cross-examination that all District Managers were sent a letter stating that they would be financially responsible for any over-budget amounts for janitorial or landscaping services. The fact that Hipp did not like complying with company-wide budget limitations does not mean he was constructively discharged in violation of ADEA. Hipp testified he felt forced to resign, but he did not know at the time that age was a factor. He said the day he left, he did not know why he was not being treated well; his “biggest problem” was that he “could not figure out why” he was being treated poorly. He discussed his unhappiness with Horton, and Horton told him things were "worse in the home office. Hipp testified that it took him “almost a year” before he could “figure it out — what was going on in the company.” He started seeing other older District Managers leaving the company. The more he thought about it, he realized Liberty National was forcing out the older District Managers. At trial, under examination by his own counsel, Hipp made the following statements: Q: At the time you resigned, were you having physical and emotional reactions to the situation that you were under? A: Yes. Q: Okay. Did you consider those — the situation you were under to be something you could tolerate? A: Yes. Q: And do you consider yourself a reasonable person? A: Yes. As noted above, to prove constructive discharge, a plaintiff must prove his working conditions are so intolerable that a reasonable person would feel he had no choice but to resign. Hipp stayed for several months after the allegedly discriminatory confrontations, suggesting the confrontations were not so intolerable as to leave him with no choice but to resign. As quoted above, Hipp himself admitted at trial that his working conditions were tolerable. He also testified he did not believe at the time he resigned that he was being discriminated against on the basis of his age. Hipp’s termination fo