Cohen v. Brown University, 879 F. Supp. 185 (D.R.I. 1995)
March 29, 1995
BROWN UNIVERSITY, Vartan Gregorian, and David Roach, Defendants.
United States District Court, D. Rhode Island.
*186 *187 Lynette J. Labinger, Roney & Labinger, Amato A. DeLuca, Mandell, DeLuca & Schwartz, Raymond A. Marcaccio, Blish & Cavanagh, Providence, RI, Arthur H. Bryant, Trial Lawyers for Public Justice, Washington, DC, Herbert B. Newberg, Sandra L. Duggan, Kronfeld, Newberg & Duggan, Philadelphia, PA, Leslie A. Brueckner, Trial Lawyers for Public Justice, Washington, DC, for plaintiffs.
Julius C. Michaelson, Jeffrey S. Michaelson, Michaelson & Michaelson, Beverly E. Ledbetter, Providence, RI, Walter B. Connolly, Miller, Canfield, Paddock and Stone, Detroit, MI, for defendants.
OPINION AND ORDER
PETTINE, Senior District Judge.
This is a class action lawsuit charging Brown University, its president, and its athletic director (collectively "Brown") with violating Title IX of the Education Amendments of 1972, 20 U.S.C. งง 1681-1688 (1988) ("Title IX"). Specifically, the plaintiff class, which consists of all present and future Brown University women students and potential students who participate, seek to participate, and/or are deterred from participating in intercollegiate athletics funded by Brown, contends that defendants have discriminated against women in the operation of Brown's intercollegiate athletic program. After carefully considering the legal arguments and evidence presented throughout the thirty day trial on the merits, I find Brown University to be in violation of Title IX.
This suit was initiated in response to the demotion of the women's gymnastics and volleyball teams at Brown from full varsity to club varsity status in May of 1991. Up until that time, both teams were fully funded by the University. At the same time that Brown demoted these two women's teams, and apparently in an effort to comply with its understanding of Title IX's directives, Brown also demoted two men's fully funded varsity teams, water polo and golf, to club varsity status. At that time, all four teams were stripped of their university funding and most of their varsity privileges. Plaintiffs allege that, against a background in which men at Brown already enjoyed a disproportionately large share of the resources expended on athletics and of the intercollegiate participation opportunities afforded to student athletes, the facially even-handed demotions perpetuated Brown's discriminatory treatment of women.
Prior to the trial on the merits, this Court granted plaintiffs' motion for class certification and denied defendants' motion to dismiss. This Court subsequently heard fourteen days of testimony on plaintiffs' motion for a preliminary injunction. I ordered that the women's gymnastics and volleyball teams be reinstated to fully funded varsity status and prohibited Brown from eliminating or reducing the status or funding of any existing women's intercollegiate varsity team until the case was resolved on the merits. Cohen v. *188 Brown Univ., 809 F. Supp. 978 (D.R.I.1992). "After mapping Title IX's rugged legal terrain and cutting a passable swath through the factual thicket that overspreads the parties' arguments," the First Circuit affirmed. Cohen v. Brown Univ., 991 F.2d 888, 891 (1st Cir. 1993).
At the time of the preliminary injunction, there was virtually no case law on point. Since issuance of the First Circuit's opinion, a number of other circuits have been faced with Title IX athletic discrimination suits. See, e.g., Horner v. Kentucky High Sch. Athletic Ass'n., 43 F.3d 265 (6th Cir.1994); Kelley v. Board of Trustees, 35 F.3d 265 (7th Cir.1994), cert. denied, ___ U.S. ___, 115 S. Ct. 938, 130 L. Ed. 2d 883 (1995); Favia v. Indiana Univ. of Pennsylvania, 7 F.3d 332 (3d Cir.1993); Roberts v. Colorado State Bd. of Agriculture, 998 F.2d 824 (10th Cir.), cert. denied, ___ U.S. ___, 114 S. Ct. 580, 126 L. Ed. 2d 478 (1993). The Third, Sixth, Seventh, and Tenth Circuits are in agreement with the First Circuit's interpretation of the law and relevant agency documents. These developments have been explored in a number of recent law review articles. In the instant case, defendants have advanced several provocative arguments that require a thorough examination of the relevant law. In addition, the unusual two-tiered structure of Brown's intercollegiate athletic program presents a unique factual situation requiring this Court to engage in an exhaustive analysis of Title IX and its regulatory complements.
First, I will chronicle the factual background of this case. Second, I will address several preliminary matters. Third, I will outline the legal framework of Title IX and the implementing regulations and interpretation. Fourth, I will review the degree of deference due these agency documents. Fifth, I will set forth this Court's interpretation of the law and will explain why the alternative interpretations offered by counsel must be rejected. Finally, I will discuss the specifies of this case in light of my legal conclusions.
II. FACTUAL BACKGROUND
Brown University is a Division I institution within the National Collegiate Athletic Association ("NCAA"). As such, Brown participates at the highest level of NCAA competition. Brown currently offers an extensive athletic program for its students. At the *189 intercollegiate level it funds 13 sports for women and 12 sports for men. Additionally, it recognizes, but does not fund, several sports as "donor-funded" varsities (four men's teams and three women's teams). Although the number of varsity sports offered to men and women are equal, the selection of sports offered to each gender generates far more individual positions for male athletes than for female athletes.
Brown provides the financial resources to sustain the budgets of the "university-funded" varsities, whereas, it requires donor-funded teams to raise their own funds through private donations in order to exist. Brown also provides certain services and privileges to the university-funded varsities but not to the donor-funded varsities.
A consequence of this two-tiered system is that most donor-funded varsities have found it difficult to maintain a level of competitiveness as high as their ability would otherwise permit. Their competitive disadvantage in comparison to university-funded teams is due, in part, to the reluctance of some schools to include donor-funded teams in their varsity schedules and in part to the inability of the teams to obtain varsity-level coaching and recruits, or to obtain funds *190 for travel, post-season competition and equipment.
Extensive testimony demonstrated that several donor-funded teams do have the interest and ability to compete at the top varsity level and would benefit from university-funded status. I find that the donor-funded women's gymnastics, women's fencing, and women's ski teams have demonstrated this interest and ability. The women's gymnastics team was a thriving university-funded varsity team prior to the 1991 demotion; in fact, the team won the Ivy League championship in 1989-90. See, e.g., testimony of Athletic Director Roach, Prelim.Inj. Hr'g Tr. 11/4/92 at 86 (acknowledging that at the time gymnastics was demoted, the team had the interest and ability to compete at the varsity level). The women's fencing team has been successful for many years against a number of club and varsity competitors, and, prior to the demotions, John Parry, who was the athletic director at that time, supported the team's request to be upgraded to varsity status. Prelim.Inj. Hr'g Tr. 10/27/92 at 122-26. Despite a meager budget, the women's ski team has consistently fielded a competitive team in the U.S. Collegiate Ski Association Osborne Division. See Trial Tr. 9/28/94 at 12-14 (head coach of Smith College women's varsity ski team testifying to strength and stability of the Brown University women's ski team).
Additionally, women's water polo, a club team, has demonstrated the interest and ability to compete at the highest varsity level. The women's water polo team, according to the coach of men's and women's water polo, is able to compete at a varsity level. See Trial Tr. 10/3/94 at 65 (testifying to varsity ability and schedule of the women's water polo team). See also Trial Tr. 10/5/94 at 18-20 (head coach of Slippery Rock University men's and women's varsity water polo teams testifying to stability and achievements of Brown University women's water polo team).
*191 I also find that although all four teams would benefit from university-funded varsity status, only two of these teams, volleyball and skiing, would be able to sustain a competitive intercollegiate varsity schedule if supported at donor-funded status. The testimony demonstrated that gymnastics would effectively cease to exist as an intercollegiate varsity team if it were denied university funding. The gymnastics team does not attract enough private donations to afford quality coaching. See supra note 9. In addition, a donor-funded gymnastics team will draw less talented individuals to the team, which will, as a result, suffer competitively. The testimony also established that elevation of women's water polo, currently a club team, to donor-funded status would actually be financially disadvantageous for the team. As a club sport, the team currently receives $3000 from the Student Activities organization; as a donor-funded sport, the team would be required to raise all of its own funds. Trial Tr. 10/3/94 at 26.
In Cohen, I summarized the history of athletics at Brown University:
Nearly all of the men's varsity teams were established before 1927. Baseball was created first in 1869, followed by football in 1878 and track in 1879. The only men's teams established after 1927 were crew in 1961, water polo in 1974 [elevated to varsity in 1981], and squash in 1989 [provided with varsity services but fully endowed through private donations]. By comparison, virtually all of the women's varsity teams were created between 1971 and 1977. The only women's varsity team created after this period was winter track in 1982. Before 1971, all women's sports were operated out of a separate athletic program at Pembroke College, a sub-unit of Brown University until its merger with Brown College during that year. Before the merger, the women's athletic program at Pembroke bore no resemblance to the program which Brown provided to its male varsity athletes. While Pembroke did have few intercollegiate teams (e.g., field hockey, basketball, tennis), the women's program received very little financial or institutional support from the university.
Cohen, 809 F. Supp. at 981.
At the present time, Brown's intercollegiate athletic program consists of 32 teams. The list of varsity teams and their respective participants for the last complete season (1993/94), with program offering updates, are as follows:
*192 WOMEN MEN University-funded University-funded Basketball: 12 Baseball: 30 Crew: 50 Basketball: 17 Cross-Country: 18 Crew: 45 Field Hockey: 36 Cross-Country: 20 Ice Hockey: 22 Football: 126 Lacrosse: 34 Ice Hockey: 41 Soccer: 22 Lacrosse: 44 Softball: 15 Soccer: 32 Squash: 15 Swimming: 24 Swimming: 27 Tennis: 12 Tennis: 10 Track: 56 Track: 39 Wrestling: 32 Volleyball: 09 Donor-funded Donor-funded: Fencing: 10 Fencing: 26 Gymnastics: 10 Golf: 14 Skiing: 09 Squash: 15 Water polo: 21 Miscellaneous Men's Golf: 01 Men's Crew: 03 _____________________________ _______________________ Total women Total men varsity: 342 varsity: 555
I find that 342 women (38.13% of athletes) and 555 men (61.87% of athletes) were members of varsity teams for the majority of the last complete season and therefore count as "participants" in intercollegiate athletics for the purpose of Title IX analysis.
Defendants argue that there is no consistent measure of actual participation rates because team size varies throughout the athletic season. Defendants observe that injuries, "cuts," and "quits" result in different team numbers depending on when in the season team rosters are tallied. Counting the number of participants at the end of the last complete season, rather than at some arbitrary point in the season, addresses this concern. A team member who participated for the majority of the season should be acknowledged as a participant. Once the season is over, it is possible to assess which of the individuals listed on the roster at the beginning of the season remained team members.
Defendants further argue that there is no consistent measure of actual participation rates because there are alternative definitions of "participant" that yield very different participation totals. For instance, a "participant" could be defined to include every member of the team, even habitual "bench-warmers," or to include only "core" players. Including as "participants" all of the students who were members of the team for a majority of the season addresses this concern. Where both the athlete and coach determine that there is a place on the team for a student, it is not for this Court to second-guess their judgment and impose its own, or anyone else's, definition of a valuable or genuine varsity experience. It is the nature of a team that each student makes a different contribution to the team's success and takes from it a unique experience. Every varsity team member is therefore a varsity "participant."
Thus, in 1993/94, there were 897 student athletes participating in varsity athletics. Of this total, 555 (61.87%) were men and 342 (38.13%) were women. During that same year, the undergraduate enrollment was 5722 students; this included 2796 men (48.86%) and 2926 women (51.14%). Pls.' Ex. 33.
III. PRELIMINARY MATTERS
A. Partial Settlement of the Case
On December 16, 1994, during the course of the trial on the merits, this Court entered *193 a Settlement Agreement and Stipulation of Dismissal in Regard to Equality of Treatment ("Settlement Agreement"). The agreement settled plaintiffs' allegations that significant disparities exist in the relative financial support of and benefits given to men's and women's university-funded varsity teams. The treatment issues settled by this agreement concern only university-funded varsity teams as to which there is no dispute regarding status and does not preclude consideration by the Court of any evidence relevant to issues in contention. Thus, the instant Opinion and Order focuses primarily on plaintiffs' alternative claim that significant disparities exist in the number of intercollegiate participation opportunities available to men and those available to women. Therefore, in assessing Brown's compliance with Title IX, I will address whether the University accommodates effectively "the interests and abilities of students to the extent necessary to provide equal opportunity in the selection of sports and levels of competition available to members of both sexes." Policy Interpretation, 44 Fed.Reg. 71,413, at 71,417 (1979). In addition, this Opinion and Order deals briefly with the treatment of teams as to which there is a dispute concerning status.
B. The Precedential Effect of the First Circuit's Preliminary Injunction Ruling
Brown has advanced the argument that any findings of fact and conclusions of law made by the First Circuit during the preliminary injunction phase of this case are not binding on this Court's ruling on the merits. They note that "[a] preliminary injunction is `by its very nature, interlocutory, tentative, provisional, ad interim, impermanent, mutable, not fixed or final or conclusive, characterized by its for-the-time-beingness.'" Defs.' Mem. of Law Re Law of the Case at 2 (quoting Hamilton Watch Co. v. Benrus Watch Co., 206 F.2d 738, 742 (2d Cir.1953)). Further, they reason that a court confronted with different facts, arguments, and issues may reach different conclusions about legal standards or statutory and regulatory interpretations. Defs.' Reply Mem. Re Law of the Case at 3. During the trial on the merits, defendants urged this Court to find that it had the authority to disregard the interpretation of Title IX adopted by the First Circuit in upholding the preliminary injunction issued by this Court.
Defendants are correct that the factual determinations set forth in Cohen, 809 F. Supp. 978, and upheld by the First Circuit on appeal are not binding on the decision currently before me. Similarly, this Court is not bound by the First Circuit's application of the law to the facts then in evidence. It is well-settled that at the preliminary injunction stage, an appellate court's "`findings' and `holdings' as to the merits of the case are not final but should be understood to be merely statements of probable outcomes based on the record as it existed before the district court." LeBeau v. Spirito, 703 F.2d 639, 643 (1st Cir.1983) (citing Planned Parenthood League of Mass. v. Bellotti, 641 F.2d 1006, 1009 (1st Cir.1981)).
Defendants incorrectly extend this general principle to support their position that this Court is not bound to follow the First Circuit's legal pronouncements. A district court is not free to determine anew issues of law already presented to and decided by the appellate court. Statements of law made by an appellate court, whether set forth in an opinion reviewing a preliminary injunction or in an opinion reviewing the outcome of a trial on the merits, become the law of the circuit and, as established precedent, bind the district courts.
*194 In any case, the question is academic because this Court is in accord with the First Circuit's interpretation of Title IX as set forth in Cohen, 991 F.2d 888. Nothing in the record before me, now fully developed, undermines the considered legal framework established by the First Circuit at the preliminary injunction stage. However, the existence of new facts and legal arguments presented at trial does require a full examination of and expansion upon the foundations laid by the First Circuit.
Naturally, as defendants contend, to the extent that the appellate court has not been presented with and has not decided any legal issue now material to the case, this Court must interpret the law without the benefit of guidance from the appellate court.
IV. LEGAL FRAMEWORK
A. Title IX
The text of Title IX does not explicitly prohibit gender discrimination in athletics. Rather, Title IX is a general prohibition of sex discrimination in all aspects of educational institutions receiving federal funding:
No person in the United States shall, on the basis of sex, be excluded from participation in, be denied the benefits of, or be subjected to discrimination under any education program or activity receiving Federal financial assistance [with exemptions not applicable to this case].
20 U.S.C. ง 1681(a). In order for this prohibition to apply to an institution's athletic programs, the athletic department need not be the direct recipient of federal funding; rather it is well-settled that Title IX applies to all of an institution's programs if any part of an educational institution receives federal funds. Cohen, 809 F. Supp. at 982-83 (citations omitted).
In addition, Title IX specifies that its proscription of gender discrimination in athletics should not be "interpreted to require any educational institution to grant preferential or disparate treatment to the members of one sex on account of [a gender] imbalance" between the persons participating in the program and the total number of persons in the relevant community. 20 U.S.C. ง 1681(b). However, subsection (b) also provides that it "shall not be construed to prevent the consideration in any ... proceeding ... of statistical evidence tending to show that such an imbalance exists with respect to the participation in, or receipt of the benefits of, any such program or activity by the members of one sex." 20 U.S.C. ง 1681(b). Thus, Title IX "does not mandate strict numerical equality between the gender balance of a college's athletic program and the gender balance of its student body." Cohen, 991 F.2d at 894. The practical effect of subsection (b) is that:
[A] court assessing Title IX compliance may not find a violation solely because there is a disparity between the gender composition of an educational institution's student constituency, on the one hand, and its athletic programs, on the other hand.
That is not to say, however, that evidence of such disparity is irrelevant. Quite the contrary: under the proviso contained in section 1681(b), a Title IX plaintiff in an athletic discrimination suit must accompany statistical evidence of disparate impact with some further evidence of discrimination, such as unmet need amongst the members of the disadvantaged gender.
Cohen, 991 F.2d at 895. Thus, evidence of gender imbalance in Brown's athletic programs is relevant to a determination of non-compliance but not sufficient by itself to mandate a finding of Title IX violation.
The U.S. Department of Education ("DED") acting through its Office of Civil Rights ("OCR") is responsible for administering Title IX. Pursuant to the statutory *195 mandate, DED promulgated regulations controlling "Nondiscrimination on the Basis of Sex in Education Programs and Activities Receiving or Benefitting from Federal Financial Assistance." 34 C.F.R. ง 106 (1994). The regulations specifically address athletic program administration at 34 C.F.R. งง 106.37(c) and 106.41.
Paralleling the language of Title IX, section 106.41(a) states:
No person shall, on the basis of sex, be excluded from participation in, be denied the benefits of, be treated differently from another person or otherwise be discriminated against in any interscholastic, intercollegiate, club or intramural athletics offered by a recipient, and no recipient shall provide any such athletics separately on such basis.
34 C.F.R. ง 106.41(a). The regulations go on to clarify, however, that institutions are permitted to "operate or sponsor separate teams for members of each sex" under certain circumstances detailed therein. 34 C.F.R. ง 106.41(b).
Section 106.41(c), entitled "Equal Opportunity," addresses the most critical issue in this case. Regardless of what teams, if any, an institution offers, its athletic program must afford equal opportunities to male and female athletes. The section reads, in pertinent part:
In determining whether equal opportunities are available the Director will consider, among other factors:
(1) Whether the selection of sports and levels of competition effectively accommodate the interests and abilities of members of both sexes;
(2) The provision of equipment and supplies;
(3) Scheduling of games and practice time;
(4) Travel and per diem allowance;
(5) Opportunity to receive coaching and academic tutoring;
(6) Assignment and compensation of coaches and tutors;
(7) Provision of locker rooms, practice and competitive facilities;
(8) Provision of medical and training facilities and services;
(9) Provision of housing and dining facilities and services;
34 C.F.R. ง 106.41(c). The trial on the merits focused on the first of these ten factors, the "effective accommodation" element. The other nine factors concern "treatment issues," which were settled by the parties only as to equivalence between men's and women's teams voluntarily maintained by Brown at the university-funded level. See discussion supra part III.A.
C. Policy Interpretation
Several years after the regulations were promulgated, DED's Office of Civil Rights proposed a policy interpretation designed to resolve confusion concerning Title IX compliance. After considering public comments, it published the final Policy Interpretation in the Federal Register. 44 Fed.Reg. 71,413 (1979). The Policy Interpretation:
clarifies the obligations which recipients of Federal aid have under Title IX to provide equal opportunities in athletic programs. In particular, this Policy Interpretation provides a means to assess an institution's *196 compliance with the equal opportunity requirements of the regulation which are set forth at [34 C.F.R. งง 106.37(c) and 106.41(c)].
44 Fed.Reg. at 71,415. The Policy Interpretation is divided into three categories: "Athletic Financial Assistance (Scholarships)," 44 Fed.Reg. at 71,415; "Equivalence in Other Athletic Benefits and Opportunities," 44 Fed. Reg. at 71,415; and "Effective Accommodation of Student Interests and Abilities." 44 Fed.Reg. at 71,417. Litigation focused on the "Effective Accommodation" section.
The "Effective Accommodation" section of the Policy Interpretation interprets the first of the ten factors found in the regulations at 34 C.F.R. ง 106.41(c) (1) (stating that the Director should consider "[w]hether the selection of sports and levels of competition effectively accommodate the interests and abilities of members of both sexes"). The Policy Interpretation measures compliance in the area of effective accommodation first by applying a three prong test and second by applying a two part test. The three prong test assesses compliance by determining:
(1) Whether intercollegiate level participation opportunities for male and female students are provided in numbers substantially proportionate to their respective enrollments; or
(2) Where the members of one sex have been and are underrepresented among intercollegiate athletes, whether the institution can show a history and continuing practice of program expansion which is demonstrably responsive to the developing interest and abilities of the members of that sex; or
(3) Where the members of one sex are underrepresented among intercollegiate athletes, and the institution cannot show a continuing practice of program expansion such as that cited above, whether it can be demonstrated that the interests and abilities of the members of that sex have been fully and effectively accommodated by the present program.
44 Fed.Reg. at 71,418. The two part test assesses compliance by examining:
(1) Whether the competitive schedules for men's and women's teams, on a program-wide basis, afford proportionally similar numbers of male and female athletes equivalently advanced competitive opportunities; or
(2) Whether the institution can demonstrate a history and continuing practice of upgrading the competitive opportunities available to the historically disadvantaged sex as warranted by developing abilities among the athletes of that sex.
44 Fed.Reg. at 71,418.
The "Effective Accommodation" section of the Policy Interpretation, which includes both the three prong test and the two part test, also provides that a school need not establish an intercollegiate team where there is no reasonable expectation of competition in that sport within the institution's normal geographic area of competition. 44 Fed.Reg. at 71,418.
Finally, the Policy Interpretation's "Effective Accommodation" section provides that the following factors should be considered in an overall determination of compliance pursuant to 34 C.F.R. ง 106.41(c):
(a) Whether the policies of an institution are discriminatory in language or effect; or
(b) Whether disparities of a substantial and unjustified nature in the benefits, treatment, services, or opportunities afforded male and female athletes exist in the institution's program as a whole; or
(c) Whether disparities in individual segments of the program with respect to benefits, treatment, services, or opportunities are substantial enough in and of themselves *197 to deny equality of athletic opportunity.
44 Fed.Reg. 71,418.
D. Title IX Athletics Investigator's Manual
The 1990 Title IX Athletics Investigator's Manual, published by OCR, was designed to assist OCR investigators in determining whether an institution is in compliance with Title IX. The Manual serves as a practical guide in the agency's enforcement of the Policy Interpretation. Valerie M. Bonnette & Lamar Daniel, Department of Education, Title IX Athletics Investigator's Manual (1990).
V. DEFERENCE TO AGENCY REGULATIONS AND INTERPRETATION OF TITLE IX
In Cohen, the First Circuit held that the regulations promulgated pursuant to Title IX deserve controlling weight. 991 F.2d at 895. This degree of deference is appropriate because Congress "explicitly delegated to the agency the task of prescribing standards for athletic programs under Title IX." Cohen, 991 F.2d at 895 (citations omitted). See Javits Amendment, Pub.L. No. 93-380, ง 844, 88 Stat. 612 (1974). The appellate court also concluded that the Policy Interpretation, because it is a "considered interpretation" of the agency's own regulation, warrants "substantial deference." Cohen, 991 F.2d at 896-7.
As I have discussed above, supra part III.B, this Court is bound by the law of the circuit. As a result, this Opinion and Order must accord controlling weight to the regulations and substantial deference to the Policy Interpretation. However, I revisit the issue of deference in order to address defendants' position that the First Circuit did not have the benefit of newly developed arguments and information or that "no court has carefully examined the issue of the degree of deference, if any, to which the OCR's Policy Interpretation and Investigator's Manual are entitled." Defs.' Post-Trial Mem. at 25.
It is well-settled that where Congress has expressly authorized an agency to issue regulations, those agency regulations are binding on the courts "unless they are arbitrary, capricious, or manifestly contrary to the statute." Chevron, U.S.A. v. Natural Resources Defense Council, 467 U.S. 837, 844, 104 S. Ct. 2778, 2782, 81 L. Ed. 2d 694 (1984). See also National Latino Media Coalition v. F.C.C., 816 F.2d 785, 788 (D.C.Cir. 1987) (stating that "[w]hen Congress delegates rulemaking authority to an agency, and the agency adopts legislative rules, the agency stands in the place of Congress and makes law").
Here, Congress explicitly delegated rulemaking authority to DED, see supra note 23, stating:
The Secretary shall prepare and publish, not later than 30 days after the date of enactment of this Act, proposed regulations implementing the provisions of title IX of the Education Amendments of 1972 relating to the prohibition of sex discrimination in federally assisted education programs which shall include with respect to intercollegiate athletic activities reasonable provisions considering the nature of particular sports.
Javits Amendment, Pub.L. No. 93-380, ง 844, 88 Stat. 612 (1974). This delegation of authority specifically directed DED to include within its intercollegiate athletic regulations "reasonable provisions considering the nature of particular sports." Thus, Congress included within its conferral of rulemaking authority a specific directive with respect to the intercollegiate athletic regulations; this directive mandates the inclusion of one particular subject but does not otherwise limit the broad scope of authority delegated to the agency.
The agency's regulations were duly approved by President Ford in 1975. This approval was required by Title IX itself, which provides that no "rule, regulation, or order" an agency is empowered to enact pursuant to Title IX shall become effective *198 unless and until approved by the President. 20 U.S.C. ง 1682. The regulations, which were designed to address extensive evidence of sex discrimination in intercollegiate athletics, clearly meet the Chevron standard; they are not arbitrary, capricious or manifestly contrary to Title IX. Kelley, 35 F.3d at 270.
An agency's interpretation of its own regulations, while not given the force of law, is entitled to substantial deference. Lyng v. Payne, 476 U.S. 926, 939, 106 S. Ct. 2333, 2341, 90 L. Ed. 2d 921 (1986) (citations omitted). See also Udall v. Tallman, 380 U.S. 1, 16, 85 S. Ct. 792, 801, 13 L. Ed. 2d 616 (1965) (noting that "[w]hen the construction of an administrative regulation rather than a statute is in issue, deference is even more clearly in order"). The Supreme Court explained its reasoning for this conclusion in Martin v. Occupational Safety and Health Review Comm'n, 499 U.S. 144, 151, 111 S. Ct. 1171, 1176, 113 L. Ed. 2d 117 (1991) (citing Ford Motor Credit Co. v. Milhollin, 444 U.S. 555, 556, 558, 100 S. Ct. 790, 792, 793, 63 L. Ed. 2d 22 (1980)):
Because applying an agency's regulation to complex or changing circumstances calls upon the agency's unique expertise and policymaking prerogatives, we presume that the power authoritatively to interpret its own regulations is a component of the agency's delegated lawmaking powers.
Thus, for all practical purposes, an agency's interpretation of its own regulations is accorded the force of law unless such interpretation is clearly erroneous or inconsistent with the regulation. United States v. Larionoff, 431 U.S. 864, 872, 97 S. Ct. 2150, 2155, 53 L. Ed. 2d 48 (1977) (quoting Bowles v. Seminole Rock Co., 325 U.S. 410, 414, 65 S. Ct. 1215, 1217, 89 L. Ed. 1700 (1945)). See also McCuin v. Secretary of Health and Human Servs., 817 F.2d 161, 168 (1st Cir.1987) (the interpretation "must be reasonable in view of the language of the regulations and the policies they were meant to implement").
As the First Circuit has held, the Policy Interpretation should be given substantial weight. Cohen, 991 F.2d at 896-97. The Policy Interpretation is a "considered" document: the original draft of the Policy Interpretation was published for public comment, the agency received over 700 comments, and agency staff visited eight universities to evaluate how the Policy Interpretation and other alternatives would be applied in practice. 44 Fed.Reg. at 71,413. Cf. Udall, 380 U.S. at 17, 85 S. Ct. at 801 (noting with approval that the agency's interpretation had been a matter of public record and discussion). In addition, since publication of the Policy Interpretation, Congress has had the opportunity to disapprove of the Policy Interpretation, see Civil Rights Restoration Act of 1987, 20 U.S.C. ง 1687 (1988); however, Congress instead chose to reaffirm its intent that Title IX's prohibition against discrimination be broadly construed. See Cohen, 991 F.2d at 894 (recognizing that record of the floor debate on Restoration Act "leaves little doubt that the enactment was aimed, in part, at creating a more level playing field for female athletes") (citations omitted). Cf. North Haven Bd. of Ed. v. Bell, 456 U.S. 512, 535, 102 S. Ct. 1912, 1925, 72 L. Ed. 2d 299 (1982) (where "an agency's statutory construction has been `fully brought to the attention of the public and the Congress,' and the latter has not sought to alter that interpretation although it has amended the statute in other respects, then presumably the legislative intent has been correctly discerned") (citations omitted).
Defendants dispute the First Circuit and this Court's determination of the degree of judicial deference appropriately accorded to the agency's regulations and interpretations. They argue that the agency exceeded its delegated rulemaking authority, and that the documents are therefore "interpretive" rather than "legislative." Defendants argue that the rulemaking authority Congress delegated to the agency was limited by the Javits Amendment to provisions concerning the nature of particular sports. Thus, defendants conclude that to the extent the regulations went beyond that specific directive they are "interpretive" rather than "legislative" and are thus entitled to the less deferential standard of review articulated in the germinal cases Skidmore v. Swift & Co., 323 U.S. 134, 65 S. Ct. 161, 89 L. Ed. 124 (1944) and General Electric Co. v. Gilbert, 429 U.S. 125, 97 S. Ct. 401, 50 L. Ed. 2d 343 (1976). Defendants' reliance *199 on Skidmore and Gilbert is misplaced because these cases concern agency guidelines issued in the absence of an explicit Congressional delegation of authority. Here, however, Congress did authorize the agency to promulgate regulations implementing Title IX. This Court is persuaded that "the directive that those regulations `include reasonable provisions considering the nature of particular sports' does not limit the scope of the delegation; it merely compels the agency to include such provisions in its broader regulatory framework." Proposed Findings of Fact and Conclusions of Law of Pls. at 28. As I noted above, these regulations must therefore be given controlling weight. Similarly, the Policy Interpretation was issued as a result of the duly delegated rulemaking authority entrusted to the agency, and is therefore entitled to substantial deference. Cf. Martin, 499 U.S. at 151, 111 S. Ct. at 1176 (where the Court acknowledged that an agency's power authoritatively to interpret its own regulations is a component of the agency's delegated lawmaking powers).
Defendants further contend that the Policy Interpretation, because it was never approved by the President, does not have the binding effect of "rules, regulations, or orders" authorized by 20 U.S.C. ง 1682. The Policy Interpretation, however, is not a rule, regulation, or order, but is a guideline designed to interpret a rule, regulation, or order, namely, the agency's own regulations published at 34 C.F.R. ง 106. The Policy Interpretation therefore need not be approved by the President in order to become effective.
In sum, this Court must abide by the Policy Interpretation unless it is clearly erroneous or inconsistent with the regulations. See Larionoff, 431 U.S. at 872, 97 S. Ct. at 2155. Defendants protest that the Policy Interpretation contravenes the intent of Title IX in at least two respects. First, they claim that the Policy Interpretation, as interpreted by this Court, renders Title IX an affirmative action statute, in derogation of Title IX, which specifically provides that it shall not be interpreted to require preferential or disparate treatment to members of one sex. 20 U.S.C. ง 1681(b). This claim, however, is premised on a misunderstanding of the Policy Interpretation. The Policy Interpretation's three prong test does not mandate statistical balancing. In fact, the test is designed to avoid an absolute requirement of numerical equality. Where substantial proportionality has not been achieved (prong one), an institution must be found in compliance if it demonstrates that it has a continuing practice of expanding the athletic opportunities of the underrepresented sex (prong two) or if its existing program fully and effectively accommodates the interests and abilities of the underrepresented sex (prong three). Kelley, 35 F.3d at 271. Thus, the test encourages equality but recognizes that some institutions may be unable to attain this goal through no fault of their own; in these cases, the test provides alternatives to statistical parity.
Second, defendants claim that OCR ignored the Javits Amendment's instruction to consider the nature of particular sports in drafting reasonable provisions with respect to intercollegiate athletics. "[D]ue to the nature of particular sports, there are upwards [of] 150 to 175 ... participation opportunities for men (on football, wrestling, lacrosse and the like) that cannot be available to women." Defs.' Post-Trial Mem. at 27. On the contrary, the Policy Interpretation does consider the nature of particular sports. For example, it permits the operation of separate teams for members of each sex and does not require a school to sponsor a women's team for every men's team offered and vice versa. In addition, it recognizes that different expenditures on men's and women's sports may be permissible if based on factors inherent to the operation of specific sports. The fact that the Policy Interpretation does not consider and accommodate the nature of different sports in the precise manner advocated by defendants does not render it unreasonable. See Cohen, 991 F.2d at 899.
VI. LEGAL ANALYSIS
A. The Three Prong Test
The plaintiffs contend that the Brown athletic program is in violation of the "Equal Opportunity" provision of the Title IX athletics regulations. Therefore, I focus my analysis *200 on the Policy Interpretation to "clarif[y] the obligations which [Brown has] to provide equal opportunities in [its] athletic programs." 44 Fed.Reg. at 71,415. I must assess compliance by looking to the regulations applicable in this case. These regulations, as noted above, are: 34 C.F.R. ง 106.37(c) (distribution of athletic scholarships โ" not applicable to this case), 34 C.F.R. ง 106.41(a) and (b) (general non-discrimination pronouncement but allowing for separate teams โ" not at issue in this case), and 34 C.F.R. ง 106.41(c) (requiring institutions to provide equal opportunity to both sexes in athletics programs offered, and listing ten factors for consideration in assessing compliance). The subject of this lawsuit is whether or not Brown complies with the "Equal Opportunity" provision found at 34 C.F.R. ง 106.41(c). Litigation has primarily focused on the first of the ten "equal opportunity" factors which asks "whether the selection of sports and levels of competition effectively accommodate the interest and abilities of members of both sexes." 34 C.F.R. ง 106.41(c) (1). Determining compliance with this factor requires application of the Policy Interpretation's three prong test, recited supra part IV.C, which delineates the standards for enforcement of 34 C.F.R. ง 106.41(c) (1). Because the proper interpretation of the three prong test is the most hotly contested legal issue in this case, I will extensively analyze and address the relevant arguments presented by the parties.
1. Prong One
An institution complies with the three prong test if it meets prong one of the analysis and no other. Plaintiffs bear the burden of proving that the institution does not satisfy prong one. Prong one asks:
Whether intercollegiate level participation opportunities for male and female students are provided in numbers substantially proportionate to their respective enrollments ...
44 Fed.Reg. at 71,418. I conclude that an institution satisfies prong one provided that the gender balance of its intercollegiate athletic program substantially mirrors the gender balance of its student enrollment. "Thus, a university which does not wish to engage in extensive compliance analysis may stay on the sunny side of Title IX simply by maintaining gender parity between its student body and its athletic lineup." Cohen, 991 F.2d at 897-98. I now discuss the meaning of three contested elements of prong one.
a. Intercollegiate Level Athletics
The Policy Interpretation expressly states that it is intended to apply to intercollegiate athletics. It explains, however, that because the regulations distinguish between club sports and intercollegiate sports, "under this Policy Interpretation, club teams will not be considered to be intercollegiate teams except in those instances where they regularly participate in varsity competition." 44 Fed. Reg. at 71,413 n. 1. It was not seriously contended until the eleventh hour, nor did the evidence show, that any of Brown's club teams should be considered to be presently operating as intercollegiate teams under this definition. Therefore, I do not treat any of Brown's club programs as intercollegiate teams under prong one. On the other hand, *201 Brown's donor-funded varsity teams resemble university-funded varsity teams in some aspects and club teams in others. See supra notes 6-13 and accompanying text. Thus, the question arises whether donor-funded teams may be considered to be "intercollegiate" teams for the purposes of prong one.
It is evident that Brown's donor-funded teams operate at a disadvantage in comparison to its university-funded varsity teams. The issue with regard to funding sources is not where the funding comes from but the extent to which funding comes at all to donor-funded teams. Donor-funded teams generally have much less money, have no commitment from the university that the team will be supported in a year when fund raising efforts are less successful, and must expend gift funds for privileges that Brown provides to university-funded teams at no cost to those teams. As a result of their unfunded status, most of the donor-funded teams are prevented from reaching their full athletic potential. However, the evidence presented at trial suggests that, on the whole, Brown's donor-funded teams do engage in "varsity level" competition. Thus, I treat both donor-funded and university-funded varsity teams as intercollegiate teams but as positioned at distinct levels within the athletic hierarchy.
b. Substantially Proportionate
The phrase "substantially proportionate" is necessarily an elusive concept. Consequently, I turn toward other factors to inform my interpretation of this term. Because a positive *202 showing on prong one terminates the inquiry, providing a "safe harbor" for the institution, Cohen, 991 F.2d at 897, logic suggests that "substantially proportionate" must be a standard stringent enough to effectuate the purposes of the statute. At the same time, "substantially proportionate" accounts for the possibility of minor fluctuations in the undergraduate population and in the athletic program from one year to the next. Thus, substantial proportionality is properly found only where the institution's intercollegiate athletic program mirrors the student enrollment as closely as possible. This definition takes into account any small variations that are beyond the institution's ability to control or predict.
Defendants argue that "substantially proportionate" must be interpreted very liberally, in favor of the institution, lest a sudden surge in numbers on one team propel a complying institution into violation. Defendants emphasize that the gender composition of the athletic program is both unpredictable and out of Brown's control, asserting that "Brown is stuck with whomever shows up on campus." Trial Tr. 12/16/94 at 61. This position ignores several significant facts. First, prong one compliance is assessed by comparing the gender ratio of the student enrollment with the gender ratio of the entire intercollegiate athletic program. It is unlikely that numerical fluctuations on an individual team will significantly alter the gender ratio of any sizable athletic program. In fact, testimony in the trial revealed that the fluctuations, from year to year, of the gender balance in the athletic program at Brown were minimal. Second, this position fails to acknowledge that when significant numerical changes did occur in the intercollegiate athletic program as a whole, these changes were within the control of the University. Plaintiff's expert, Dr. Christine Grant, Athletic Director for the women's program at the University of Iowa and a former official of a number of collegiate athletic associations, testified persuasively that a university "predetermines" the approximate number of athletic participants and the male to female ratio. I conclude that Brown does predetermine the gender balance of its athletic program through the selection of sports it offers (some sports, by their nature, require more players), the size of the teams it maintains (as dictated by each coach's preference), the quality and number of coaches it hires, and the recruiting and admissions practices it implements. Cf. Prelim.Inj. Hr'g Tr. 10/26/92 at 20. For example, coaches at Brown acknowledged the prominent role recruiting plays in sustaining Brown's varsity athletic program. Most coaches testified that they determine an ideal team size and then recruit the requisite number of athletes to reach that goal. Because recruits constitute the great majority of athletes on nearly all of Brown's university-funded varsity teams, the University should not have been surprised by the gender mix of interested athletes on campus.
c. Participation Opportunities
The First Circuit, in sustaining the preliminary injunction, did not define the term "participation opportunities" as set forth in prong one of the Policy Interpretation. It now rests upon me to formulate a definition. For the purposes of the three prong test, I hold that the "participation opportunities" offered by an institution are measured by counting the actual participants on intercollegiate teams. The number of participants in Brown's varsity athletic program accurately reflects the number of participation opportunities Brown offers because the University, through its practices "predetermines" the number of athletic positions *203 available to each gender. In addition, as noted below, any other measure of participation opportunities is infeasible. While the First Circuit did not explicitly define "participation opportunities" as participation rates, it implicitly adopted and applied this definition in its prong one analysis.
Defendants contend, however, that the concept "participation opportunity" carries a very different practical meaning from the simpler concept of "participation." They argue that the intercollegiate athletic participation opportunities offered at Brown should be measured by counting each team's filled and unfilled athletic slots. The defendants offer several methods of determining the exact number of slots that they contend are available but unfilled.
First, Brown proposes that participation opportunities should include those additional athletic slots that women's team coaches testified they were able to support, given current resources. This method is flawed because many coaches, while technically able to support more team members, restrict their team size according to their personal coaching philosophies. Thus, these theoretical opportunities are not actually available to athletic hopefuls. I also note that even if I were to accept this definition of participation opportunities, or any other measure of "unfilled but available" positions, I would be compelled to compare the women's and the men's theoretically available additional spots.
Second, Brown asserts that each team is necessarily capable of carrying, at a minimum, the number of athletic positions it has supported in recent history. They contend that each team therefore affords at least that number of opportunities to participate, even during years in which the actual participation rates fall short of their historical high. I must reject the assertion that peak numbers achieved during some year in the past are the most accurate measure of participation opportunities presently offered by an institution. Numbers from the current or most *204 recent, complete competitive season provide the most representative quantification of participation opportunities presently offered.
Third, in the alternative, Brown contends that each women's team for which there is an "equivalent" men's team is by definition able to carry at least the number of athletic positions carried by its counterpart in the men's program. Brown maintains that each women's team, therefore, affords at least as many participation opportunities as its "matching" men's team. I decline to adopt this methodology. It is not immediately apparent why the size of men's teams should set the standard for women's team sizes. Nor does the evidence presented at trial justify this approach. Men's and women's teams of the same name are sufficiently distinct from one another to invalidate any approach that rests upon an assumption of similarity. I conclude that intercollegiate "`[a]thletic opportunities' means real opportunities, not illusory ones," Horner, 43 F.3d at 274 (quoting Williams v. School Dist. of Bethlehem, 998 F.2d 168, 175 (3d Cir.1993)), and therefore should be measured by counting actual participants.
Finally, defendants offer an independent interpretation of the meaning of "participation opportunities" in the context of prong one. They define a participation opportunity as a chance for an interested person to participate. Defendants contend that where the student body is comprised of equal numbers of men and women, equality means "offering" the chance to participate in athletics to an equal number of men and women. They postulate that if students were offered a hypothetical opportunity to participate, the students would actually participate in varsity athletics in accordance with the relative interest of their respective genders. Thus, where the gender ratio of a university's interested student population is substantially proportionate to the gender ratio of its athletic program, it may be assumed that men and women in the student body were "offered" an equal "opportunity" to participate. Therefore, defendants claim, Brown provides equal "participation opportunities" if the ratio of men to women among varsity athletes is substantially proportionate to the ratio of men to women among students interested in participating in varsity athletics. Defendants *205 conclude that, in order to succeed on prong one, plaintiffs bear the burden of proving that the percentage of women among varsity athletes is not substantially proportionate to the percentage of women among students interested in participating in varsity athletics.
Defendants, through their expert Dr. Finis Welch, a prominent labor economist, borrow from Title VII employment discrimination jurisprudence. There, the relevant comparison is between the qualified applicant pool and the work place demographics, rather than between the population of the United States and the work place demographics. This analogy is the basis for defendants' argument that the relevant comparison in Title IX cases is between the interested potential varsity athlete pool, however defined, and the make-up of Brown's athletic program, rather than between the student enrollment and the athletic program. Comparison to Title VII is inapposite, however. Title VII seeks to determine whether gender-neutral job openings have been filled without regard to gender. Title IX, on the other hand, was designed to address the reality that sports teams, unlike the vast majority of jobs, do have official gender requirements, and this statute accordingly approaches the concept of discrimination differently from Title VII. Cf. Kelley, 35 F.3d at 270 (stating that "Congress itself recognized that addressing discrimination in athletics presented a unique set of problems not raised in areas such as employment and academics") (citations omitted). Title IX establishes a legal presumption that discrimination exists if the university does not provide participation opportunities to men and women in substantial proportionality to their respective student enrollments, unless the university meets one of the two exonerating situations set forth in prongs two and three. See Kelley, 35 F.3d at 271.
Entirely apart from the flaws in the Title VII-Title IX analogy, there are a number of other problems with interpreting prong one to require that a university afford athletic opportunities to men and women in proportion to their relative interests in athletics. Under this Court's interpretation of prong one, concerned parties can assess an institution's compliance simply by comparing the gender ratio of participating athletes with the gender ratio of the student body, both of which are easily ascertained. Under the defendants' theory, a concerned party must undertake a complicated assessment of "interested" students before making any comparisons. Any such assessment will be meaningless since it is an impossible task to quantify latent and changing interests. Thus, in addition to contravening the plain language of prong one, defendants' interpretation imposes a heavy burden on student-plaintiffs, the courts, and institutions who *206 wish to monitor their own compliance. In any case, it is unclear what population should be surveyed to assess the interest of the "qualified applicant pool," even if it were possible to do so.
The possible survey populations range from matriculated students, to all actual Brown applicants, to all academically able potential varsity participants. In an analogy to Title VII and the "qualified applicant pool," defendants argue that the relevant population consists only of those men and women who might be interested and able to participate in varsity athletics. Uninterested and unathletic persons are irrelevant to any assessment of the substantial proportionality between the athletic program and the student enrollment. The question, under defendants' theory, then becomes who belongs to the "qualified applicant pool" from which Brown might draw student-athletes. Each of the possible "pools" from which Brown might draw varsity athletes has inherent theoretical and practical problems as a survey population, which confirms my initial conclusion that defendants' interpretation of prong one is incorrect. I will now discuss each of the possible survey pools.
(i) Matriculated Students
Because Brown, as a Division I school, actively recruits nationwide most of the students who play on its varsity teams, the survey population of potential participants must be broader than the pool of matriculated students. What students are present on campus to participate in a survey of interests has already been predetermined through the recruiting practices of the coaches. What teams are established and can recruit or qualify for admissions preferences has already been predetermined by Brown. Thus, the interest present on campus is controlled by Brown; to then suggest that Brown must only satisfy the relative interests of students present on campus is circular.
(ii) Actual Brown Applicants
Defendants have suggested that an appropriate survey population might consist of all students who applied to Brown. They claim that several already existing questions on Brown's application would provide an easy determination of the relative interests in varsity athletics. There are two problems with using this group of students as the determinative "pool." First, it revisits the conceptual problems inherent in attempting to measure "interest." See supra note 43. It cannot be true that questions on Brown's application provide a reliable measure of interest that will be acted upon given the opportunity. Although many women who later committed to attend Brown expressed some type of interest in sports at the time of application, far fewer actually participate in Brown's intercollegiate athletic program. Second, using the pool of actual Brown applicants fails to consider the fact that college applicants interested in a sport not offered as a varsity sport at Brown may not even apply to Brown. A survey of actual Brown applicants *207 would thus fail to capture the interest of those student-athletes who choose not to apply due to the limits of Brown's program offerings. To suggest that Brown need only satisfy the interests of actual applicants where Brown's selection of program offerings affects who applies to the school in the first place is illogical.
(iii) Academically Able Potential Varsity Participants
If one were to accept the defendants' analogy to the "qualified applicant pool" of Title VII, the most appropriate survey population would consist of all academically able potential varsity participants. The "qualified applicant pool" for Brown, as a Division I institution, would consist of all prospective college applicants who might apply to or be recruited by Brown if Brown offered their preferred varsity sports. Because Brown seeks athletic recruits from across the nation, the difficulties in identifying all such persons in order to construct a representative survey population may be insurmountable.
In addition, even a successful survey of all academically able potential Brown applicants could not accurately measure interest in certain sports, crew, for example, that commonly develop only after matriculation. Nor can a survey of this population account for the extent to which opportunities drive interests. See, e.g., testimony of Dr. Welch, Trial Tr. 11/29/94 at 92 (question asking, "Would you agree with the following statement? If Brown provides far more opportunities for women, then maybe the percentage of interested women will rise?" and witness replying, "Sure, I don't see anything wrong with that").
For all the foregoing reasons, I reject defendants' alternative interpretation of prong one.
Under the Policy Interpretation, an institution may fail to provide participation opportunities substantially proportionate to student enrollment but still comply with the Title IX regulatory scheme. The Policy Interpretation recognizes that a school might make every attempt to accommodate women's interests and abilities and still not achieve substantial proportionality. Therefore, an institution is permitted to maintain an athletic program that provides substantially more opportunities to students of one gender than to the other if it comports with one of the two other prongs. In this way, the Policy Interpretation accommodates the equal opportunity mandate of Title IX without requiring strict quotas.
2. Prong 2
In the event that an institution fails prong one, it can still prevail by demonstrating compliance with prong two. Defendants bear the burden of proof on prong two, which asks:
Where the members of one sex have been and are underrepresented among intercollegiate athletes, whether the institution can show a history and continuing practice of program expansion which is demonstrably responsive to the developing interest and abilities of the members of that sex ...
44 Fed.Reg. at 71,418. Prong two illustrates that "Title IX does not require that the university leap to complete gender parity in a single bound." Cohen, 991 F.2d at 898. However, it does require an institution to demonstrate that it has continued to increase the number of athletes participating in intercollegiate athletics. An institution does not demonstrate "program expansion" by reducing men's teams so as to increase the relative percentage of female participation in intercollegiate athletics, although it may achieve compliance with prong one if it sufficiently reduces the program of the overrepresented gender. See Roberts, 998 F.2d at 830 (stating that "the ordinary meaning of the word `expansion' may not be twisted to find compliance under [prong two] when schools have increased the relative percentages of women participating in athletics by making cuts in both men's and women's sports programs").
3. Prong 3
Even where plaintiffs can prove that an institution does not offer intercollegiate athletic participation opportunities to men and women in substantial proportionality to their respective enrollments (prong one), and where defendants are unable to demonstrate *208 a history and continuing practice of program expansion (prong two), plaintiffs must additionally prove that an institution does not satisfy prong three. Prong three asks:
Where the members of one sex are underrepresented among intercollegiate athletes, and the institution cannot show a continuing practice of program expansion such as that cited above, whether it can be demonstrated that the interests and abilities of the members of that sex have been fully and effectively accommodated by the present program.
44 Fed.Reg. at 71,418. Prong three "requires a relatively simple assessment of whether there is unmet need in the under-represented gender that rises to a level sufficient to warrant a new team or the upgrading of an existing team." Cohen, 991 F.2d at 900. Thus, provided that athletes of the underrepresented gender have both the ability and interest to compete at the intercollegiate level, they must be fully and effectively accommodated. Of course, institutions do retain discretion when met with requests for the creation of a new team or the upgrading of an existing team. They need not upgrade or create a team where the interest and ability of the students are not sufficiently developed to field a varsity team. Further:
[i]nstitutions are not required to upgrade teams to intercollegiate status or otherwise develop intercollegiate sports absent a reasonable expectation that intercollegiate competition in that sport will be available within the institution's normal competitive regions.
44 Fed.Reg. at 71,418. Thus, defendants' concern that the Court's interpretation of prong three would require Brown to "accommodate each and every expressed student interest as long as there [is] any evidence of ability" is not well taken. Defs.' Post-Trial Mem. at 1.
a. Full and Effective Accommodation
Defendants press their interpretation of prong three, originally presented at the preliminary injunction stage of this litigation. Their position is that "to the extent students' interests in athletics are disproportionate by gender, colleges should be allowed to meet those interests incompletely as long as the school's response is in direct proportion to the comparative levels of interest." Cohen, 991 F.2d at 899. In other words, Brown argues that it may accommodate less than all of the interested and able women if, on a proportionate basis, it accommodates less than all of the interested and able men. This is simply not the law. This Court and the First Circuit previously rejected defendants' interpretation of prong three for a number of reasons that still hold true today. Just as with prong one, defendants' theory is inconsistent with the law, is poor policy, and presents a logistical quagmire.
First, as the First Circuit detailed:
We think that Brown's perception of the Title IX universe is myopic. The fact that the overrepresented gender is less than *209 fully accommodated will not, in and of itself, excuse a shortfall in the provision of opportunities for the underrepresented gender....
In the final analysis, Brown's view is wrong on two scores. It is wrong as a matter of law, for DED's Policy Interpretation, which requires full accommodation of the underrepresented gender, draws its essence from the statute. Whether Brown's concept might be thought more attractive, or whether we, if writing on a pristine page, would craft the regulation in a manner different than the agency, are not very important considerations....
Brown's reading of Title IX is legally flawed for yet another reason. It proceeds from the premise that the agency's third benchmark countervails Title IX. But, this particular imprecation of the third benchmark overlooks the accommodation test's general purpose: to determine whether a student has been "excluded from participation in, [or] denied the benefits of' an athletic program "on the basis of sex ..." 20 U.S.C. ง 1681(a). While any single element of this tripartite test, in isolation, might not achieve the goal set by the statute, the test as a whole is reasonably constructed to implement the statute....
As it happens, Brown's view is also poor policy, for, in the long run, a rule such as Brown advances would likely make it more difficult for colleges to ensure that they have complied with Title IX. Given that the survey of interests and abilities would begin under circumstances where men's athletic teams have a considerable head start, such a rule would almost certainly blunt the exhortation that schools should "take into account the nationally increasing levels of women's interests and abilities" and avoid "disadvantag[ing] members of an underrepresented sex ..." 44 Fed.Reg. at 71,417.
Brown's proposal would also aggravate the quantification problems that are inevitably bound up with Title IX.[] Student plaintiffs, who carry the burden of proof on this issue, as well as universities monitoring self-compliance, would be required to assess the level of interest in both the male and female populations and determine comparatively how completely the university was serving the interests of each sex....
Furthermore, by moving away from OCR's third benchmark, which focuses on the levels of interest and ability extant in the student body, Brown's theory invites thorny questions as to the appropriate survey population, whether from the university, typical feeder schools, or the regional community.[] In that way, Brown's proposal would do little more than overcomplicate an already complex equation.
Cohen, 991 F.2d at 899-900. Thus, as the First Circuit found, Brown's interpretation is logistically difficult to administer and is inconsistent with the effective accommodation mandate of the Title IX implementing regulations in two respects. It both contravenes the plain meaning of prong three and ignores the Policy Interpretation's directive that an institution determine the athletic interests and abilities of students in such a way as to "take into account the nationally increasing levels of women's interests and abilities" and avoid disadvantaging members of an underrepresented sex. 44 Fed.Reg. at 71,417. Brown's interpretation disadvantages women and undermines the remedial purposes of Title IX by limiting required program expansion for the underrepresented sex to the status quo level of relative interests.
Additionally, the trial on the merits illustrated some of the practical and conceptual difficulties identified above by the First Circuit. Although the plaintiffs bear the burden of proof on prong three, the defendants introduced a great deal of evidence in support of their position that Brown satisfies prong three, as they interpret it. Brown conducted a survey on campus, analyzed students' college applications, and assembled a variety of national studies in an attempt to quantify the relative interest of men and women in athletics. Defendants drew from a variety of populations and survey questions, generating scores of measures of what constitutes "interest." Because no one measure and no *210 identifiable population adequately establish relative interest, see supra notes 43-44 and accompanying text, defendants effectively demonstrated how their interpretation of prong three would impose an insurmountable task on Title IX plaintiffs.
Finally, contrary to defendants' arguments, this Court's interpretation of the three prong test does give independent meaning to each component of the test. Defendants posit:
[T]o suggest that prong one and prong three are the same, that they both require student body ratios is to read prong three out of existence. It becomes redundant. And as the U.S. Supreme Court said in Nordic Village a statute must, if possible, be construed in such a fashion that every word has some operative meaning. For a court to say I have to be at student body ratios in prong one, and for a court to say I also have to be at student body ratios in prong three if I have anybody that's interested on campus, has basically thrown out the raison d'etre for prong three and said we might as well only have a one prong test.
Trial Tr. 12/16/94 at 74-75. This Court's interpretation of prong three does require that the unmet interests and abilities of the underrepresented sex be accommodated to the fullest extent until the substantial proportionality of prong one is achieved. This requirement does not, however, render prong three meaningless. At Brown University the number of interested and able women may match or exceed the number of men participating in the intercollegiate athletic program, thus requiring Brown to achieve substantial proportionality. This depth of athletic talent among the underrepresented sex may not exist to the same extent at other universities. Thus, while Brown may be unable to justify its program under prong three, because of its wealth of unaccommodated female athletes, other universities may point to the absence of such athletes to justify an athletic program that does not offer substantial proportionality. In this way, the Policy Interpretation gives effect to 20 U.S.C. ง 1681(b), see supra part IV.A, and does not, as defendants' argue, impose an affirmative action/quota scheme on institutions receiving federal funds.
I now turn to the application of the Policy Interpretation to Brown University's athletic *211 program. I will first apply the three prong test to Brown's intercollegiate athletic program pursuant to 34 C.F.R. ง 106.41(c) (1) and 44 Fed.Reg. at 71,417-18 ("Effective Accommodation of Student Interests and Abilities"). I will then evaluate the "equal treatment" of Brown's teams by applying the remaining nine factors of the regulation's ten "Equal Opportunity" factors pursuant to 34 C.F.R. ง 106.41(c) (2)-(10) and 44 Fed.Reg. at 71,415-17 ("Equivalence in Other Athletic Benefits and Opportunities").
A. Prong One Applied
Plaintiffs have proven that Brown does not satisfy prong one. The gender balance of Brown's intercollegiate athletic program is far from substantially proportionate to its student enrollment. As discussed above, Brown University provides 555 (61.87%) intercollegiate athletic opportunities to men and 342 (38.13%) to women, whereas the undergraduate enrollment for the relevant year is 2796 men (48.86%) and 2926 women (51.14%). Brown currently offers 479 university-funded varsity slots for men and 312 university-funded varsity slots for women. It also provides 76 donor-funded varsity slots for men and 30 donor-funded varsity slots for women. Thus, because Brown maintains a 13.01% disparity between female participation in intercollegiate athletics and female student enrollment, it cannot gain the protection of prong one. Although Brown clearly does not meet the criteria of the first prong, defendants will still prevail if they can demonstrate that they satisfy the requirements of the second prong or if plaintiffs are unable to meet their burden of proof on prong three.
B. Prong Two Applied
Although Brown University has an impressive history of program expansion, defendants have failed to demonstrate that the University has maintained a continuing practice of intercollegiate program expansion for women, the underrepresented sex. As I noted at the preliminary injunction stage, Brown substantially expanded its athletic program for women in the 1970s. However, while I recognize that Brown has improved the quality of its women's program, the fact remains that since 1977 only women's indoor track in 1982 and women's skiing in 1994 have been added to its intercollegiate athletic program. Because merely reducing program offerings to the overrepresented sex does not constitute program "expansion," see supra part VI.A.2, the fact that Brown has eliminated or demoted several men's teams does not amount to a continuing practice of program expansion for women. In any case, Brown has not proven that the percentage of women participating in intercollegiate athletics has increased. Since the 1970s, the percentage of women participating in Brown's varsity athletic program has remained remarkably steady. See Cohen, 809 F. Supp. at 991.
C. Prong Three Applied
Prong three would excuse Brown's failure to provide substantially proportionate participation opportunities only if Brown fully and effectively accommodated the underrepresented sex. Here, the underrepresented sex is women and thus I focus my attention on the extent to which Brown's women athletes are accommodated. I find that Brown has not fully and effectively accommodated the interest and ability of the underrepresented sex "to the extent necessary to provide equal opportunity in the selection of sports and levels of competition available to members of both sexes." 44 Fed.Reg. at 71,417.
Plaintiffs have successfully established that Brown has not fully and effectively accommodated the interest and abilities of women, the underrepresented sex at Brown. In its affirmance of this Court's ruling on the preliminary injunction, the First Circuit recognized the potential difficulty in identifying "interest and ability" where student plaintiffs seek the creation or elevation to varsity status of an entirely new team. See Cohen, 991 F.2d at *212 904. Here, however, plaintiffs have introduced the testimony of student athletes, coaches and experts to verify that at least four existing teams have long been participating in competitive schedules and are capable of competing at Brown's highest varsity level. There are interested women able to compete at the university-funded varsity level in gymnastics, fencing, skiing, and water polo.
Brown fails to comply with prong three in two respects. First, Brown has failed to increase the number of intercollegiate participation opportunities available to the underrepresented sex where it could do so by elevating a team with demonstrated interest and ability from below intercollegiate status to intercollegiate status. Specifically, I find that Brown has failed to accommodate fully and effectively the underrepresented sex by maintaining women's water polo at club status and by demoting women's gymnastics where these teams have demonstrated the interest and ability to operate as varsity teams. I have already determined that the women's water polo team operates as a traditional club sport rather than as an "intercollegiate" team under the definition of the Policy Interpretation. See supra note 28 and accompanying text. I have also determined that although gymnastics is technically a donor-funded varsity at this time, it will in fact cease to exist, within a few seasons, at an intercollegiate varsity level in the absence of university funding. Neither team can compete as an intercollegiate varsity team if denied university-funded status. See supra part II. Thus, Brown violates prong three by maintaining women's water polo at club status and by, in effect, demoting women's gymnastics to club status, although officially designating it a "donor-funded" varsity team.
Second, Brown has failed to maintain and support women's donor-funded teams at Brown's highest level, thus preventing the athletes on these teams from developing fully their competitive abilities and athletic skills. Specifically, I find that Brown has failed to accommodate fully and effectively the underrepresented sex by maintaining women's fencing and women's skiing at a donor-funded level where each of these teams has demonstrated the interest and ability to operate as a university-funded varsity team and where donor-funded status has prevented each of these teams from reaching its athletic potential. Even though both athletic tiers offer participation opportunities in Brown's "intercollegiate" program, I have already determined that there are substantial qualitative differences between university and donor-funded teams. The differences preserved by Brown rise to such a level that the women participating on donor-funded varsities are not being fully and effectively accommodated. Thus, Brown violates prong three by maintaining these two women's teams at the donor-funded varsity level.
I recognize that this second basis for finding a violation is a new application of prong three. However, while prong three has not yet been read by any court to require an institution to upgrade any teams from within its intercollegiate athletic program, no other court has been presented with the factual situation of an officially maintained, two-tiered intercollegiate program.
It is true that prong three's "full and effective accommodation" language could be read to require only that an institution elevate or create athletic program offerings for the underrepresented sex from outside of its present intercollegiate program. However, Brown has created two distinct levels of athletics within the Policy Interpretation's definition of "intercollegiate" athletics; this unique factual situation calls for a more comprehensive interpretation of the "full and effective accommodation" language of prong three. As the First Circuit noted, institutions "must remain vigilant, `upgrading the competitive opportunities available to the historically disadvantaged sex as warranted by developing abilities among the athletes of *213 that sex,' [44 Fed.Reg. at 71,418], until the opportunities for, and levels of, competition are equivalent by gender." Cohen, 991 F.2d at 898.
Brown's restructured athletic program cannot be used to shield it from liability when in truth and in fact it does not fully and effectively accommodate the women athletes participating on donor-funded teams. It would circumvent the spirit and meaning of the Policy Interpretation if a university could "fully and effectively" accommodate the underrepresented sex by creating a second-class varsity status. The three prong test establishes a standard of equality that requires Brown to provide substantially equal numbers of intercollegiate athletic opportunities for men and for women, unless Brown either (1) steadily increases the number of such opportunities for women under prong two, or (2) fully and effectively meets the athletic interest and ability of women under prong three such that Brown cannot further improve the athletic opportunities for women until their interest and abilities further develop. Clearly, the potential for athletic development and the level of competition of women's donor-funded teams are much less than that of university-funded teams; thus, women on the ski and fencing teams can be more fully accommodated within the structure of Brown's established varsity athletic program.
Finally, I note that, in this instance, Brown cannot excuse its failure to accommodate the interest and ability of the women athletes on these four teams by citing an absence of "a reasonable expectation that intercollegiate competition in [these sports] will be available within the institution's normal competitive regions." 44 Fed.Reg. at 71,418. The evidence demonstrates that adequate intercollegiate competition exists within Brown's normal competitive region for each of these four teams.
D. Equal Treatment Factors Applied
There is an additional ground upon which to rest a finding of Title IX violation.
Brown's program offerings, as currently allocated by gender within the two-tiered structure of the intercollegiate varsity program violate the "treatment" aspect of the regulation. See 34 C.F.R. ง 106.41(c) (2) (10) ("Equal Opportunity" factors); 44 Fed. Reg. at 71,415-17 ("Equivalence in Other Athletic Benefits and Opportunities"). At Brown, far more male athletes are being supported at the university-funded varsity level than are female athletes, and thus, women receive less benefit from their intercollegiate varsity program as a whole than do men from their intercollegiate varsity program as a whole. This inequity is a consequence of the qualitative differences between the benefits enjoyed by university and donor-funded varsity teams. Donor-funded teams are not provided with treatment equivalent to that accorded university-funded teams with regard to at least the following factors: equipment and supplies (ง 106.41(c) (2)), travel and per diem allowance (ง 106.41(c) (4)), opportunity to receive coaching (ง 106.41(c) (5)), assignment and compensation of coaches (ง 106.41(c) (6)), and training services (ง 106.41(c) (8)). See supra notes 613, 29-30. These differences were not settled by the parties as to the donor-funded varsity teams since the Settlement Agreement does not resolve any program-wide treatment inequities. Rather, it settles only the relative support afforded to men's and women's university-funded teams as to which there is no dispute concerning their university-funded varsity status.
Having determined that Brown University is in violation of Title IX, I now address the remedy required. This Court has the authority to mandate specific relief. See Franklin v. Gwinnett County Pub. Sch., 503 U.S. 60, ___ _ ___, 112 S. Ct. 1028, 1035-37, 117 L. Ed. 2d 208 (1992). However, as I stated *214 in my earlier ruling, I have no desire to "micromanage" Brown's athletic program. Furthermore, I believe that judicial restraint is warranted at this point in light of the wide latitude afforded institutions under the regulations. Title IX does not require an institution to provide any athletic opportunities to its students. What it does require is that an institution provide equal opportunity to both genders in any program it chooses to offer. Thus, "Brown may not ... operate an intercollegiate program that disproportionately provides greater participation opportunities to one sex in relation to undergraduate enrollments, where there is no evidence of continuing program expansion [for the underrepresented sex] or effective accommodation of the interests and abilities" of the underrepresented sex. Cohen, 809 F. Supp. at 999.
Brown may achieve compliance with Title IX in a number of ways. It may eliminate its athletic program altogether, it may elevate or create the requisite number of women's positions, it may demote or eliminate the requisite number of men's positions, or it may implement a combination of these remedies. I leave it entirely to Brown's discretion to decide how it will balance its program to provide equal opportunities for its men and women athletes. I recognize the financial constraints Brown faces; however, its own priorities will necessarily determine the path to compliance it elects to take.
Defendants frequently raised the specter of being forced by financial constraints to eliminate men's athletic opportunities in order to achieve compliance under plaintiffs' interpretation of the law. I feel compelled to point out that an institution has much flexibility, even within a finite resource base. Dr. Lopiano was asked whether it would be prudent for a university confronted with severe budgetary constraints to cap the size of men's teams rather than add women's teams to its athletic program. She responded, addressing the false dichotomy posed:
I believe that philosophically in any case where you have a previously disadvantaged population that you're trying to bring up to snuff to the advantaged population, that it's a bad idea to bring the advantaged population down to the level of the disadvantaged population. [T]he whole idea [of Title IX] is to add participation opportunities for women. And it's unfortunate that across the country that in the name of maintaining the standard of living of football team[s] or the standard of living of one or two special men's sports, that men's sports are being cut and women's gender equity under Title IX [is] being blamed for that.
Trial Tr. 12/2/94 at 59.
Thus, defendants' plea that "[t]here is nothing further Brown can do except cut, cap or eliminate men's teams," Defs.' Post-Trial Mem. at 39, is simply not true. Brown certainly retains the option to redistribute its resources in a way that may slightly reduce the "standard of living" for its university-funded varsity sports in order to expand the participation opportunities for its women athletes and closer approach equal opportunity between its male and female athletes. Whether it will follow this course of action is, of course, well within its discretion.
For all of the reasons stated above, Brown University is in violation of Title IX and supporting regulations.
Brown University is hereby ordered to submit to this Court, within 120 days, a comprehensive plan for complying with Title IX. This portion of the order is stayed pending appeal.
It is hereby further ordered that, in the interim, the preliminary injunction, as outlined in my Opinion and Order of December 22, 1992, Cohen v. Brown Univ., 809 F. Supp. 978, 1001 (D.R.I.1992), will remain in full force and effect.
 Title IX applies only to those institutions that receive federal financial assistance. 20 U.S.C. ง 1681. Brown University receives federal funds and is therefore required to comply with Title IX.
 Financial considerations prompted the demotions. Brown expected to save $77,823 per year as a result of these reductions. Most of the savings realized by eliminating the funding for the four teams came from the women's athletic budget. Brown saved $62,028 by demoting the women's volleyball and gymnastics teams and $15,795 by demoting the men's golf and water polo teams. However, the status reductions did not appreciably affect the athletic participation gender ratio.
 See, e.g., Jennifer L. Henderson, Gender Equity in Intercollegiate Athletics: A Commitment to Fairness, 5 Seton Hall J.Sport L. 133 (1995); Janet Judge et al., Perspective, Gender Equity in the 1990s: An Athletic Administrator's Survival Guide to Title IX and Gender Equity Compliance, 5 Seton Hall J.Sport L. 313 (1995); Diane Heckman, The Explosion of Title IX Legal Activity in Intercollegiate Athletics During 1992-93: Defining the "Equal Opportunity" Standard, 1994 Det. C.L.Rev. 953 (1994); Teresa M. Miguel, Title IX and Gender Equity in Intercollegiate Athletics: Case Analyses, Legal Implications, and the Movement Toward Compliance, 1 Sports Law.J. 279 (1994); B. Glenn George, Miles to Go and Promises to Keep: A Case Study in Title IX, 64 U.Colo. L.Rev. 555 (1993); William E. Thro & Brian A. Snow, Cohen v. Brown University and the Future of Intercollegiate and Interscholastic Athletics, 84 Ed.Law Rep. 611 (1993); Jill K. Johnson, Note, Title IX and Intercollegiate Athletics: Current Judicial Interpretation of the Standards for Compliance, 74 B.U.L.Rev. 553 (1994); R. Lindsay Marshall, Case Comment, Cohen v. Brown University: The First Circuit Breaks New Ground Regarding Title IX's Application to Intercollegiate Athletics, 28 Ga.L.Rev. 837 (1994); Mary Beth Petriella, Note, Injunctive Relief โ" Title IX โ" Interim Preliminary Injunction Reinstating Varsity Status to Demoted Collegiate Athletic Teams is Available When That Team Alleges a Title IX Violation and Litigation is Pending โ" Cohen v. Brown ..., 4 Seton Hall J.Sport L. 595 (1994); Catherine Pieronek, Note, A Clash of Titans: College Football v. Title IX, 20 J.C. & U.L. 351 (1994); and William H. Webb, Jr., Case Comment, Sports Law โ" Cohen v. Brown University: The Promulgation of Gender Equity in Intercollegiate Athletics, 25 U.Mem.L.Rev. 351 (1994).
The defendants submitted one such piece for my consideration with their post trial memoranda. Walter B. Connolly, Jr. & Jeffrey D. Adelman, A University's Defense to a Title IX Gender Equity in Athletics Lawsuit: Congress Never Intended Gender Equity Based on Student Body Ratios, 71 U.Det.Mercy L.Rev. 845 (1994). This Court, however, can hardly accept as authoritative a law review article co-written by defense counsel.
 Brown football, however, is a Division I-AA sport and therefore participates at the second highest level of NCAA competition.
 Defendants repeatedly point to the fact that Brown's program offerings for women exceed the national average of women's sports per institution. However, Brown also exceeds the national average of men's sports per institution. The fact that Brown's athletic offerings are "extensive" cannot and does not excuse its failure to comply with Title IX. That other universities may be even more clearly in violation of Title IX's nondiscrimination mandate does not exonerate an institution that still provides unequal opportunities to its own male and female students.
 This "intermediate" program level has been in flux since its inception. Both the name and the treatment of these teams have changed over the last three years. In May 1991 the demoted teams were labeled "club varsity" teams. The athletic director announced that the four teams would no longer receive funds from the University but would continue to be treated in all other ways as a varsity team. The teams would remain eligible for NCAA competition if they continued to comply with NCAA requirements and if they were able to raise the funds to maintain a sufficient level of competitiveness. During the next season, however, it became apparent that "club varsities" were not receiving the same university privileges as "true" varsities. In December of 1991, the athletic director decided that the term "club varsity" was misleading and changed it to "intercollegiate club" in order to avoid misleading prospective student-athletes. By the time of trial, both the treatment and name of the intermediate teams had further evolved. Brown restored some of the traditional varsity privileges to these teams and now refers to them as "donor-funded," "gift-funded," or "unfunded" varsities. For ease of reference, I will use the term "donor-funded" varsity.
 While Brown, since the end of the 1993-94 season, has decided to provide donor-funded varsities with access to the weight and training rooms and with special admissions considerations, it is not yet clear to what extent all of these varsity advantages will actually be provided to donor-funded varsities on the same basis as they are provided to university-funded varsities. For example, free trainer services are not always available to donor-funded and university-funded varsity teams on an equal basis. Brown does not provide donor-funded varsities with other university amenities, such as office space, campus telephone privileges, secretarial support, and possibly laundry facilities and free interterm housing.
More importantly, Brown also does not guarantee that donor-funded teams will have coaching, equipment and supplies, money for travel (including meals and lodging) and for post-season competition. Brown guarantees all of the above to its university-funded varsities.
 For instance, when volleyball was originally demoted, two schools called to cancel their competitions with Brown because of the team's demotion. The volleyball coach also cancelled one tournament because she was unsure if the team would be able to raise enough money to afford the trip. Prelim.Inj.Hr'g Tr. 11/2/92 at 138-39.
 The women's ski team, for example, during the 1993-94 season could only afford to hire one coach to train the men's and women's ski teams two nights a week and another to show up for competitions. Currently, the donor-funded women's ski team can only afford to contribute $1,000 toward coach's compensation; the men's club ski team will match this amount, and the two teams will share the coach. The team captain is responsible for any other training the team does. Trial Tr. 9/29/94 at 11-12, 22, 31.
Similarly, the gymnastics team, in the absence of the preliminary injunction, would have been drastically underfunded. The gymnastics team has never been able to raise close to the amount of money required to retain its coach at her current salary, much less to meet other operating expenses of the team. See Trial Tr. 9/27/94 at 171.
 Eileen Rocchio, a member of the gymnastics team, testified that she "wouldn't have come to Brown unless [she] knew that it was a varsity team." Prelim.Inj.Hr'g Tr. 10/28/92 at 137. See also Prelim.Inj.Hr'g Tr. 10/29/92 at 19-20 ("I would never have gone to Brown knowing and ... having to worry about fund raising and lockers and ... trainers and not getting to fulfill my potential as ... an athlete. Those are things that I ... would go to a school so that they could provide me with, not that I would have to worry about finding them on my own").
The inability to recruit team members of a high caliber greatly affects the competitive level of a team. For example, when the volleyball team was demoted and lost its recruiting appeal, the team had no alternative but to accept far less accomplished athletes, and it effectively ceased to exist as a varsity level team. Trial Tr. 9/29/94 at 188-90.
 Elliot Lilien, former head coach and current assistant coach of men's and women's fencing testified as follows:
Q: So did you consider the amount of finances that you had in picking a schedule and deciding which teams against which to compete?
. . . . .
A: Now if we were to aim at a higher level of competition, no, we didn't have enough [funds] then. (Trial Tr. 9/27/94 at 102).
. . . . .
A: I've always thought Brown was a gold mine for fencing and if there was sufficient commitment made to it, Brown could compete with anyone in the country. (Trial Tr. 9/27/94 at 104).
See also Trial Tr. 9/27/94 at 31-33 (fencing team did not join Ivy League round robin due to insufficient commitment from Brown; team could not compete successfully in the round robin without coaching, travel allotments and practice space comparable to that afforded to the other teams participating in that league).
 Gymnast was unable to attend the NCAA Eastern Regional Championships due to team's financial constraints. Prelim.Inj. Hr'g Tr. 10/29/92 at 18.
 The captain of the women's ski team testified, for example, that the gates the team uses in practice have not been replaced in over ten years due to the team's lack of funds. "[I]f they break, we're in big trouble because we have no money to replace any of our training gates. So we would be at a loss, a big loss if one of the gates breaks which is very plausible." Trial Tr. 9/29/94 at 34.
A coach of the women's fencing team expressed similar concerns about the team's limited access to electric equipment. He explained that "electric weapons are somewhat differently weighted and feel differently [from conventional weapons]." Trial Tr. 9/27/94 at 107. Electric equipment is used in competition, but the team "seldom used electric equipment in practice because of the fear that it would break and [the team] wouldn't be able to โ" to replace it." Trial Tr. 9/27/94 at 20.
 This is a non-inclusive list; there may be other women's club sports with sufficient interest and ability to warrant elevation to varsity status. Plaintiffs, however, did not introduce any substantial evidence at trial to prove the existence of other women's club teams meeting the criteria.
 For a complete definition of intercollegiate athletics as defined by the relevant agency document, see infra part VI.A.1.a.
 Although the 1994/95 season has already started, with new program offerings in place, I have chosen to use data from 1993/94 to calculate participation numbers. The 1994/95 participation numbers are necessarily incomplete and unreliable given that Spring sports had not yet begun at the close of the trial. However, the lists below do reflect the 1994/95 upgrade in status of women's volleyball to university-funded varsity and women's skiing to donor-funded varsity. I have used this combination of 1993/94 numbers and 1994/95 program offerings in order to reflect most accurately the current program.
I derived the 1993/94 participation numbers from the NCAA Squad Lists, Pls.' Ex. 32, which I find to be the most accurate, if imperfect, representation of varsity participants available. These lists are maintained by the athletic department and periodically updated by team coaches. I did not include student-athletes who either quit or were cut from the team, except where the change occurred very late in the season. I did, however, count participants marked as injured under the assumption that they remained team members.
I decline to rely upon the NCAA Sport Sponsorship Forms submitted to the NCAA by the athletic department at the end of the year. Joan Taylor, Brown's Associate Athletic Director and the Senior Women's Administrator, testified that the number of varsity participants on the sponsorship forms is a rough figure "because the N.C.A.A.'s reason for this form is only to make sure that we have at least met the minimum criteria for Division I sport sponsorship." Trial Tr. 10/3/94 at 169. The NCAA Squad Lists, on the other hand, report on the individual eligibility of each member of the team. For that reason, the NCAA Squad Lists, unlike the NCAA Sports Sponsorship Forms, list the names of individual athletes, thus permitting verification of team participation numbers. The Squad Lists are maintained by the athletic department and are updated periodically throughout the season.
 During the 1994/95 season Brown guaranteed the volleyball team university-funded varsity status for the next five years.
 Gymnastics is currently supported as a university-funded team as required by court order. However, in the absence of this order, Brown has acknowledged that it would demote gymnastics to donor-funded status. See Trial Tr. 11/21/94 at 25. I therefore include gymnastics as a donor-funded team.
It is arguable that gymnastics should be counted only as a club team because the testimony demonstrated that gymnastics would, within a few seasons, effectively cease to exist as an intercollegiate varsity team if it were denied university funding.
 Women's skiing was upgraded in 1994/95 to donor-funded status. Prior to this season, the athletic department did not maintain an NCAA Squad List for the team because the team was a club sport. I therefore rely on the 1994/95 NCAA Squad List to count ski team participants.
 There are four women who participate on men's teams: three female coxswains on the men's crew team and a female golfer on the men's golf team.
 The treatment of men's and women's teams addressed by the Settlement Agreement is governed by 34 C.F.R. ง 106.41(c) (2)-(10). See infra part IV.B.
 District and circuit courts commonly accord appellate court pronouncements of law rendered in reviewing preliminary injunctions precedential value sub silentio. For example, Planned Parenthood League of Mass. v. Bellotti, 641 F.2d 1006 (1st Cir.1981) considered a district court's denial of a preliminary injunction and has been relied upon by district courts and by the Court of Appeals. See, e.g., Massachusetts v. Secretary of Health and Human Servs., 899 F.2d 53, 66 (1st Cir.1990) (stating "[t]he regulations also run afoul of our holding in [Bellotti], in which we invalidated a twenty-four hour waiting period requirement contained in a Massachusetts statute regulating abortions"), vacated and remanded sub nom. Sullivan v. Massachusetts, 500 U.S. 949, 111 S. Ct. 2252, 114 L. Ed. 2d 706 (1991); and Women's Medical Center of Providence, Inc. v. Roberts, 530 F. Supp. 1136, 1143, 1145, 1147, 1149, 1150-51, 1153 (D.R.I.1982).
 For the shift in agency responsibility for Title IX administration, see Cohen, 991 F.2d at 895 (describing the shift of authority from Health Education and Welfare ("HEW") to its successor agency, DED, when HEW was split into the Department of Health and Human Services ("HHS") and DED). The regulations and the Policy Interpretation discussed below were originally promulgated by HEW, in 1975 and 1979, respectively. For the purposes of this discussion, however, I will refer to DED as the promulgating agency "because the agency adopted the very same regulation which the Policy Interpretation was issued to interpret." Horner, 43 F.3d at 273 n. 6 (citing Cohen v. Brown Univ., 991 F.2d 888, 896 n. 10 (1st Cir.1993)).
 This provision governs the distribution of athletic scholarships under the general caption of financial assistance in education programs. Brown University, as an Ivy League institution, does not grant athletic scholarships to its students. Therefore, section 106.37(c) is not at issue in this case, and I do not address it.
 At the time the Policy Interpretation was originally published, the Office of Civil Rights operated within HEW. DED subsequently inherited the final Policy Interpretation. When DED repromulgated the exact same regulations which the Policy Interpretation was issued to interpret, see supra note 23, DED sent "an unmistakably clear signal of the agency's satisfaction with the Policy Interpretation." Cohen, 991 F.2d at 896 n. 10.
 The two part test is not at issue in this case and is listed solely for the purpose of providing a more complete context for the provisions that are central to this case. At the preliminary injunction stage I found, tentatively, that "the competitive schedules at Brown provide men and women with `equivalently advanced competitive opportunities.'" Cohen, 809 F. Supp. at 994. Nothing introduced at the trial on the merits alters my opinion that the competitive schedules offered to men and women at the university-funded varsity level are equivalent to each other and that the competitive schedules offered to men and women at the donor-funded varsity level are equivalent to each other.
 Because I do not rely on the Investigator's Manual, except to refute an incorrect reliance upon it, see infra note 51, I decline to address specifically the degree of deference to which this document is entitled.
 However, "its general principles will often apply to club, intramural, and interscholastic athletic programs, which are also covered by regulation." 44 Fed.Reg. at 71,413. Here, plaintiffs have not challenged Brown's non-varsity programs.
 For the majority of the trial, defense counsel represented that intercollegiate athletic participation opportunities at Brown relate only to university-funded and donor-funded varsity teams. Trial Tr. 11/1/94 at 89.
The testimony later offered in support of the position that some of the club teams engage in "intercollegiate competition" was given by Brown's Assistant Athletic Director, Jeffrey Ward. Conceding his uncertainty, he testified that he believed that the following club teams might engage in substantial competition with varsity teams at other colleges: women's water polo, women's soccer, women's skiing, women's and coed sailing, and men's skiing. Trial Tr. 12/8/94 at 139-40. However, in addition to being unable to ascertain from the record which of the clubs' competitors were varsity teams, this Court has no way of determining the number of participants (with the exception of women's water polo) on each of these club teams. As Brown acknowledged in its answer to plaintiffs' interrogatories, club team rosters are not maintained by either the athletic department or the student activities office. Pls.' Ex. 1 at 6-7.
 While both men's and women's teams at the donor-funded level may operate at a disadvantage compared to university-funded teams, the burden of this status falls more heavily on women's teams. There was ample testimony that women's teams are less able than men's teams to raise necessary funds from private donations. Although defendants attributed this fund raising disparity to lack of effort on the part of women's coaches and teams, I find Coach Robert Rothenberg's explanation more persuasive. As head coach of both the men's and women's track teams, he testified that he and his staff worked equally hard to raise money for both teams but were never able to raise as much money on behalf of the women's team as they were able to raise for the men's team. He attributed this disparity to the financial bases available and to traditional patterns of giving, stating:
[The] women's track and field program at Brown started about 1978, the men's started about a hundred years before. When you look at the giving, you are not going to get major contributions [from] men or women that are in their twenties or thirties. Your biggest contributions will come from people that have raised their children, that are out, that have accumulated some money and are looking for [a] way to [give] back to the university in the sport that he or she participated in. I think traditionally men have probably controlled more of that money available in families. But for whatever the reason, there's a tremendous difference in the ability to fund raise for men and for women and I've been doing [it] for 13 years.
Trial Tr. 12/1/94 at 188-89.
 Kim Havell, the 1994-95 women's ski team captain illustrated this point:
Q: Have you enjoyed all of the aspects of competing to become a top-flight women's ski program at Brown?
A: I would say no to that in the sense that it's been a lot of โ" it's hard to get everything together. Every year the captains have to put in a lot of work to make sure everyone knows when the races are, when we have to meet. All the details we have to worry about too, the team members have to consider. And it's not just set up, you go, you do the best you can[. Y]our mind is on the duties you have to the team, not just on competing and doing the best that you can.
Trial Tr. 9/29/94 at 75.
 The Policy Interpretation applies separately to each level of athletic program within an institution. The Policy Interpretation states that it "is designed specifically for intercollegiate athletics, however its general principles will often apply to club, intramural, and interscholastic athletic programs...." See 44 Fed.Reg. at 71,413. Brown's unusual two-tiered treatment of its varsity program defies easy categorization. Because Brown has created another layer of athletic offerings, it is arguable that this Court should assess compliance by independently applying the Policy Interpretation to each tier of its intercollegiate varsity program. This method would give effect to the "Overall Determination of Compliance" element of the Policy Interpretation's "Effective Accommodation" provision. See 44 Fed. Reg. at 71,418 (asking "[w]hether disparities in individual segments of the program with respect to benefits, treatment, services, or opportunities are substantial enough in and of themselves to deny equality of athletic opportunity"). See supra part IV.C.
I have chosen not to adopt this method; however, I note that the factual record could only support a finding of noncompliance pursuant to this alternative application.
 "[T]hink of the university that in one year has 54 men track participants and in year two has 70 men's track participants. Does that mean that [the athletic director] has to wave a magic wand over the athletic department and create a new team of 16 women?" Trial Tr. 12/16/94 at 52.
 At the preliminary injunction hearing, John Parry, Brown's Athletic Director from January 1979 to January 1990, was asked, "When you were at Brown, was it true, in your view, that Brown's activities effectively did predetermine the percentage of โ" the percentage of men and women participating in intercollegiate athletics?" He responded, "Yes." Prelim.Inj. Hr'g Tr. 10/27/92 at 130.
 Brown argues that under this "predetermination" analysis it is inconsistent to find that there are club team members who are interested and able to compete at the varsity level. If only recruited athletes have the ability to compete at the Division I varsity level for the purposes of prong one, defendants argue, plaintiffs cannot at the same time maintain that non-recruited club team athletes have the ability to compete at the varsity level for the purposes of prong three. Defs.' Reply Br. at 4.
Defendants err in conflating the analyses of prong one and prong three. Under prong one, the ability of recruited athletes is relevant only because Brown effectively limits participation on its varsity teams to recruited athletes, which in turn logically requires this Court to equate "participation opportunities" with actual participation rates, for the purposes of prong one. It is not the position of this Court that only recruited athletes could compete on Brown's varsity teams, rather the evidence has demonstrated that it is in fact the practice at Brown to rely almost exclusively on recruited athletes to field varsity teams. Under prong three, the ability of club athletes is a factor of the test itself, but the standard of prong three is generous: the under-represented sex need only demonstrate the interest and ability to compete in an "intercollegiate schedule" as defined in the Policy Interpretation. See 44 Fed.Reg. at 71,413 n. 1. A club team can demonstrate the interest and ability to compete in an "intercollegiate schedule" without presently operating at the same level of competitiveness maintained by longstanding university-funded Division I varsity teams.
 I have found that Brown "predetermines" the approximate number of varsity positions available to men and women; the concept of any measure of unfilled but available athletic slots does not comport with reality.
 Under the same rationale, it is incorrect to equate participation opportunities with the number of athletes permitted to travel with the team to "away" games, as determined by the NCAA, because Brown does not require coaches to maintain their teams at travel squad size. Thus, I reject Brown's proposal to measure participation opportunities by NCAA team travel squad sizes.
 A number of factors support this conclusion. First, Brown conceded that each coach employs his or her own philosophy in limiting team size. Because a current coach may make adjustments in his or her coaching practices over the years, and because a newly hired coach may implement different governing policies from that of a predecessor, only the current team size accurately reflects the number of participation opportunities actually offered to students. For instance, although previous swimming coaches had maintained a team with as many as 36 members, the new head coach of men's swimming testified that he did not plan to coach such a large team. Trial Tr. 12/8/94 at 114.
Second, the coaches' testimony demonstrated that the number of positions open to newcomers in any given year depended upon the number of members expected to graduate, the strength of the existing team, and the abilities of the prospective athletes. Because these latter variables may change team size slightly, team numbers from past years may not be a reliable indication of current participation opportunities.
 Although many men's and women's teams share the same name, they may not require the same number of team members in order to compete effectively. Different rules of play, including but not limited to, rules of substitution, amount of contact permitted, size of field, and number of specialized positions involved determine appropriate team size. For example, Dr. Donna Lopiano, Executive Director of the Women's Sports Foundation, and an accomplished four sport athlete honored in the softball hall of fame, discussed some of the differences between baseball and softball:
I think softball and baseball are two completely different sports, that baseball will have a larger team size than softball will primarily because of the [effect of] pitching. It is typical on a baseball team to carry anywhere from six to ten pitchers because of the stress that is involved in throwing a baseball.... [I]n the sport of softball a single pitcher can pitch 44 consecutive innings, and I have done so on a number of occasions, without any stress on the shoulder joint because your arm is closer to your body and in a much more well supported position when you're throwing the ball. Where a baseball pitcher typically cannot throw more than 11 or 12 innings at a time and needs two or three days rest to recover. So just that factor alone will result in participation differences of 10 to 20 percent on the size of a baseball to the size of a softball team.
Trial Tr. 12/2/94 at 14-15. See also Trial Tr. 9/27/94 at 39 (men's and women's fencing coach testifying that women fence in only one weapon [and now in two] whereas men fence in three, thus resulting in many more positions available for men on a fencing team).
 Defendants explain their theory with a hypothetical. In this hypothetical, there are 1000 men and 1000 women in the student body at a university that has 150 gender-neutral athletic slots available. 500 men and 250 women in the student body (50% of men enrolled and 25% of women enrolled) are interested in filling these spots, and a random lottery is conducted in which students are offered a guaranteed position on the team. Where equal numbers of randomly selected men and women are offered such an opportunity, 50% of the men and 25% of the women will accept. Thus, although equal numbers of men and women are offered a position, 100 men and 50 women, due to the relative interest of their gender, will fill the 150 slots. See testimony of Dr. Finis Welch, Trial Tr. 11/22/94 at 13-15 and 19-24.
 At the preliminary injunction stage, defendants similarly sought to make the relative interests of men and women the relevant inquiry. However, at that time, defendants propounded this theory under the authority of prong three. See infra part VI.A.3.a. The First Circuit squarely rejected this analysis with respect to prong three, Cohen, 991 F.2d at 899-900, and I now decline to adopt it with regard to prong one.
 Cf. Cohen, 991 F.2d at 902 (rejecting analogy to Title VII with respect to allocation of burden of proof on elements of three prong test and noting the substantial differences between Title IX and Title VII statutory purposes and between the education and employment contexts). The factors articulated by the First Circuit in declining to import Title VII burdens of proof also support my rejection of defendants' attempt to superimpose the meaning of discrimination in Title VII upon the plain language of Title IX.
 Defendant's distinguished expert, Dr. Finis Welch, acknowledged that there is no single factor by which to measure that degree of athletic interest that will reliably be acted upon when the opportunity is present. "[T]he interest measures that we get vary. I mean there is an overall pattern and they fall within the pattern, but there's โ" there's no one place that we can go to get a rock solid measure of interest." Trial Tr. 11/29/94 at 115.
Dr. Welch illustrated the difficult in quantifying interest when he was asked by the Court, "[W]hat constitutes an interested man or an interested woman?" Dr. Welch responded, "That's hard. And that's what we're going to do is draw information from various sources, because we don't know." Dr. Welch listed a variety of survey sources and concluded, "I would try to combine the attitudinal information, the survey kind of question with the realization, `What do I see by way of people showing up.'" Trial Tr. 11/22/94 at 15-17.
 Given the difficulty of measuring the relative interests of men and women, it would be almost impossible for an institution to remain in compliance with Title IX by staying abreast of the ever-changing relative "interests" of its male and female students and adjusting its program offerings accordingly. Because defendants' interpretation would require substantial proportionality between the gender balance of its athletic program offerings and the gender balance of interested prospective student-athletes, constant rebalancing would be necessary to maintain compliance, thereby eliminating the ability of an institution to verify easily that it falls within the "safe harbor" that prong one provides.
 Question 9 on the application form asked applicants to list activities that they might pursue at Brown. Pls.' Ex. 112. Dr. Welch acknowledged that the answers to question 9 could not be used to measure precisely those interested in participating in varsity athletics, as opposed to club, intramural, and recreational sports. Trial Tr. 11/29/94 at 114-15.
 As Dr. Welch noted in his report:
The fact that only small minorities of those indicating interests actually participate in varsity sports is indicative that reported interests are not necessarily firm commitments. These numbers suggest that applicants expressing potential interests in participating should be viewed as a mixture consisting of some who are seriously committed to participating and others who are, perhaps, not as committed. The alternative treatments of these groups can result in major shifts in calculated female/male interest ratios.
Defs.' Ex. IIIII, Table 30A at 92.
 Defendants colorfully illustrated this concern in their closing statement: "I can go onto any campus in this country, Your Honor, and probably find somebody interested in playing Russian roulette. That doesn't mean that that's an unmet need that ought to be raised to the dignity of an intercollegiate varsity." Trial Tr. 12/16/94 at 71. This portrayal of the practical consequences of this Court's interpretation fails to acknowledge the safeguards embodied in the Policy Interpretation as a whole. An institution is not required to create or elevate a team in the absence of the reasonable expectation of intercollegiate competition within the institution's normal competitive region. 44 Fed.Reg. at 71,418. It follows that there must be both a sufficient number of other teams competing in that region to provide intercollegiate competition and enough interested and able individuals for the institution to draw upon and build a competitive team. But see 44 Fed. Reg. at 71,418 (institutions may be required by Title IX regulations to "actively encourage" the development of such competition). As plaintiffs noted in their closing statement: "We represent a class. We don't represent any one individual and [we don't say] this person alone needs this opportunity. If you provide equal opportunities to the class members, to the potential athletes, we're satisfied even though one athlete here or one athlete there is disappointed." Trial Tr. 12/16/94 at 34.
 Thus, for example, under defendants' interpretation, where a school with a student enrollment of 1000 men and 1000 women has 500 interested men and 250 interested women, the school satisfies prong three if it provides 100 athletic positions for men and 50 for women. Contra Cohen, 991 F.2d at 899.
 See supra note 44 and accompanying text.
 See supra note 43 and accompanying text.
 Contrary to this Court's view of the inherent difficulties with requiring a determination of relative interests, defendants argue that OCR plainly advocates the use of surveys as a simple measure of the relative accommodation of interest and abilities:
[w]e have looked at what the Office of Civil Rights has instructed us to look at in terms of determining interests and abilities as they change in a university context over time.... [I]f any court seeks to give deference to the Office of Civil Rights, it must give deference to how they tell us to determine interest and abilities.... And I quote [page 27 of the 1990 Investigator's Manual, which], points out [that] to determine interest and abilities [an institution should look to], quote, "involved club, intramural, feeder school, community, regional, physical education, and other programs." And most importantly, your Honor, they say on page , quote, "a survey is mentioned most often since it is usually the simplest method for the institution and O.C.R. to determine interest and abilities."
Trial Tr. 12/16/94 at 85-87. In fact, however, a closer reading of the Athletics Investigator's Manual reveals that OCR identifies surveys as a simple way to identify which additional sports might appropriately be created to achieve compliance. The Manual states on page 27 "[a] survey or assessment may be required as a part of a remedy when OCR has concluded that an institution's current program does not equally effectively accommodate the interests and abilities of students." Valerie M. Bonnette & Lamar Daniel, Department of Education, Title IX Athletics Investigator's Manual 27 (1990). Thus, a survey of interests would follow a determination that an institution does not satisfy prong three; it would not be utilized to make that determination in the first instance.
 In fact, however, plaintiffs did not prove that there are women in addition to those participants in gymnastics, skiing, fencing and water polo who have the interest and ability to compete at the intercollegiate level in a sport for which there is a reasonable expectation of competition in Brown's normal competitive region. Even if Brown were to upgrade each of these teams to university-funded status, a 12% disparity would still exist between female participation in intercollegiate athletics and female representation in the student body, resulting in a failure to satisfy prong one. Therefore, it is most likely that even Brown could find that prong three offers protection for its intercollegiate athletic program where prong one does not.
 As I determined above, see supra part VI.A.1, "intercollegiate athletic participation opportunities" are measured by counting the number of participants on both university and donor-funded varsity teams. In Brown's case, the most accurate, if imperfect, participation numbers appear on the NCAA Squad Lists. See supra note 16 and accompanying text.
 The record demonstrates that the women's water polo team has the interest and ability to compete as a university-funded varsity team; however, the evidence does not demonstrate that this club team presently operates as an intercollegiate team. I note that the fact that this team does not engage in prong one's definition of "intercollegiate competition" does not mean that it, by definition, cannot demonstrate the interest and ability under prong three to do so if upgraded.
 For discussion of Brown's varsity and club competitors in water polo, see Trial Tr. 10/3/94 at 24; Trial Tr. 10/3/94 at 61-2; and Pls.' Ex. 39. For discussion of Brown's varsity and club competitors in skiing, see Trial Tr. 9/28/94 at 29-30; and Trial Tr. 9/29/94 at 12-13. For discussion of Brown's varsity and club competitors in fencing, see Trial Tr. 11/21/94 at 26-29; Trial Tr. 9/27/94 at 8, 76, 78. For discussion of Brown's varsity and club competitors in gymnastics, see Trial Tr. 9/27/94 at 184-86.