MEMORANDUM OPINION & ORDER
BABCOCK, District Judge.
Plaintiffs, who are all former employees of defendant, Storage Technology Corporation [677]*677(Storage Tek), assert claims for wrongful discharge and for violations of the Employee Retirement Income Security Act of 1974 (ERISA), the Age Discrimination in Employment Act, 29 U.S.C. § 621 et seq. (ADEA), and the Americans with Disabilities Act, 42 U.S.C. § 12101 et seq. (ADA). Plaintiffs move to certify an ADEA collective action under 29 U.S.C. § 216(b) and an ERISA class under Fed.R.Civ.P. 23. A hearing was held on the motions. For the following reasons, I will grant the motion to certify an ADEA collective action and deny the motion to certify an ERISA class action.
I.
Storage Tek, headquartered in Louisville, Colorado, produces high technology computer equipment. Plaintiffs allege in their Sixth Amended Complaint that layoffs of Colorado workers between 1989 and 1996 were based on age and Storage Tek’s perception of plaintiffs’ use or anticipated use of company benefits, such as the self-funded medical plan. At the hearing, plaintiffs limited the relevant period to between April 13,1993, and December 31,1996. Plaintiffs claim violations, inter alia, of the ADEA and ERISA. In addition, plaintiffs seek certification of a class of plaintiffs for each claim. Defendant opposes certification of either class.
II.
COLLECTIVE ACTION PURSUANT TO 29 U.S.C. § 216(b)
Plaintiffs move for an order permitting this ease to proceed as a collective action pursuant to 29 U.S.C. § 216(b) with respect to the Age Discrimination in Employment Act (ADEA) claim. Eleven of the twelve named plaintiffs request that they be designated as representative of the proposed ADEA class. Only plaintiff Pamela Wilson, who is under the age of 40, has not requested to represent the ADEA class.
The ADEA, 29 U.S.C. § 623(a) provides, in pertinent part:
It shall be unlawful for an employer—
(1) ... to discharge any individual or otherwise discriminate against any individual with respect to his compensation, terms, conditions, or privileges of employment, because of such individual’s age;
(2) 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.
Generally, plaintiffs allege the following. From April 13, 1993, through the present, Storage Tek engaged in corporate strategic planning that identified as a major corporate financial problem the high cost and perceived decreased productivity of older employees. Based upon these concerns, through its reductions-in-force, Storage Tek engaged in a pattern and practice of age discrimination that resulted in the unlawful termination of several hundred employees over the age of forty, including the representative plaintiffs. Throughout the reductions-in-force, defendant treated younger employees more favorably than older employees.
The ADEA class proposed by the plaintiff is defined as follows:
Ail non-bargaining unit employees age 40 and older and who were terminated as part of a company-wide reduction-in-force, which occurred between April 13,1993 and December 31,1996, pursuant to a series of Strategic Plans designed to decrease the average age of employees and their associated costs, as well as increase productivity, by targeting older workers for job elimination.
A. Legal Standards for § 216(b) Collective Action
The ADEA incorporates the Fair Labor Standards Act, 29 U.S.C. § 216(b), which provides that:
[a]n action [under the ADEA] ... may be maintained against any employer ... in any Federal or State court of competent jurisdiction by any one or more employees for and in behalf of himself or themselves and other employees similarly situated. No employee shall be a party plaintiff to any such action unless he gives his consent in writing to become such a party and such [678]*678consent is filed in the court in which such action is brought.
Therefore, 29 U.S.C. § 216(b) permits plaintiffs to proceed under the ADEA “for and in behalf of ... themselves and other employee similarly situated,” and I must determine whether plaintiffs and those they seek to represent are “similarly situated” for purposes of the ADEA claim. The statute does not define “similarly situated,” nor has the Tenth Circuit explained its meaning.
In Bayles v. American Medical Response of Colorado, Inc., 950 F.Supp. 1053, 1066-67 (D.Colo.1996), I adopted an ad hoc method of determining whether plaintiffs were similarly situated under § 216(b). In particular, I employ the two-step approach to § 216(b) certification adopted by several other courts. First, I must determine whether a collective action should be certified for notice purposes. Then, after discovery is completed and the case is ready for trial, I revisit the issue of certification.
At the notice stage, courts following the ad hoc method “require nothing more than substantial allegations that the putative class members were together the victims of a single decision, policy, or plan____” Id. at 1066 (citing Sperling v. Hoffman-La Roche, Inc., 118 F.R.D. 392, 407 (D.N.J.), judgment ajfd. in part, appeal dismissed in part, 862 F.2d 439 (3d Cir.1988), judgment ajfd. and remanded, 493 U.S. 165, 110 S.Ct. 482, 107 L.Ed.2d 480 (1989)). The court then makes a second determination after discovery has been completed and the case is ready for trial. Bayles, 950 F.Supp. at 1066. At this second stage, although not specifically deemed, the “similarly situated” standard is higher. Id. In determining whether plaintiffs are similarly situated after discovery is completed, courts address several factors: (1) disparate factual and employment settings of the individual plaintiffs; (2) the various defenses available to defendant which appear to be individual to each plaintiff; (3) fairness and procedural considerations; and (4) whether plaintiffs made the filings required by the ADEA before instituting suit. Bayles at 1066 (citing See Lusardi v. Xerox Corp., 118 F.R.D. 351 (D.N.J.1987), vacated in part on other grounds, 122 F.R.D. 463 (D.N.J.1988)).
I must also consider whether certification would serve the purposes and putative benefits of a collective action under § 216. The Supreme Court has identified main benefits of a collective action under § 216(b):
A collective action allows ... plaintiffs the advantage of lower individual costs to vindicate rights by the pooling of resources. The judicial system benefits by efficient resolution in one proceeding of common issue of law and fact arising from the same alleged ... activity.
Hoffmann-La Roche Inc. v. Sperling, 493 U.S. 165, 170, 110 S.Ct. 482, 486,107 L.Ed.2d 480 (1989). I must balance the putative benefits against any prejudice to the defendant and any judicial inefficiencies that may result from allowing plaintiffs to proceed collectively. Bayles at 1067.
B. Section 216(b) Analysis
In Bayles, I conditionally certified a § 216(b) class under the FLSA for notice purposes. After the plaintiffs had received the benefits of collective discovery and trial preparation, however, I decertified the class because I found and concluded that they were not similarly situated within the plain meaning of the statute. In part, I found that the benefits of proceeding collectively to trial were significantly outweighed by the judicial inefficiencies and prejudice to the defendant resulting from the disparate circumstances in which the collective plaintiffs worked. Id. at 1066-67.
Defendants argue that I should deny certification here for the same reasons. The § 216(b) class proposed by plaintiffs would include former employees of defendant who worked in markedly different circumstances and positions within the corporation. Defendants, therefore, contend that the proposed class of plaintiffs are not similarly situated, and plaintiffs’ motion should be denied. I disagree for several reasons.
First, plaintiffs’ motion seeks certification only for notice purposes. Therefore, the lower standard for conditional certification ap[679]*679plies. Plaintiffs certainly have made “substantial allegations that the putative class members were together the victims of a single ... plan.” See Sixth Am.Comp. 1HI21-28. In particular, plaintiffs allege they were the victims of a “Strategic Plan” by defendant to eliminate older employees and their associated costs. Therefore, conditional certification for notice purposes is warranted.
Defendant argues that because significant discovery has already been taken, I should treat plaintiffs motion as one for final certification and deny the motion under the higher standard articulated in Bayles. Again, I disagree that certification should be denied. First, discovery is not complete and significant economies can still be achieved by collective discovery and trial preparation. More importantly, even under the higher standard used for final certification, a collective action is appropriate.
Plaintiffs have alleged, inter alia, that defendant engaged in a “pattern and practice” of discrimination against its employees during the reductions in force from April 13, 1993, to December 31, 1996. Sixth Am. Comp. UK 27, 49. Although originally pattern and practice cases were brought only by the government, the Supreme Court has expanded the pattern and practice claim to include private litigants. See Bazemore v. Friday, 478 U.S. 385, 106 S.Ct. 3000, 92 L.Ed.2d 315 (1986). To the extent an ADEA claim is based on an allegation of a pattern and practice of discrimination, the typical burden-shifting analysis of McDonnell Douglas Corp. v. Green, 411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973), is altered. International Brotherhood of Teamsters v. United States, 431 U.S. 324, 97 S.Ct. 1843, 52 L.Ed.2d 396 (1977); EEOC v. Sandia Corp., 639 F.2d 600 (10th Cir.1980). Unlike typical disparate treatment cases, pattern and practice cases are divided into two phases: liability and remedy. Teamsters, supra at 357-62, 97 S.Ct. at 1865-68.
In the liability phase, plaintiffs carry the burden to demonstrate the existence of a pattern and practice of discrimination. In particular, plaintiffs “must establish by a preponderance of the evidence that ... discrimination was the company’s standard operating procedure — the regular rather than the unusual practice.” Id. at 336, 97 S.Ct. at 1855. Within this first phase, the burden-shifting analysis of McDonnell Douglas applies. Id. at 360, 97 S.Ct. at 1867; Sandia, 639 F.2d at 621. Plaintiffs must establish a prima facie case supporting the inference that the defendant engaged in a pattern and practice of discrimination. Id. “The burden then shifts to the employer to defeat the prima facie showing of a pattern and practice by demonstrating that the [plaintiffs’] proof is either inaccurate or insignificant.” Teamsters, supra at 360, 97 S.Ct. at 1867. The plaintiffs may then offer evidence of pretext. Sandia, supra at 621.
If plaintiffs prove a pattern and practice of discrimination, the court may fashion an award of class-wide prospective relief without further evidence. If, however, individual relief is also sought, the court must conduct a second, “remedial” phase. The term “remedial” is somewhat of a misnomer because the defendant’s liability to any individual class member must still be determined. However, for purposes of that remedial phase:
The proof of the pattern or practice supports an inference that any particular employment decision, during the period in which the discriminatory policy was in force, was made in pursuit of that policy. The [plaintiff] need only show that an alleged individual discriminatee [suffered an adverse employment decision] and therefore was a potential victim of the proved discrimination____[T]he burden then rests on the employer to demonstrate that the individual [suffered the adverse employment decision for lawful reasons.
Teamsters, supra at 362, 97 S.Ct. at 1868. It is unclear from the opinion whether the burden of persuasion or merely the burden of going forward with the evidence is shifted to the defendant in the remedial phase. See Teamsters, supra at 362 n. 50, 97 S.Ct. at 1868 n. 50; Sandia, supra at 622 (contradictorily stating: “Teamsters ... holds that in pattern and practice cases the burden of proof in the sense of the burden of persuasion shifts to the employer once a discrimina[680]*680tory pattern and practice is shown. It is not necessary here to enunciate a conclusion as to whether there is indeed a shifting of the burden of proof. ”); Gavalik v. Continental Can Co., 812 F.2d 834 (3d Cir.), cert. denied, 484 U.S. 979, 108 S.Ct. 495, 98 L.Ed.2d 492 (1987) (holding that burden of persuasion is shifted to the defendant in the remedial phase).
Defendant argues that certification is inappropriate because plaintiffs have adduced no evidence to show a pattern and practice of discrimination. I disagree. Plaintiffs have provided significant evidence to support their claim that defendant engaged in a pattern and practice of discrimination. More importantly, whether plaintiffs can meet their burden in the liability phase of the Teamsters procedure is irrelevant to the question of § 216(b) certification. I am not deciding a dispositive motion by defendant. Rather, it is critical here that the liability phase of a pattern and practice action is unaffected by the number of plaintiffs asserting the claim.
The sole issue in the liability phase is whether defendant engaged in a pattern and practice of discrimination. At that stage, plaintiffs are not required to provide specific evidence that they were individually discriminated against. Teamsters, supra at 361, 97 S.Ct. at 1867-68. Therefore, phase one of the trial will not be affected by the expansion of this claim from eleven named plaintiffs to the hundreds who might be encompassed by the proposed class. Any disparities among the employment situations of the individual plaintiffs are irrelevant during the liability phase, judicial economies are clearly served by proceeding collectively, and the defendant is not prejudiced. This represents the epitome of similarly situated plaintiffs.
Defendant’s arguments against certification are not persuasive. First, defendant argues that plaintiffs’ pattern and practice claim is a type of disparate impact claim and that disparate impact claims may not be maintained under the ADEA. I agree that disparate impact claims are not cognizable under the ADEA. Ellis v. United Airlines, 73 F.3d 999, 1007 (10th Cir.), cert. denied, — U.S. - — •, 116 S.Ct. 2500, 135 L.Ed.2d 191 (1996). However, plaintiffs’ pattern and practice claim, like the pattern and practice claim in Sandia, supra, is a disparate treatment claim.
“Disparate impact claims ... challenge employment practices that are facially neutral in their treatment of different groups but that in fact fall more harshly on one group than another and cannot be justified by business necessity.” Ellis, supra at 1006 (citations omitted). Plaintiffs here, as in Sandia, allege that defendant had a pattern and practice of intentionally discriminating against employees on the basis of age, not that a neutral policy of defendant happened to affect older employees adversely. See Sandia, supra at 610. Accordingly, plaintiffs’ claims are based on disparate treatment, and Sandia stands squarely for the proposition that a pattern and practice claim may be maintained under the ADEA for disparate treatment.
Defendant next argues that plaintiffs have failed to demonstrate compliance with the predicate administrative filing requirements. Again, I disagree. In the Tenth Circuit, it is sufficient that the representative plaintiffs file timely complaints with the EEOC on behalf of themselves and others similarly situated. Mistretta v. Sandia Corp., 639 F.2d 588, 593-95 (10th Cir.1980). All representative plaintiffs have filed such timely complaints with the EEOC. Pltf. Reply Exh. Q. Accordingly, the administrative filing requirements are satisfied for all putative class members who could have filed an EEOC complaint at the time one of the representative plaintiffs filed such a complaint. See Burt v. Manville Sales Corp., 116 F.R.D. 276, 277-78 (D.Colo.1987).
Finally, defendant argues that certification should be denied because the interests of the ADEA class will contradict the interests of the ERISA class, and the same named plaintiffs seek to represent both classes. I am not persuaded. The Supreme Court in Hazen Paper Co. v. Biggins, 507 U.S. 604, 612-13, 113 S.Ct. 1701, 1707, 123 L.Ed.2d 338 (1993), stated:
[681]*681We do not preclude the possibility that an employer who targets employees with a particular pension status on the assumption that these employees are likely to be older thereby engages in age discrimination. Pension status may be a proxy for age, not in the sense that the ADEA makes the two factors equivalent, but in the sense that the employer may suppose a correlation between the two factors and act accordingly. Nor do we rule out the possibility of dual liability under ERISA and the ADEA where the decision to fire the employee was motivated by both the employee’s age and his pension status.
See also Krueger v. New York Tel. Co., 163 F.R.D. 433 (S.D.N.Y.1995) (certifying ERISA and ADEA classes with overlapping representative plaintiffs).
Plaintiffs state that their evidence supporting the ADEA claim is essentially identical to their evidence supporting the ERISA claim. Plaintiffs intend to show that defendant terminated employees based on (1) age and (2) perceived use of benefits, including health care. Plaintiffs theory of age discrimination is dependent in large part on a showing that defendant perceived older employees as having higher health-care costs. Therefore, I agree that plaintiffs’ ADEA and ERISA claims are complimentary and do not conflict so as to preclude the same named plaintiffs from representing each class.
Accordingly, even applying the higher “similarly situated” standard articulated in Bayles, supra, I conclude that final § 216(b) certification is appropriate for the liability phase of plaintiffs’ pattern and practice claim. The liability phase will determine: (1) whether defendant engaged in a pattern and practice of discrimination against its employees in violation of the ADEA from April 13, 1993, to December 31, 1996; (2) if so, whether defendant acted willfully and wantonly; and (3) whether plaintiffs are entitled to injunctive or declaratory, class-wide relief. None of these issues is affected by the number of plaintiffs added by the proposed collective action. These added plaintiffs, however, could potentially garner significant declaratory relief that would allow them to enter the remedial phase with a presumption that they were terminated in violation of the ADEA. The efficiencies of using the collective action device for the liability phase are, therefore, plainly evident.
The question remains, then, whether certification of the ADEA class should extend only to the liability phase or should also include the remedial phase. As plaintiffs concede, the remedial phase, if necessary, would involve numerous issues individual to particular plaintiffs. However, the liability phase of the trial may bring into focus workable subclasses of plaintiffs who may be sufficiently similarly situated to proceed collectively. Therefore, if plaintiffs prevail in the liability phase, I will revisit whether the class should be decertified for the remedial phase or whether appropriate subclasses can be crafted. My only conclusion here is that certification is appropriate through the liability phase of the pattern and practice claim.
Notably, the named plaintiffs assert several other claims for relief, including a claim under the ADEA independent of whether a pattern and practice of discrimination is proved. The opt-in plaintiffs will not be permitted to assert non-pattern and practice claims because plaintiffs have made no showing that they are similarly situated to the proposed class outside of the pattern and practice context. Nevertheless, I believe that judicial economy is best served by allowing the named plaintiffs to litigate completely in the first trial all of their claims. Hence, the first trial will include both the liability phase for the class-plaintiffs’ pattern and practice claim and all claims of the representative plaintiffs, including the remedial phase of their pattern and practice claims. This will not only provide complete relief to the representative plaintiffs, but also provide “test cases” that may aid in the settlement of the class-plaintiffs’ claims should plaintiffs prevail in the liability phase.
C. Notice and Consent Forms
Before a similarly situated employee can become a party plaintiff, he or she must file a written consent with this court. 29 U.S.C. § 216(b). Upon certification of a collective action, the named plaintiffs and their counsel may contact potential members [682]*682of the class by sending them eourt>-approved notice and consent forms. See Bay les, supra; Hoffman-La Roche, supra. In addition, plaintiffs are entitled to specific discovery to determine the names and addresses of the putative class members. See Hoffman-La Roche, supra at 170, 110 S.Ct. at 486 (“The District Court was correct to permit discovery of the names and addresses of the discharged employees.”).
Plaintiffs included within their motion a proposed notice to potential class members. That notice, however, is now inaccurate. For example, the proposed notice states that only seven named plaintiffs represent the ADEA class. There are now eleven representative ADEA plaintiffs. Further, plaintiffs have limited the relevant time period for their claims to between April 13,1993, and December 31,1996, and the proposed notice reflects a period between 1989 and 1996. Further, the notice should reflect that I have certified a class only through the liability phase of the pattern and practice claim, and decertification remains possible before the remedial phase.
Accordingly, I will order the following. The ADEA class proposed by plaintiffs will be certified through the liability phase of the pattern and practice claim. Defendant shall provide plaintiffs within twenty days a list of all persons discharged from Storage Tek between April 13,1993, and December 31,1996, as part of reductions-in-force. That list will be provided, if possible, in electronically usable form. Plaintiffs shall file within twenty days thereafter a revised notice and consent form to be sent to prospective class members reflecting the extent to which I have certified a collective action pursuant to this order. Defendant will then have ten days to object to the form of that notice, and plaintiffs will have ten days to respond to the objections. I will revise the notice and consent form, if necessary, before it is distributed.
III.
CLASS CERTIFICATION PURSUANT TO FED.R.CIV.P. 23
Plaintiffs seek certification of a class pursuant to Fed.R.Civ.P. 23 for violation of ERISA. Plaintiffs each request that they be designated as representative of the ERISA class. ERISA, 29 U.S.C. § 1140 (section 510) prohibits:
any person to discharge, fine, suspend, expel, discipline, or discriminate against a participant or beneficiary for exercising any right to which he is entitled under the provisions of an employee benefit plan ... or for the purpose of interfering with the attainment of any right to which such participant may become entitled.
To establish a prima facie case of ERISA benefits discrimination, plaintiffs must demonstrate: “(1) prohibited employer conduct; (2) taken for the purpose of interfering (3) with the attainment of any right to which the employee may become entitled.” Maez v. Mountain States Telephone and Telegraph, Inc., 54 F.3d 1488, 1504 (10th Cir.1995) (quoting Gavalik v. Continental Can Co., 812 F.2d 834, 852 (3rd Cir.1987)). Again, plaintiffs seek to represent the proposed class only to the extent their claim is based on a pattern and practice of discrimination, which is subject to the bifurcated approach of Teamsters, supra. Gavalik, supra. If the representative plaintiffs show a pattern and practice of discrimination in the liability stage of the trial, the class members are entitled to a presumption that they were individually discriminated against.
Generally, plaintiffs allege the following. From April 13, 1993, through the present, Storage Tek engaged in corporate strategic planning which identified as a major corporate financial problem the high cost of employees with health problems or potential health problems. Sixth Am. Comp. 1Í 22. Based upon these concerns, Storage Tek engaged in a pattern and practice of benefits discrimination, through its layoffs, which resulted in the unlawful termination of thousands of employees. Sixth Am. Comp. fU 27, 29-40. For example, the named plaintiffs allege they were laid off for a variety of reasons relating to health-care costs, including age, personal illness, illness of an insured family member, etc. In general, Storage Tek treated employees who were not perceived as placing a burden on the company’s benefits plans more favorably than those who [683]*683were perceived as placing a burden on the company’s benefits plans. Sixth Am. Comp, f f 29-42.
A. Legal Standards for Class Certification under Fed.R.Civ.P. 28
Rule 23 sets forth a two-part test for the maintenance of a class action. First, plaintiffs must satisfy the four prerequisites of Rule 23(a). Second, one of the three subsections of Rule 23(b) must also be satisfied. Plaintiffs bear the burden of demonstrating that all the prerequisites for class designation are met. Rex v. Owens, 585 F.2d 432, 435 (10th Cir.1978). Class action certification is discretionary with the trial court judge, Anderson v. City of Albuquerque, 690 F.2d 796, 799 (10th Cir.1982), and may be altered, expanded, subdivided or abandoned as the case develops. See Daigle v. Shell Oil Co., 133 F.R.D. 600 (D.Colo.1990); Dubin v. Miller, 132 F.R.D. 269, 270-75 (D.Colo.1990). Because of the flexible nature of class certification, I am to favor the procedure. Esplin v. Hirschi, 402 F.2d 94, 99 & 101 (10th Cir.1968), cert. denied, 394 U.S. 928, 89 S.Ct. 1194, 22 L.Ed.2d 459 (1969). I will certify, however, only after rigorous analysis of the Rule 23(a) prerequisites. General Tel. Co. of Southwest v. Falcon, 457 U.S. 147, 161, 102 S.Ct. 2364, 2372-73, 72 L.Ed.2d 740 (1982); Reed v. Bowen, 849 F.2d 1307, 1309 (10th Cir.1988).
I agree with plaintiff that at least the liability phase of plaintiffs ERISA claim is ideal for class treatment under Fed.R.Civ.P. 23(b)(2) because it involves only class-wide declaratory and/or injunctive relief. Nevertheless, unlike in plaintiffs’ ADEA collective action, I conclude that plaintiffs have failed to define adequately the proposed class. ERISA class certification, therefore, is not warranted at this time.
B. Rule 28 Analysis
To proceed as a class action, the prerequisites of Fed.R.Civ.P. Rule 23(a) must be satisfied. Rule 23(a) provides that:
One or more members of a class may sue or be sued as representative parties on behalf of all only if:
1. the class is so numerous that joinder of other members is impracticable (numerosity);
2. there are questions of law or fact common to the class (commonality);
3. the claims or defenses of the representative parties are typical of the claims or defenses of the class (typicality); and
4. the representative parties will fairly and adequately protect the interest of the class (adequacy of representation).
For the following reasons, I conclude that the prerequisites of Rule 23(a) are not satisfied.
To satisfy the requirements of Rule 23(a), plaintiffs must first adequately define the class so that potential class members can be identified. Joseph v. General Motors Corp., 109 F.R.D. 635, 637 (D.Coio. 1986). Indeed, this fundamental consideration of class existence is the most basic consideration of all and the one that unifies and relates the narrower, more technical requirements of class certification. Cyr v. Walls, 439 F.Supp. 697, 703 (N.D.Tex.1977). However, the class does not have to be so ascertainable that every potential member can be identified at the beginning of the action. Joseph v. General Motors Corp., supra at 637. Nevertheless, the class description must be sufficiently definite so that it is “administratively feasible” for the court to determine whether a particular individual is a member. Davoll v. Webb, 160 F.R.D. 142, 144 (D.Colo.1995). Plaintiffs must then establish that it is so numerous that joinder of all members is impracticable. Id. at 144.
Plaintiffs propose the following ERISA class definition:
All non-bargaining unit employees who were laid off by Defendant because of their perceived use or perceived anticipated use of company benefits, due to their age, actual utilization of corporate benefit plans, and life style choices, including smoking and weight, pursuant to company wide layoffs between April 13,1993, and December 31,1996.
Storage Tek states that plaintiffs’ class definition is “nebulous at best and is, argu[684]*684ably, entirely illusory.” Resp. p. 8. For example, defendant objects to defining the class in terms of “perceptions” and contends that the class is overinclusive because “life style choices” is undefinable.
I am not concerned with defining the class in terms of perceptions. Other courts have permitted class actions based on the alleged “perceptions” of the defendant. For example, a class defined as “persons who are alcoholics, or who are believed by the [defendant] to be alcoholics or alcohol abusers ...” was certified. Smith v. McGriff, 434 F.Supp. 673, 677 (M.D.Ala., 1976). Also, class action certification was granted based on the beliefs and perception of defendants that plaintiffs were homosexual after the district court refined the proposed class definition by establishing sub-classes. Cyr v. Walls, 439 F.Supp. 697, 703 (N.D.Tex.1977). Logically, unlawful discrimination based on perceptions of the defendant is no less harmful to the recipient than unlawful discrimination based upon the actual existence of the basis for the discrimination. Importantly, ERISA prohibits discrimination based on actual use or perceived use of company benefits. See Gavalik, supra.
Nevertheless, the proposed class definition is overinclusive. As I understand plaintiffs’ claims, they assert that a certain class of persons were discharged because defendant perceived them an above-average burden on the company’s benefits plan; however, the proposed class definition does not delineate that group with sufficient precision. Rather, the putative class is defined as persons who were “laid off by Defendant because of their perceived use or perceived anticipated use of company benefits.” The proposed class is further defined by the reasons for defendant’s perceptions, specifically — age, actual use of company benefits, and life style choices. Smoking and weight are included as examples of life style choices but the class is not limited by them.
Nothing in the proposed class definition distinguishes between an employee who is perceived to use a “small” or “average” amount of corporate benefits from one who is perceived to use a “large” amount of corporate benefits. This is a critical distinction. Pursuant to the Teamsters regimen, I must decide first whether plaintiffs can prove a class-wide pattern and practice of discrimination. Before I can decide that, I need to define the precise class that the pattern and practice of discrimination is alleged to have affected.
Moreover, if plaintiffs’ prevail in the liability phase of the ERISA claim, they must also prevail at the remedial phase before they can obtain any individual relief. During the remedial phase, individual plaintiffs will have the initial burden to demonstrate that they are members of the class. Only then will the presumption attach that the defendant laid that individual off because of his membership in the class. Teamsters, supra at 362, 97 S.Ct. at 1868.
In the remedial phase of an ADEA claim, individual class members need only prove that they were employed by defendant and over the age of forty when laid off to gain the presumption attendant a showing of a pattern and practice of discrimination. Id. As currently constituted, however, I cannot ascertain what putative ERISA class members would need to prove to avail themselves of the presumption.
For example, assume: (1) I grant declaratory judgment that defendant engaged in a pattern and practice of discriminating against persons “because of their perceived use or perceived anticipated use of company benefits, due to their ... actual utilization of corporate benefit plans”; (2) an individual who was laid off during the relevant time period receives notice of the judgment; and (3) she had no major health problems during her tenure but she did “use” the company health plan. What does she need to prove during the remedial phase to receive the benefit of the pattern and practice presumption? Does she need to prove only that defendant perceived that she was using or would use the company benefits based on her past use of the plan? Plaintiffs’ allegations and evidence suggest that she would need to prove more — perhaps that defendant perceived that she was costing or would cost the company an abnormally high amount in company benefits. A class definition including that limitation would make proof of class [685]*685membership difficult, but not impossible. The current proposed class definition, however, is simply too broad.
It is not sufficient that plaintiffs define the class in terms of persons who were perceived to cost the company enough to be targeted for lay off. Such a definition would eviscerate the presumption provided by Teamsters. Individual putative class members would be required to show that they were laid off because of their perceived costs to the company to receive the presumption that they were laid off because of their perceived costs to the company. Joseph Heller, himself, might buckle under the futility of that arrangement.
Although I believe that certification of an ERISA class is possible in this case, it is not for me to revise the proposed class definition for plaintiffs. I suggest, however, that class definition might best be focused by anticipating what precise declaratory relief is sought following the liability phase. In their reply brief, plaintiffs state: “For clarification, Plaintiffs are seeking an injunction prohibiting violations of § 510 of the ERISA by Defendant and a declaration that Defendant is and has been violating § 510.” That type of relief, however, is useless unless accompanied by a declaration of how and against whom such a violation is being or has been implemented. That is why class definition is so critical. It is not clear whether plaintiffs seek a declaration (1) only that defendant discriminated against persons who they perceived as having high benefit-usage or (2) that defendant both perceived certain classes of persons as having high benefit-usage and discriminated on that basis.
I anticipate that the revised class definition chosen by the plaintiffs may be driven by the evidence they have adduced in discovery. If, for example, plaintiffs believe the evidence supports that defendant unilaterally discriminated against all employees it perceived as a using an appropriately defined level of present or future benefits, they may rework the “global” class definition presently proposed. If, however, the evidence adduced provides for greater specificity the plaintiffs may choose to narrow the class definition.
For example, if the evidence supported it, plaintiffs could limit the class to those employees who either (1) were over a certain age, (2) actually used greater than an appropriately defined level of corporate benefits in the past, or (3) made specific life style choices, such as choosing to smoke, maintaining a certain level of obesity, etc. Thus, plaintiffs may choose to craft the class in terms of the subclasses they allege served as the basis for defendant’s perceptions regarding benefit-usage. This would facilitate notice if I chose to certify' a Rule 23(b)(3) class for the remedial phase and would simplify proof of class membership for individuals seeking relief in the remedial phase.
Because plaintiffs seek only a Rule 23(b)(2) certification for the liability phase, notice to class members is not necessary, and precise class definition is not as critical as it might otherwise be. In addition, because this claim will be tried to the court, I will have the opportunity to grant declaratory relief to whatever portion of the proposed class I find has been discriminated against. Nevertheless, I will not certify a class that, if plaintiffs prevail in the liability phase, would be unworkable for the remedial phase. At a minimum, plaintiffs must narrow the class to include only those who defendant perceived as using, at that time or in the future, greater than an appropriately defined of company benefits. Plaintiffs bear the burden of establishing the precise boundaries of that limitation in light of their allegations and the evidence adduced in discovery.
Concluding, as I have, that the class definition is inadequate, I cannot address the technical requirements of Rule 23. I will permit plaintiffs to submit a second motion for certification of an ERISA class pursuant to this order. If plaintiffs choose to file such motion, I will address the technical requirements of Rule 23 in relation to the revised class definition.
Accordingly, it is ORDERED that:
1. ' Plaintiffs’ motion to certify a collective action for their ADEA claim is GRANTED.
2. Defendant shall provide plaintiffs within twenty days a list of all persons discharged from Storage Tek between April 13, [686]*6861993, and December 31, 1996, as part of reductions-in-force. That list will be provided, if possible, in electronically usable form.
3. Plaintiffs shall file within twenty days thereafter a revised notice and consent form to be sent to prospective class members reflecting the extent to which I have certified a collective action pursuant to this order. Defendant will then have ten days to object to the form of that notice, and plaintiffs will have ten days to respond to the objections.
4. Plaintiffs’ motion to certify an ERISA class action is DENIED without prejudice.