Hopwood v. State of Tex.

999 F. Supp. 872, 1998 U.S. Dist. LEXIS 5339, 1998 WL 180320

• Court: District Court, W.D. Texas
• Decided: March 20, 1998
• Docket: A 92 CA 563 SS
• Status: Published

Opinion

MEMORANDUM OPINION

SPARKS, District Judge.

This is the continuing case of four white students who contend they were denied admission to the University of Texas School of Law in 1992 as a result of procedures granting preferences in admission to black and Mexican-American applicants.

I.

On September 29,1992, the plaintiffs Cheryl J. Hopwood, Douglas W. Carvell, Kenneth R. Elliot, and David A. Rogers filed suit under 42 U.S.C. §§ 1981 and 1983 (West 1994) and Title VI of the Civil Rights Act of 1964, 42 U.S.C. § 2000d (West 1994) (“Title VI”), 1 against the defendants the State of Texas, the University of Texas Board of Regents, the University of Texas, the University of Texas School of Law (“the law school”), and various University of Texas officials in their official capacities. 2 The plaintiffs sought injunctive and declaratory relief, as well as compensatory and punitive damages. The Court held an eight-day bench trial in the case in May 1994.

On August 19, 1994, the Court issued its memorandum opinion in Hopwood v. State of Texas, 861 F.Supp. 651 (W.D.Tex.1994) (“Hopwood I ”). In deference to controlling Supreme Court precedent, the Court declined to declare the law school’s use of racial preferences in its admissions system unconstitutional per se, see id. at 553-54, and instead applied strict scrutiny to the law school’s admissions system, see id. at 568-69. Relying primarily on Justice Powell’s opinion in Regents of the University of California v. Bakke, 438 U.S. 265, 98 S.Ct. 2733, 57 L.Ed.2d 750 (1978), the Court found that the law school’s use of racial preferences for the purpose of achieving a diverse student body served a compelling state interest under the Fourteenth Amendment. See Hopwood I, 861 F.Supp. at 569-571. Additionally, the Court found that the remedial nature of the admissions process, in which racial classifications were used as a means of overcoming the present effects of past- race discrimination, served a compelling governmental interest. See id. at 571-73. The Court ultimately concluded, however, that the law school’s use of separate admissions procedures for minorities and nonminorities 3 was not narrowly tailored to achieve those compelling interests because the process prevented any meaningful comparative evaluation among applicants of different races. See id. at 573-579. The Court therefore entered declaratory judgment that the law school’s 1992 admissions procedures violated the Fourteenth Amendment. See id. at 582.

' The Court then considered whether any of the four plaintiffs was denied admission in 1992 as a result of the constitutionally impermissible method in which the law school considered race in its admissions procedures. In determining which party bore the burden of persuasion on that issue, the Court adopted a burden-shifting scheme similar to that used in employment discrimination cases brought pursuant to Title YII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e (West 1994) (“Title VII”), and placed the ultimate burden of persuasion on the plaintiffs. 4 See id. at 579-80. The Court found that the plaintiffs failed to establish by a preponderance of the evidence that they would have been offered admission to the law school under a constitutional admissions process. See id. at 580-82. The Court therefore declined to award the plaintiffs injunctive relief (that is, an immediate order of admission) or compensatory damages. 5 See id. at 582-83. Furthermore, given the undisputed remedial goals of the admissions program, as well as the fact that the law school adopted the program in a good faith effort to comply with federal guidelines under Title YE, the Court declined to issue an award of punitive damages. See id. at 583. Finally, because the law school had substantially modified its admissions procedures by the end of the trial to provide for individual comparison among minority and nonminority applicants (thereby remedying the infirmity identified by the Court in its opinion), the Court declined to issue any permanent injunctive relief against the law school. See id. at 582.

A three-member panel of the Fifth Circuit Court of Appeals reversed and remanded the Court’s decision in part in Hopwood v. State of Texas, 78 F.3d 932 (5th Cir.), cert. denied, 518 U.S. 1033, 116 S.Ct. 2581, 135 L.Ed.2d 1095 (1996) (“Hopwood II ”). 6 The Fifth Circuit declared that the law school’s use of racial preferences served no compelling state interests under the Fourteenth Amendment. 7 See id. at 941-55. The divided Fifth Circuit panel therefore directed the law school not to use race as a factor in admissions, although it declined to order any permanent injunctive relief to that effect. See id. at 958. Furthermore, the Fifth Circuit disagreed with the Court’s allocation of the burden of proof on the issue of causation — whether any of the four plaintiffs would have been admitted to the law school under a constitutional system. Using the burden-shifting scheme of Mt. Healthy City School District Board of Education v. Doyle, 429 U.S. 274, 97 S.Ct. 568,50 L.Ed.2d 471 (1977), the Fifth Circuit determined the law school bore the burden of proof on that issue because the plaintiffs had successfully established the unconstitutionality of the admissions system. See Hopwood II, 78 F.3d at 955-57.

On remand, the Fifth Circuit directed the Court to reconsider two issues. First, the panel directed the Court to apply the proper burden and to reevaluate whether any of the four plaintiffs would have been admitted to the law school in the absence of admissions procedures which took into account an applicant’s race or ethnicity. Second, the Fifth Circuit instructed the Court to “revisit” the issue of damages in the event the law school fails to meet its burden: “In the event that the law school is unable to show (by a preponderance of the evidence) that a respective plaintiff would not have been admitted to the law school under a constitutional admissions system, the court is to award to that plaintiff any equitable and/or monetary relief it deems appropriate.” 8 Id. at 957.

In accordance with the Fifth Circuit’s instructions on remand, this case was tried before the Court on March 31 and April 2, 3, and 7,1997. Having carefully considered the evidence presented at both trials and the arguments and briefing of counsel, the Court finds the law school has proved by a preponderance of the evidence that none of the plaintiffs would have been admitted to the law school under a constitutional admissions system. In the event any of the plaintiffs successfully appeals that decision, the Court makes several alternative factual findings and legal conclusions regarding the issue of damages. Finally, the Court enters its findings of fact and conclusions of law regarding the attorneys’ fees to which the plaintiffs are entitled as prevailing parties under 42 U.S.C! § 1988 (West 1994).

II.

The admissions process employed by the law school in 1992 and the plaintiffs’ qualifications for admission are summarized here for the convenience of the reader and to provide a context for the opinions of the defendants’ expert on causation, Professor Olin Guy Wellborn, III. 9 In 1992, the law school received 4,494 applications for admission to fill approximately 500 available seats. 10 See Hopwood I, 861 F.Supp. at 563; Hopwood II, 78 F.3d at 935 n. 2. Given the rather large volume of applications it receives, the law school devised and for several decades has used an administrative procedure by which applicants are categorized according to their Texas Index (“TI”) score. An applicant’s TI score is calculated by the Law School Data Assembly Service (“LSDAS”). It is a composite number reflecting both the applicant’s grade point average (“GPA”) and score on the Law School Aptitude Test (“LSAT”), and it is generally considered a rough predictor of one’s probability of success in law school. See Hopwood I, 861 F.Supp. at 557 n. 9; Hopwood II, 78 F.3d at 935 & n. 1. In 1992, the law school placed each applicant, as dictated by his or her TI score, into one of the following three categories: presumptive admit, presumptive deny, and the discretionary zone. See Hop-wood I, 861 F.Supp. at 558; Hopwood II, 78 F.3d at 935.

The admissions committee in 1992 comprised nine professors, two assistant deans, and four students. Hopwood I, 861 F.Supp. at 560. Three members of the admissions committee, Professor Stanley Johanson and Assistant Deans Susana Alemán and Laquita Hamilton, formed a subcommittee to review the files of minority applicants. 11 See id. As chair of the admissions committee, Johnson had the responsibility of setting the presumptive admission and presumptive denial lines throughout the admissions process. 12 See id. He set the first of several presumptive admission lines in late January 1992 and began extending offers at that time to ensure that the most desirable applicants received offers as quickly as possible. See id. at 561. Because the law school application deadline was not until February 1,1992 — and thus the quality of the entire applicant pool had not yet been defined — Johanson set this initial presumptive admit line relatively high. 13 See id. at 560-61. Johanson then reviewed the files within the presumptive admission category to determine whether the applicant’s TI score was inflated by high grades in a noncompetitive university and/or major, or whether there was some other questionable aspect to the applicant’s file. See id. at 561. “Those applications with a high TI reflecting a high LSAT and high grades in a rigorous major at a leading undergraduate institution were admitted by Johanson, who had unilateral authority to admit any applicant in this category without further consultation with the full admissions committee.” Hopwood I, 861 F.Supp. at 561. Questionable files (roughly five to ten percent of all presumptive admit files) were placed in the discretionary zone for further review. See Hop-wood II, 78 F.3d at 936.

Essentially the same procedure occurred at the other end of the spectrum. Johanson set the initial presumptive denial line, and one or two members of the admissions committee then reviewed aE of the appheations in this category. See Hopwood I, 861 F.Supp. at 561. Apphcants with TI scores that understated their competitive standing relative to the rest of the pool were upgraded to the discretionary zone. See id. Johanson testified at the first trial that he could not recall the number of presumptive deny files that were upgraded in 1992, although generally twenty to forty files were moved to the discretionary zone as a result of this review. See id. The discretionary zone, therefore, comprised “those apphcants whose TIs feE between the presumptive denial Ene and the presumptive admission Ene, those applicants who Johanson had moved down from the presumptive admission category, and those apphcants who reviewers had moved up from the presumptive,denial category.” Id.

The law school’s admissions procedures treated minority applicants differently in two ways. First, the presumptive admissions and denial Enes were lower for minorities than they were for nonminorities. 14 By March 1992, the presumptive admission Ene for resident nonminorities was lowered from 202 to 199, whüé the presumptive denial line was set at 192. See id. at 561-62. For Mexiean-Americans, however, the presumptive admission line was lowered from 196 to 189, and for blacks this Ene was lowered from 192 to 189. For both minority groups, the presumptive denial line was 179. See id. at 562; Hopwood II, 78 F.3d at 936. Second, minority and nonminority files within their respective discretionary zones were segregated and subjected to different procedures of review. The minority subcommittee was to meet as a group to review and discuss all the minority files. Although the minority subcommittee provided the fuE admissions committee with summaries of the files they considered to be good candidates, the subcommittee’s admissions decisions were virtuaEy final. See Hopwood I, 861 F.Supp. at 562. Nonminority applications, on the other hand, were separated into weighted stacks of thirty files, and each stack w;as reviewed by a random three-member panel of the admissions committee. 15 See id. Rather than review the piles as a group, each member of the panel conducted an independent and secret screening of the stack of thirty files and voted to offer admission to an average of nine apphcants from the stack. 16 See id. Johanson then tallied the number of votes each applicant received within his or her stack of thirty. Applicants receiving two or three votes were offered admission, applicants receiving no votes were immediately denied admission, and applicants receiving one vote were offered a position on the waiting list. See id.

The plaintiffs were each reviewed by a three-member panel in the nonminority discretionary zone screening, and all were considered Texas residents by the admissions committee. Hopwood received an associate’s degree in accounting from Montgomery County Community College in 1984, and she earned a bachelor’s degree in accounting from California State University-Sacramento in 1988. She graduated with a GPA of 3.80 and had an LSAT score in the 83rd percentile. 17 See Hopwood I, 861 F.Supp. at 564; P-145. Although Hopwood had a TI score of 199 which, by March 1992, placed her just within the presumptive admit line, Johanson concluded her GPA overstated her educational background and therefore downgraded her file to the discretionary zone. See id. at 564. Carvell, Elliot and Rogers each had a TI score of 197. Carvell attended Hendrix College in Conway, Arkansas, where he graduated with a bachelor’s degree in political science in 1991. He had a GPA of 3.28 and an average LSAT score in the 76th percentile. 18 See id. at 566-67; P-151, Elliot graduated with a bachelor’s degree in accounting from the University of Texas in 1984, where he earned a GPA of 2.98, and he scored in the 95th percentile on the LSAT. Following graduation, he became a certified public accountant (“CPA”) and worked for various state agencies in Texas as an auditor or examiner. See id. at 565-66. Rogers earned an undergraduate degree in professional writing from the University of Houston-Downtown in 1990. In the early to mid-1980s, he attended the University of Texas, where he was placed on academic probation once and dismissed twice for poor scholastic performance. He had a cumulative GPA of 3.13 and an LSAT score in the 94th percentile. Rogers earned a master’s degree in professional writing from the University of Southern California in 1992. See Hopwood I, 861 F.Supp. at 567; P-171. By contrast, the median GPA for white students in the 1992 entering class was 3.56, and the median LSAT score was in the 91st percentile. See Hopwood II, 78 F.3d at 936-37 & n. 7. For all students (minority and nonminority) in the 1992 entering class, the median GPA was 3.52, while the median LSAT score was in the 89th percentile. See id.

Hopwood and Carvell each received one vote in screening, were offered positions on the waiting list, and were eventually denied admission to the law school in Spring 1992. See Hopwood I, 861 F.Supp. at 564-66. Elliot and Rogers received no votes in screening and were immediately notified of their rejection in April 1992. 19 See id. at 565-67. By the end of the admissions process, the law school had extended offers to 936 resident and nonresident applicants. See id. at 563 n. 32. Of the 637 offers extended to Texas residents, 96 went to blacks and Mexiean-Americans and 541 went to whites and nonpreferred minorities. See D-519. The plaintiffs were therefore among over 3,500 individuals, including approximately 1,500 Texas resi dents, who were denied admission to the law school in 1992.

III.

The Fifth Circuit characterizes Mt. Healthy as devising “a test of ‘causation’ that plaee[s] the burden of proving no harm on the defendant” when the plaintiff establishes the defendant intended to discriminate or otherwise' acted unconstitutionally. See Hopwood II, 78 F.3d at 956. The Court respectfully suggests that the Mt.- Healthy framework is inapplicable to this case and the plaintiffs should retain the burden of proof. In Mt. Healthy, an untenured teacher claimed a school board’s decision not to rehire him was the result of the comments he had made on a radio show. See Mt. Healthy, 429 U.S. at 281-3, 97 S.Ct. at 573-74. Although the board conceded the comments had influenced its decision not to rehire the teacher, the board asserted the teacher would not have been rehired in any event because of legitimate, nondiseriminatory reasons unrelated to the exercise of his First Amendment rights.

The district court found that the teacher’s comments were protected by the First Amendment and that they played a “substantial part” in the board’s decision not to rehire him. See id. at 283, 97 S.Ct. at 574. The district court ordered the teacher’s reinstatement, which ultimately resulted in his receiving tenure, and the circuit court affirmed the trial court’s findings and conclusions. See id. at 283-86, 97 S.Ct. at 574-75. Mt. Healthy was unique in that it presented the Supreme Court with a “mixed motives” case in which the governmental entity admitted it considered both legitimate and illegitimate factors in reaching its employment decision, in contrast to the typical “pretext” case in which the only issue is whether any improper or discriminatory motive exists. 20 Therefore, the question before the Supreme Court was whether the plaintiff had necessarily shown a constitutional violation justifying remedial action simply by establishing that the protected conduct was a “substantial” factor in the board’s decision not to rehire him. See id. at 285, 97 S.Ct. at 575.

The Supreme Court answered that question in the negative. The Supreme Court reasoned that allowing a plaintiff to recover under those circumstances might “place the employee in a better position as a result of the exercise of constitutionally protected conduct than he would have occupied had he done nothing,” because it would require holding the defendant liable even though the same employment decision would have been made for entirely legitimate reasons. Id. In response to this dilemma, the Supreme Court devised a burden-shifting framework in which the plaintiff initially bears the burden of proving that the constitutionally protected conduct was a “substantial” or “motivating” factor behind the employer’s discriminatory action. See id. at 287, 97 S.Ct. at 576; Village of Arlington Heights v. Metropolitan Housing Dev. Corp., 429 U.S. 252, 270 n. 21, 97 S.Ct. 555, 566 n. 21, 50 L.Ed.2d 450 (1977) (characterizing the plaintiff’s burden as a “required threshold showing”). Only after the plaintiff makes this showing does the burden then shift to the defendant to prove that the violation was “largely harmless.” See Hopwood II, 78 F.3d at 957. Therefore, unlike the McDonnell Douglas framework used in Title VII pretext cases (in which the burden of persuasion remains on the plaintiff at all times), the burden of persuasion shifts to the defendant in mixed-motives cases once it is either conceded by the defendant or determined by the fact finder that the employer considered impermissible factors in reaching its adverse employment decision. Mt. Healthy therefore confirms the rather uneontroversial proposition that plaintiffs are required to prove injury-in-fact in order to collect monetary damages. See Mt. Healthy, 429 U.S. at 285, 97 S.Ct. at 575.

Unlike the adverse employment action at issue in Mt. Healthy, there are two types of injury in a case involving the unlawful use of racial preferences. As the Court recognized in Hopwood I, 861 F.Supp. at 583, there is an intangible injury resulting from the government’s discriminatory classification which prevents a plaintiff from “competing on an equal footing” with other applicants. 21 See Hopwood II, 78 F.3d at 957 (quoting Adarand Constructors, Inc. v. Pena, 515 U.S. 200, 211, 115 S.Ct. 2097, 2105, 132 L.Ed.2d 158 (1995)). All nonminorities who applied to the law school in 1992, including nonminority students who were admitted, suffered that kind of injury. A second, tangible type of injury — an injury-in-fact— occurs when a plaintiff is actually denied some right or benefit, such as admission to the law school, as a direct result of the use of unlawful racial preferences. In applying the Mt. Healthy framework to this case, the Fifth Circuit affirmed this Court’s conclusion that the law school’s 1992 admissions procedures were constitutionally flawed. With virtually no explanation, however, the Fifth Circuit then asserted that, as a result of that determination alone, the defendants bear the burden of proving that the constitutional violation against the four plaintiffs was “largely harmless.” See Hopwood II, 78 F.3d at 957 (“In this case, there is no question that a constitutional violation occurred (as the district court found) and that the plaintiffs were harmed thereby.”). The Fifth Circuit’s analysis in this regard was incomplete at best; at worst, it was a misapplication of the Mt. Healthy framework because it presupposed that race was a substantial or motivating factor in every nonminority applicant’s denial of admission to the law school in 1992, regardless of the applicant’s qualifications to enter law school. As a matter of common sense and rudimentary mathematics, that cannot be the case.

There is no basis in fact or logic to suggest, as the Fifth Circuit apparently does in Hopwood II, that all resident nonminority applicants who were denied admission to the law school in 1992 were denied admission substantially — or, for that matter, even in small part — because of race. Even assuming all 96 offers of admission made to resident minorities in 1992 were available, there would still remain approximately 1,400 resident applicants in 1992 (the overwhelming majority of whom, presumably, were nonminorities) who would have been denied admission without regard to race. At most, therefore, only 7% of resident nonminority applicants were affected by the law school’s use of racial preferences. 22 Suppose this case had been brought by the least qualified nonminority who. applied to the law school in 1992. Under Adarand, that applicant clearly would have standing to challenge the affirmative action program. See Adarand, 515 U.S. at 212, 115 S.Ct. at 2105. However, an applicant who has no conceivable chance of admission cannot possibly show that race was a substantial or motivating factor in the law school’s decision to deny him or her admission. The Mt. Healthy burden-shifting framework is never implicated. In contrast, Hopwood II always shifts the burden to the defendant — and would shift the burden even in the example of the candidate with no chance of admission — once the plaintiff establishes the unconstitutionality of the affirmative action program. But being subjected to admissions procedures that take into account racial preferences does not, in and of itself, establish or even imply that the applicant’s race was a substantial or motivating factor in his or her denial. Importantly, in this case the law school, unlike the University of California in Bakke, never conceded that the plaintiffs would have been admitted in the absence of an affirmative action program, and this Court specifically found the plaintiffs failed to carry their burden of persuasion on that point. See Hopwood I, 861 F.Supp. at 581-82.

The harm in the Fifth Circuit’s interpretation of Mt. Healthy is not readily apparent in the Court’s example because it would be quite easy for the law school in that instance to establish the applicant’s non-admission. The harm occurs, for example, when the suit is brought by a candidate who has some chance of admission under a race-neutral procedure but who would not be able to prove that he or she would have been admitted by a preponderance of the evidence. The Fifth Circuit decision potentially gives those candidates a windfall by placing the ultimate burden of proof on the defendant: close calls must always be decided in favor of the plaintiff. And yet the function and purpose of the Mt. Healthy burden-shifting scheme are just the opposite — to ferret out instances of discrimination and to ensure that the plaintiff is not put in a better position than he or she ordinarily would have occupied. Indeed, the Fifth Circuit recognized in Hopwood II that Mt. Healthy was intended to give the defendant “a second chance of prevailing by showing that the violation was largely harmless.” See Hopwood II, 78 F.3d at 957. Instead of giving the defendant a second chance of prevailing, the Fifth Circuit gives the defendant little chance of prevailing when the applicant is an extremely close call. The Fifth Circuit decision essentially allows a plaintiff to recover monetary damages without ever having to establish any injury-in-faet. 23

Mt. Healthy does not apply to this ease. One might be able to make. a normative argument that the defendants should bear the ultimate burden of persuasion once the plaintiffs establish the unconstitutionality of the law school’s affirmative action program, but the legal argument cannot be justified. This is a classic pretext ease in which the trier of fact — in this case, the Court — must determine why each of the four plaintiffs was denied admission to the law school. If the plaintiffs were denied admission because of their race, they should have to prove it, as plaintiffs must in every other pretext case brought in federal and state court. The Court therefore respectfully but strenuously objects to this portion of the Fifth Circuit opinion. The Court is cognizant, however, of the panel’s instructions on remand, and it will faithfully and responsibly execute them.

A.

The defendants presented one expert on causation, Olin Guy Wellborn, a tenured professor who has taught at the law school since 1974. 24 Wellborn has served on the law school admissions committee for over 15 years and was a member of the admissions committee in 1992. In accordance with the Fifth Circuit decision, Wellborn analyzed whether the plaintiffs would have been admitted to the law school under a constitutional admissions system. See Hopwood II, 78 F.3d at 957 & n. 55. In preparation for his expert reports, Wellborn testified he carefully examined the LSDAS sheets of approximately 450 applicants, including the LSDAS sheets of the 96 minorities admitted in 1992. Wellborn, vol. 1 at 45. He then narrowed the field to approximately 200 of the “most promising” minority and nonminority candidates, as reflected by their LSDAS sheets, and he reviewed those application files in their entirety. Id. Wellborn prepared two expert reports, employing a different methodology in each one. See D-519 and D-520. In both reports he concluded that none of the plaintiffs would have been admitted in a race-blind admissions procedure. See id.

(1) Wellborn’s First Report

In his first report, Wellborn examined the 1992 admissions process and then considered how that process would have been modified in a manner that would yield the same total number of admissions on a race-blind basis. Wellborn first compared the relation between an applicant’s TI score and the statistical rate of admittance in the actual 1992 admissions process. Wellborn detected the emergence of a pattern in which candidates with TI scores of 203 or higher were virtually all admitted and candidates with TI scores of 184 or lower were virtually all denied. A discretionary zone candidate’s statistical chance of admittance generally increased by an additional 10 or 20 percentage points with each successive TI score. 25 Hopwood’s TI score of 199 corresponded to an acceptance rate of 89%, and a TI score of 197 (the TI score belonging to Carvell, Elliot, and Rogers) corresponded to an acceptance rate of 59%. This pattern of percentages reflected a typical, yearly phenomenon. 26 See Wellborn, vol. 1 at 145.

Because the actual 1992 presumptive admission and denial lines were lower for minorities than for nonminorities, Wellborn had to adjust the overall presumptive admission and denial lines in creating a hypothetical admissions system. 27 Wellborn therefore lowered the overall presumptive admit and deny lines to 198 and 190, respectively, and retained the 1992 average of nine votes per pile of thirty files. Wellborn assumed the law school would extend the same number of resident offers that it had extended in 1992. 28 See D-520; Wellborn, vol. 1 at 237. Using this methodology, Wellborn predicted a pattern of percentages in admissions almost identical to the 1992 admissions percentages, except that each percentage of admission would be “notched down” by one TI score throughout. In other words, Wellborn predicted that virtually all candidates with a TI score of 202 and above would have been admitted; 90% of applicants with TI scores ranging from 198 to 201 would have been admitted; 70% of applicants with TI scores of 197 would have been admitted, and so on down the Texas Index. 29 Wellborn then re examined the application files of the four plaintiffs, the admitted minority resident applicants, and the denied nonminority resident applicants at their respective TI scores to determine who would have received the additional offers as dictated by the predetermined percentages of admission.

Wellborn predicted that one additional offer of admission would be made at Hop-wood’s TI score. Of the ten admitted minority and denied nonminority resident applicants at the 199 TI level, Wellborn concluded that seven applicants, including the four minority applicants, were “clearly stronger” than Hopwood and that the remaining two applicants were “comparable” to Hopwood. As in the actual admissions process in 1992, Wellborn determined that Hopwood would most likely have been placed in the discretionary zone by Professor Johanson and that she would not have received the two or three votes necessary to be admitted. At the 197 TI score of Carvell, Elliot, and Rogers, Well-born estimated that the law school would have extended seven additional offers. 30 Wellborn identified the seven applicants he predicted would have been offered admission over Carvell, Elliot, and Rogers, and he also rated those seven applicants superior to Hopwood. Wellborn then identified four additional applicants who, although not as strong as the original seven, were still stronger than Carvell, Elliot, and Rogers. Indeed, Wellborn indicated that Rogers and Elliot were “among the very weakest” at that index and that Carvell, although a stronger applicant than Elliot and Rogers, does not compare favorably because of his relatively low combined LSAT score in the 76th percentile. 31 See D-519.

(2) Wellborn’s Supplemental Report

Using his first report as a hypothesis for how the law school would have extended offers of admission in a race-blind procedure, Wellborn undertook the more difficult task in his supplemental report of specifically identifying (1) the admitted resident minorities in 1992 who probably would have been denied admission in a race-blind procedure, and (2) the denied resident nonminority applicants who most likely would have been admitted in their place. Wellborn made several basic assumptions about a hypothetical race-blind admissions procedure. First, as he had in his first report, Wellborn projected that the presumptive admission line would be set at 198 and that the presumptive denial line would be set at 190. He then assumed that all of the nonminority residents who had been admitted in 1992 would have been admitted in a hypothetical race-blind admissions procedure. Therefore, Wellborn only examined the application files of the 96 admitted minority applicants and the approximately 450 LSDAS reports (which he then narrowed down to about 200 application files) of the denied residents (minority and nonminority) who had TI scores above 190. 32 See D-520; Wellborn, vol. 1 at 45. In examining each application, Wellborn primarily considered the applicant’s college record and LSAT score, which he weighed “about equally,” 33 and he applied identical standards to minority and nonminority candidates. See D-520. In evaluating the college record, he considered the caliber of the school, as manifested by the LSAT college mean, 34 and the applicant’s major, rank in class, and college transcript. Wellborn also considered an applicant’s personal statement, letters of recommendation, and other factors and materials such as the applicant’s age and background; the relevance and importance of these factors varied, of course, with each file.

In selecting a particular applicant for his supplemental report, Wellborn attempted to predict those candidates who he believed would emerge as probable admittees. He did not, in other words, only select candidates for whom he necessarily would have voted, nor did he attempt to predict how a particular- admissions committee member would have voted. See Wellborn, vol. 1 at 231-33, 249-51. Wellborn explained his approach as follows:

[I]f I say these candidates are the ones that I judge most likely to be admitted, and these I judge to be relatively unlikely to be admitted, that’s not going to describe in exact detail the offers that go out. I’m making an overall estimate of probabilities, based upon many years of experience in watching this process and having a judgment about on the average which criteria are going to be most influential and most important to the reviewers as a group. And so, of course, there are going to be those vagaries in the voting, but I still believe in the overall validity of the prediction.

It certainly is not the case — absolutely is not the case that ... [all] denied non-minorities [within the discretionary zone] ... have an equal chance [of admission] ... or that a person, as familiar as I am with the process, couldn’t tell you that these people have a better chance than these other people. I think ... it would be preposterous to suggest that ... it’s just random, or that it’s completely unpredictable.

Wellborn, vol. 1 at 250.

The law school made 96 offers of admission to resident minorities in 1992. Wellborn evaluated all 96 minority admittees and concluded that 19 would have been admitted without regard to race. Wellborn named this group of applicants “Group A.” Wellborn then selected 2 additional resident minority admittees who, though weaker than the original 19, would still have a better chance of admission than any of the plaintiffs in a color-blind procedure (“Group B”). Subtracting the Group A applicants from the actual number of resident minority admittees in 1992 leaves 77 seats remaining for nonminority resident applicants who were denied in 1992. There were 398 nonminority residents with TI scores of 190 or above who were denied admission in 1992; therefore, fewer than 20% of candidates in Wellborn’s hypothetical discretionary zone would have been offered admission. Of those 398 files, Well-born selected 78 resident nonminority applicants 35 who he believed probably would have been offered admission in a race-blind procedure (“Group C”). As with the minority applicants reflected in Group B, Wellborn also selected an additional 20 resident nonminority applicants who he believed had a reasonable chance of admission as alternatives to the Group C candidates (“Group D”). 36 Wellborn testified that none of the plaintiffs merited inclusion in either Group C or D. See Wellborn, vol. 1 at 115.

The plaintiffs have several objections to the methodology Wellborn used in his supplemental report. 37 First, the plaintiffs argue the random admissions procedure employed in 1992 effectively prevents the defendants from carrying their burden of proving that the plaintiffs would not have received offers of admission in a race-neutral procedure. Second, Hopwood and Carvell argue that, in contravention of the Fifth Circuit decision, Wellborn made no attempt to reconstruct the actual admissions process used in 1992 to determine which nonminority residents would have been offered admission under a constitutional system, nor did he devise a methodology revolving around either the piles of thirty or the waiting list actually generated in 1992. Third, the plaintiffs criticize Wellborn for making the assumption that all nonminority residents who were given offers of admission in 1992 would nevertheless have been admitted under a constitutional system. Fourth, Hopwood and Carvell argue that Wellborn’s testimony is insufficient to carry the defendants’ burden as a matter of law because Well-born relied on “mere statistical likelihoods” in concluding that the four plaintiffs would have been denied admission under a race-neutral system. The Court considers each objection seriatim.

As this Court recognized in Hopwood I, reconstructing the actual 1992 admissions procedure is a conceptually difficult, if not “virtually impossible,” task: “[T]he difficulty does not stem from the unconstitutional aspects of the procedure alone but from the random shuffle of files into stacks of thirty, each stack reviewed by different subcommittees of three.” 38 Hopwood I, 861 F.Supp. at 582 n. 86. Although the stacks were weighted as closely as possible with the same number of high to low TI scores, see Wellborn, vol. 1 at 41, an applicant’s chances for admission may in some instances have depended upon the personal tastes of the reviewers who evaluated the applicants file and whether the applicant was placed in a stack with an unusually strong or weak pool of applicants. To the extent that the plaintiffs argue that the admissions procedure used in 1992 was' inherently unfair due to the random composition of the piles of thirty and the subjectivity of each reviewer, the point is very well taken. It goes without saying, however, that this unfairness does not implicate a constitutional defect with the law school’s admissions procedures. To the extent that the plaintiffs argue that the law school can never meet its burden for that reason, the point is not well taken. A particular applicant’s admission to the law school in 1992 was not the result of a random and unpredictable process — after all, applicants were not admitted by a flip of a coin. And it is nonsense to suggest that all nonminorities who were denied admission in 1992 would have an equal chance of admission under a race-blind procedure. With or without the law school’s ill-advised piles-of-thirty approach, even a superficial comparison of the four plaintiffs; for instance, reveals the varying quality of students who applied to the law school in 1992 and confirms that concrete distinctions can be made among those applicants — if not to a virtual certainty, then at least by a preponderance of the evidence.

The plaintiffs’ argument that Well-born should have reproduced an admissions procedure by which application files were randomly placed into piles of thirty and then reviewed by randomly formed screening committees is also without merit. Wellborn is a very talented law professor, but he cannot turn back the hands of time, nor does the Fifth Circuit decision require him to do so. Rather, the inquiry posed by the Fifth Circuit is whether any of the four plaintiffs would have been admitted under '“a constitutional admissions system.” See Hopwood II, 78 F.3d at 957 (emphasis added). The Court, therefore, does not read the Fifth Circuit opinion as instructing the defendants to replicate exactly the 1992 admissions procedure or to use the 1992 piles-of-thirty approach. First, how would one go about reproducing a random process? By the,very definition of the word random, it cannot be done. The only logical, reasonable, and fair way to determine who would have received offers of admission under a constitutional system is to compare the application files of all of the admitted minorities and the denied nonminorities, and that is precisely the methodology Wellborn used in his supplemental report. 39 Furthermore, using the 1992 piles of thirty would perpetuate the unconstitutional aspects of the 1992 admissions procedure. The admissions procedure employed in 1992 was fraught with constitutional error precisely because the original piles of thirty did not include any minority applicants. Those piles would never exist under a constitutional admissions system, and as a result, it would make no sense to use them now to determine whether any of the plaintiffs would have gained admission to the law school in 1992. The Court finds it more than a little baffling that the plaintiffs assert Wellborn should have utilized, as the very premise of his methodology, the admissions system found unconstitutional as a result of this lawsuit.

Likewise, using the 1992 waiting list to determine who would have been admitted under a constitutional system is equally problematic. In support of this method, Hop-wood and Carvell argue that the one vote they actually received in 1992 placing them on the waiting list is more probative than any subsequent attempt to reconstruct a hypothetical race-blind admissions process. While this method certainly has some intuitive appeal, it is not the best way to determine which nonminority applicants would have gained admission in 1992 in a race-blind procedure. First, there were two waiting lists in 1992, each of which was segregated by race. See Hopwood I, 861 F.Supp. at 574 n. 68. Second, this methodology suffers from the same defect identified with respect to the original piles of thirty: the waiting list was generated by an unconstitutional system in which minority and nonminority applicants were reviewed separately. Therefore, the waiting list created in 1992 would not exist under a constitutional admissions system. Third, due to the change in the LSAT scoring system, 1992 was somewhat of an anomaly in that the law school had not extended a sufficient number of offers through its normal procedures. As a result, the admissions committee had to select an unusually high number of candidates — 42 out of 123 applicants — from the waiting list. 40 See Wellborn, vol. 1 at 52-53, 144. Wellborn testified that the waiting list had been pretty well “picked over” and, consequently, the remaining eighty-one applicants on the waiting list had, in a sense, a third “no” vote denying them admission. 41 See Wellborn, vol. 1 at 52-54. Fourth, using the waiting list to determine who would have gained admission is, in the Court’s opinion, self-defeating for all of the plaintiffs: the methodology would unfairly exclude Elliot and Rogers from consideration because they received no votes in screening, and the one vote Carvell and Hopwood each received is not necessarily probative of any increased likelihood that they would have been admitted in a race-blind procedure. Carvell’s vote was from a student member of the admissions committee, not a faculty member, see Hopwood I, 861 F.Supp. at 566, and Hopwood’s vote was essentially a “sympathy vote” from Hamilton, then the Assistant Dean of Admissions for the law school. 42

The plaintiffs also argue that Wellborn arbitrarily and impermissibly “picks and chooses” among actual events in 1992 in that he (1) gave no weight to applicants who were placed on the waiting list by virtue of their one vote and (2) assumed that all nonminority applicants who were admitted in 1992 would have been admitted in a race-blind procedure. The plaintiffs urge the Court to disregard Wellborn’s supplemental report entirely because, they argue, he did not reconsider all offers of admission in hypothesizing a constitutional admissions system. Not only is this argument unpersuasive, it mischaraeterizes Wellborn’s methodology. The plaintiffs acknowledge it was more difficult for nonminorities to achieve offers of admission in 1992; indeed, the plaintiffs brought this lawsuit for that very reason. The fact that those nonminority applicants received offers of admission under the University’s admissions program is very strong evidence to support the inference that they would emerge as probable admittees under a race-neutral system. See Wellborn, vol. 1 at 42. Wellborn did not, therefore, arbitrarily pick and choose among real events in 1992; rather, he drew logical inferences based upon the admissions procedure in force in 1992. The fact that an applicant was presumptively admitted, received two or three votes in the screening process, or was an applicant chosen from the waiting list in 1992 correlates with a strong likelihood that the applicant would have been admitted in a race-blind admissions procedure; for the reasons indicated above, however, there is not a sufficiently strong correlation between receiving a position on the waiting list and an increased probability of admittance. 43 Therefore, the distinction between nonminority admittees and waiting list candidates is warranted. Furthermore, it is conceptually difficult to see the harm in the inference — to the extent that these individuals were- offered admission over the four plaintiffs, it certainly was not the result of any unlawful or invidious discrimination.

Finally, the Court rejects the plaintiffs’ argument that Wellborn relied on “mere statistical likelihoods” in concluding that the four plaintiffs would have been denied admission. In support of this argument, the plaintiffs cite Smith v. Rapid Transit, 317 Mass. 469, 470, 58 N.E.2d 754, 755 (1945), a case in which the Massachusetts Supreme Court held it is insufficient as a matter of law to prove a proposition merely by showing the mathematical, chances favor the veracity of the proposition. This argument might have been valid had Wellborn conelusorily asserted, for instance, that Carvell would not have been admitted to the law school based solely on the fact that his “true” TI score of 191 or 192 correlated to a 5% to 15% probability of admission in 1992. But Wellborn did not use mathematical probabilities to prove Carvell would not have been admitted to the law school under a constitutional system. Instead, Wellborn evaluated Carvell’s application according to certain enumerated and, to a large extent, quantifiable criteria, compared his application to the applications of 450 other denied nonminorities, and concluded that Carvell was less qualified than the 119 applicants chosen for. inclusion in his supplemental report. See D-520. Based on the applications in evidence, as well as his knowledge of the law school’s admissions procedures, it was Wellborn’s considered opinion that it was more likely than not that Carvell would not have been admitted to the law school under a race-neutral admissions system. Indeed, Wellborn reached the same conclusion with respect to the other three plaintiffs despite the fact that their relatively high TI scores correlated with a probability of admittance of greater than 50%. This objection is nugatory.

The Court therefore concludes that the methodology Wellborn employed in his supplemental report is sound. Wellborn did not attempt to handicap the chances of admission of each and every applicant who was denied admission to the law school in 1992. Rather, Wellborn presented evidence of 119 applicants who were better candidates for the study of law than the plaintiffs and who were therefore more likely to receive offers of admission under a constitutional system. In hypothesizing a constitutional admissions system, Wellborn made reasonable projections regarding the lowering of the presumptive admission and denial lines and the probable number of offers to be extended. His assumption that all nonminority- resident admittees in 1992 would be admitted under a race-neutral system is not only reasonable but supported by the record. Finally, Well-born identified reasonable criteria to evaluate the application files, and as discussed below, he applied the criteria in a fair, consistent, and nondiscriminatory way.

B.

Before Wellborn’s conclusions are explored and explained, the Court offers the following disclaimer. This opinion provides more information than anyone ever needed or wanted to know about the qualifications required for admission to the law school. The Court once again undertook a painstaking review of the application files of hundreds of students, and, unfortunately, this opinion reflects the tedious and arduous process that it was. One principle the Court emphasized in its first opinion bears repeating here:

[T]he Court appreciates the difficulty of the task facing the admissions committee each year. Evaluation of applications involves both objective and subjective factors, and the Court is aware that some evaluators could use subjectivity to conceal discriminatory motives. As a general rule, however, judges are not as well suited to evaluate qualifications of applications as those who are familiar with the process and have many years of experience evaluating applications.

See Hopwood I, 861 F.Supp. at 581 (citing Odom v. Frank, 3 F.3d 839, 847 (5th Cir. 1993)).

The plaintiffs openly accused Well-born of manipulating certain criteria and of disingenuously using the. subjectivity of the admissions process to rig the desired outcome of this case. The Court emphatically finds that allegation to be false. The fact that subjective criteria necessarily informed Wellborn’s analysis does not render Well-born’s opinions arbitrary or untenable. Part of Wellborn’s everyday responsibilities as a law professor is to make subjective assessments of students, whether on a law school exam or as a member of the admissions committee, and the Court will not cavalierly disregard the conclusions Wellborn reached in his supplemental report simply because they involve subjective judgments. Furthermore, Wellborn credibly supported the applicants he selected, and the faculty members who originally reviewed the plaintiffs’ files in 1992 universally support Wellborn’s judgment that the four plaintiffs were weak candidates for admission to the law school in 1992. See Johanson Declaration D-332; Declaration of Laquita Hamilton D-333; Sharlot Declaration D-334; Declaration of Mark Gergen D-335; Goode Declaration D~ 336.

Before delving into the specific weaknesses of each plaintiff’s application, the Court makes the following observations and generalizations. First, TI scores are, to a certain degree, inherently unreliable because they - do not weigh the student’s GPA by major or quality of school. For example, Hopwood’s GPA of 3.80, with her high number of hours from community colleges and subsequent transfer to a relatively weak undergraduate institution, does not compare favorably with a student having a lower GPA from, say, Rice, Trinity, the University of Texas, or Texas A & M. Wellborn’s evaluation of Carvell’s chances for admission also illustrates the artificial nature of the Texas Index. Although Carvell’s “true” TI score was 191 or 192, Wellborn rated Carvell “roughly comparable” to Hopwood, “slightly better” than Elliot, and “significantly better” than Rogers. See Wellborn, vol. 1 at 117. Large institutions such as the University of Texas use TI scores to make general distinctions among the many students who apply. But a TI score is only as good as its two components — the student’s GPA and LSAT score. They are not hard-and-fast numbers determining who is admitted, nor should they be treated as such, because to do so would unfairly reward students who attended weak undergraduate institutions or who took less-than-ehallenging curricula. In light of the fact that TI scores are not weighted by quality of school or difficulty of major, it is not surprising that the plaintiffs’ relatively high TI scores significantly overstate their chances for admission to the law school and that Wellborn’s selections ran the gamut from high to low TIs.

Second, the law school admissions process is incredibly competitive. See Hopwood II, 78 F.3d at 935. In fact, the Court was astonished to discover the number of applicants with drastically better credentials than the four plaintiffs who were denied admission in 1992. For instance, of the 78 nonminority Group C applicants Wellborn selected, 31 were in the top quarter of their classes at competitive universities such as the University of Texas and Texas A & M and had LSAT scores in the 80th percentile and above. See D-543, 550-51, 555, 557, 560-63, 565, 568, 570, 572-73, 570-77, 580, 582, 584-85, 587-89, 591-93, 599, 609, 611, 613, and 619. Over one third of Wellborn’s Group C selections had a college class rank in the 80th percentile or higher; unlike Hopwood, however, these applicants attended competitive schools and, except for one applicant, had few, if any, hours from a community or junior college. See D-543, 551, 556, 558-60, 562, 564-68, 572, 574-76, 578-84, 609, and 611. Applicants similar to Carvell who had a class rank below the 70th percentile virtually all came from colleges with an LSAT mean of 34 or better such as Duke, Cornell, Rice, Tufts, and the University of Texas, and their LSAT scores in all eases were better — and in most eases were considerably better — than Carvell’s combined LSAT score in the 76th percentile. See D-542, 544-45, 547-49, 552, 554, 569, 594-98, 602-03, 614-15, and 618. Moreover, of the 97 Group A and C applicants Wellborn predicted would have been probable admit-tees under a constitutional system, only 2 graduated in the bottom half of their classes, as did Elliot. One applicant earned an LSAT score in the 99th percentile and graduated in the 31st percentile of the class at the University of California at Berkeley, a 36 LSAT mean college. See D-524. The other applicant attended the University of Texas, majored in a rigorous liberal arts honors program called Plan II, was in the 48th percentile of the class, and earned an LSAT score in the 96th percentile. 44 See D-529. Well-born did not select a single applicant who, like Rogers, had been dismissed from a four-year university or who had graduated from a university with an LSAT college mean as low as that of the University of Houston-Downtown.

Third, the applicants Wellborn selected for inclusion in Groups A, B, C, and D have superior credentials to the four plaintiffs even when one looks exclusively at a combination of objective factors. After making his 119 selections, Wellborn discovered that his selections tended to satisfy the following four criteria: each applicant (1) had an LSAT score in the 80th percentile or higher; (2) had a rank in class in his or her university of at least the 60th percentile; (3) graduated from a college with an LSAT mean of at least thirty; and (4) had no more than one year in a community or junior college. None of the plaintiffs satisfies all four criteria, while the vast majority of Wellborn’s selections do. Moreover, applicants who did not comply with all four criteria counterbalanced that weakness with a specific strength. For instance, five applicants in Group C have LSAT scores in the 78th percentile, and three applicants in Group A and one applicant in Group D have LSAT scores in the 74th percentile. Across the board, however, these applicants had exceptionally strong college records at high quality institutions. 45 Likewise, applicants with low class ranks offset that weakness with strong LSAT scores coupled with very competitive universities and/or particularly rigorous majors such as engineering. 46

Fourth, the inherently subjective nature of the admissions process does not, as the plaintiffs contend, provide a reason to believe that distinctions cannot be made among different applicants. Of course, reasonable minds can and do differ on the quality of particular institutions or courses of study. For instance, whether accounting is a more rigorous major than political science, economics, or English is, to the Court’s chagrin, an issue of raging debate among the parties in this case. In the Court’s opinion, all four courses of study provide a good background for the study of law, and the Court, like Wellborn, tended to focus more on the overall quality of the applicant’s undergraduate institution and curriculum in evaluating the application files. And, the plaintiffs protestations to the contrary, Wellborn’s opinion regarding the quality of certain universities is not entirely subjective. He supported his assessment of each institution with an objective criterion, the LSAT college mean. 47 Cf. P-414 (discretionary zone screening instructions stating that “[t]he LSAT college mean should be used as a general indicator of the strength of the undergraduate institutions’ student body” and indicating that reviewers should “ Svatch out’ for inflated GPA’s due to community or junior college grades”). Even without such quantifiable data supporting his assessments, the Court is confident Wellborn can fairly evaluate the quality of colleges and universities around the nation given his twenty-three years as a law professor and fifteen years on the admissions committee. In addition, Hop-wood, Carvell, and Eogers do not appear to object to Wellborn’s use of the LSAT college mean as a legitimate means of distinguishing among universities or, significantly, to the resulting inference Wellborn draws that the applicant’s GPA is inflated. Instead, they argue that Wellborn exaggerates the importance of this criterion. The Court finds that argument unpersuasive. One of Wellborn’s constant themes throughout his testimony was the notion that the applicant be exposed to a “rigorous testing ground” to prepare him or her for the study of law at an elite law school. See, e.g., Wellborn, vol. 1 at 84. The evidence establishes an applicant’s undergraduate institution can be of paramount importance in determining admission, particularly when the student attended a very weak school.

On the other hand, Hopwood and Carvell also criticize Wellborn for failing to give greater weight to or for unevenly evaluating subjective and nonacademic factors such as post-graduate work experience, good personal statements, and improvement in grades over time. 48 First, some of these factors, such as improvement in grades and working through school, áre “very common,” see Well-born, vol. 1 at 68, and/or are-generally not given much weight, see Goode Declaration, D-336. 49 Second, although these idnds of sub jeetive factors tended to support Wellborn’s assessment of a particular applicant, see Wellborn, vol. 1 at 249, they did not appear to impact Wellborn’s selections significantly except in one unusual case. 50 See D-538. Indeed, Wellborn testified that, with respect to marginal candidates in particular, members of the admissions committee tend to focus primarily on an applicant’s balanced and consistent college performance and LSAT:

In general, I think we are talking here about the margin____The unimpeachable candidates have already been admitted. I’m not asserting that these files I’ve selected are without fault or without weaknesses. It’s a matter of relative weaknesses —

[I]t’s a question, I think, of minimizing risk that over my years on the committee what I see members doing at this margin is looking for belts and suspenders ---- That’s why I put the emphasis on some kind of balance here with the LSAT and the college score. Those are the two biggest things____[0]ur feeling is that especially in this middle zone where the LSAT and the college record kind of balance one another, they reinforce one another, it’s a little more reassuring, and reassurance is what you are looking for. You’re not going to get excited about these margin candidates.

Wellborn, vol. 1 at 87-88.

Part of the plaintiffs’ strategy throughout the remand trial has been to find the files of a small handful of 1992 admittees who the law school faculty members opined were as qualified, or in a few instances, perhaps less qualified for admission than the plaintiffs. 51 This evidence does not persuade the Court to disregard the conclusions Well-born reached in his expert report or the assessments made by faculty members who originally reviewed the plaintiffs’ files in 1992. The fact that a small number of admittees 52 had credentials similar to or worse than the plaintiffs does not lead one to the conclusion that the plaintiffs would have been admitted in a race-neutral process. At most, this evidence proves that the plaintiffs had some chance of admission, however slim, a proposition the law school has conceded all along. 53 See Wellborn, vol. 1 at 115. But having “some” chance of admission does not correlate to having a reasonable chance, which is the inquiry required by the burden of proof in this case. Moreover, the plaintiffs’ argument rests on a fallacious premise. The plaintiffs have identified the lowest common denominator and, after determining that the plaintiffs meet or exceed that, .standard, have concluded that the plaintiffs would therefore have a reasonable chance of admission (or, conversely, that the defendants cannot prove that the plaintiffs do not have a reasonable chance of admission). The comparison to be drawn, however, is not among individual applicants but among the entire applicant pool. It simply does not follow that anyone on a par with the least qualified admittee would have a reasonable chance of admission, and it certainly does not reflect the way in which the law school selects its entering classes. Frankly, the Court draws quite the opposite inference — the fact that so few applicants comparable to the plaintiffs were actually admitted in 1992 is evidence that the plaintiffs probably would not'have been offered admission in a constitutional process.

Finally, each of the plaintiffs has one or more significant weaknesses in his or her application that are not counterbalanced by a specific strength. Hopwood earned 70 of her undergraduate hours at community colleges. See Johanson Declaration D-332 (indicating that community colleges tend to be “noncompetitive institutions with academically weak student bodies and faculties”); Declaration of Elizabeth Chambliss, D-338 (indicating that less than 1% of resident nonminorities admitted in 1992 had more than 70 hours of undergraduate education at a community college or junior college). Hopwood required 6 years to complete her associate’s degree, resulting in an average of 10 credit hours per year. See P-145 (LSDAS record); 1994 trial transcript, Johanson, vol. 5 at 15 (stating that Hopwood earned her GPA on “a fairly slow track”). She subsequently obtained a bachelor’s degree from a university that is uncompetitive with the universities attended by the vast majority of students who are admitted to the law school. See Chambliss Declaration, D-338 (indicating that less than 2% of resident nonminorities admitted in 1992 graduated from schools with college LSAT mean scores of 28 or below).

Wellborn testified that Hopwood’s LSAT score was acceptable for a marginal candidate. See Wellborn, vol. 1 at 67. Likewise, Hopwood’s non-academic strengths (her age, the, fact that she was the mother of a handicapped child, had worked her way through college, and later became a CPA) merit some “preferential consideration.” See Sharlot Declaration D-334. However, according to Wellborn and the admissions committee members who reviewed Hopwood’s application in 1992, these positive attributes ultimately do not outweigh the negative aspects of her application, the most important factor of which is the weakness of her undergraduate education. See Wellborn, vol. 1 at 67-68; Johanson Declaration D-332 (“Her transcript shows not only that she attended weak schools, but that the bulk of her education was in technical ‘how-to’ courses rather than academic courses requiring analytic skills l ... Hopwood is simply not well-prepared academically for Law School.”); Hamitton Declaration D-333 (stating that Hop-wood’s undergraduate record “reflects little analytic preparation for law school”); Sharlot Declaration (concluding that Hopwood’s file “is very weak in comparison with the overwhelming credentials of so many of our applicants”). 54

As the Court noted in Hopwood I, there is little else to Hopwood’s file. She provided no letters of recommendation, no personal statement, and the handwritten application is among the least impressive in appearance of all of the files examined by the Court. 55 See Hopwood I, 861 F.Supp. at 581; Johanson Declaration D-332 (“It is also striking to see no letters of recommendation, particularly for an older applicant who has been out of school a while.”). Wellborn summed up her application as follows:

So the overall picture here I think is one that would raise a lot of concern because we don’t have a candidate who has been tested at a highly competitive college environment as many, many of our applicants have. A lot of it is kind of like high school, frankly, community college, and so forth, and we — of course, we all know from our own personal experience of the high school honor student who goes to the major state university and doesn’t fair [sic ] very well. It’s a lot tougher. And, of course, we feel that the University of Texas Law School is a lot tougher still. So that would be a concern here.

Wellborn, vol. 1 at 68.

Carvell’s “true” TI score of 191 or 192 placed him with the group of applicants who were presumptively denied admission in 1992. Only approximately 5% of candidates with TI scores of 191 and 15% of candidates with TI scores of 192 were admitted to the law school in 1992; therefore, it is not surprising that Carvell was denied admission, as he was among an overwhelming majority of applicants with the same TI score who were not admitted. Although Carvell’s personal statement was “well-written,” see Wellborn, vol. 1 at 127, Wellborn testified that it did not compensate for the overall weakness of his application. Both Carvell’s GPA of 3.28 and LSAT score in the 76th percentile were significantly below the 1992 overall GPA median of 3.52 and overall LSAT median in the 89th percentile. See Hopwood I, 861 F.Supp. at 563 n. 32; Hopwood II, 78 F.3d at 937 n. 7; see also Sharlot Declaration D-334 (“The most striking feature [of his file] is his undergraduate GPA at a school with an extremely low mean LSAT score----lam also struck by the fact that this relatively unimpressive performance was by a graduate of what is reputedly one of the best public high schools in the state [Highland Park High School in Dallas, Texas].”); Gergen Declaration D-335 (characterizing Carvell as “an easy no vote” and indicating that his GPA was weak “even at a school that I don’t suspect is very demanding”); Goode Declaration D-336 (“Mr. Carvell’s transcript reveals a mediocre academic record from a mediocre school____If Mr. Carvell could not finish in the top two-fifths of his college class, what chance is there that he would perform well in a much more competitive environment?”).

Carvell provided two faculty and two professional letters of recommendation, which one of his 1992 faculty reviewers described as “strikingly weak.” See Goode Declaration D-336. At the remand trial, Wellborn described one of the faculty letters as “not very strong” and “vague.” See Wellborn, vol. 1 at 74. The other faculty letter by all accounts is downright negative, characterizing Car-veil’s classroom performance as disappointing, mediocre, and uneven. See P-151; Ger-gen Declaration D-335 (“I also cannot help but be impressed by a rare honest letter of recommendation intimating that Mr. Carvell is either lazy or unfocused.”); Goode Declaration D-336 (“It is a letter that first damns with faint praise and then damns without faint praise. In this day of inflated grades and recommendations, I would definitely take notice of such a tepid assessment of Mr. Carvell’s performance as an undergraduate.”). In addition, Carvell was denied ad mission to Vanderbilt Law School, a school which was generally considered inferior to the University of Texas in terms of national reputation. Carvell was also denied admission to the University of Texas Graduate School of Business, which is less competitive than the law school in terms of admissions, as well as Vanderbilt Business School. See Johanson Declaration D-332; Carvell, vol. 4 at 112.

Elliot’s weakness, according to Wellborn, was his lack of “belts and suspenders,” that is, his “consistently weak undergraduate record.” See Wellborn, vol. 1 at 82; Johanson Declaration D-332 (“[I]t is preposterous to suggest that the ... challenged affirmative action program was the cause of his [denial] ____ Elliot is simply a weak candidate.”). Wellborn testified that this type of “disparity” candidate — an applicant with a high LSAT score and low GPA, or vice versa — is occasionally admitted, but only when the scores are high enough to place the individual in the presumptive admit zone. See Well-born, vol. 1 at 88. Under those circumstances, Wellborn testified, “there is a reason to take risks on [such] candidates.” Well-born, vol. 1 at 88. Elliot did not provide any faculty letters of recommendation, 56 leaving the committee members with no information about his college record except his “rather unimpressive 40th percentile class rank.” See Wellborn, vol. 1 at 81-82. Even Elliot himself acknowledged his weak college record in his personal statement: “I was an average student, studying when I needed to, partying more than I should, and not managing my time efficiently____[My GPA] is not of a caliber expected by the University of Texas School of Law.” P-153. Furthermore, Elliot was denied admission to Baylor Law School, a school which is less competitive in terms of admission standards and, importantly, one that did not operate a significant affirmative action program in 1992. See Johanson Declaration D-332.

One of the issues the Court decided in the first trial'was whether Elliot had standing to sue the law school. Following Elliot’s rejection by the law school in April 1992, Elliot’s father sent a letter to Dean Yudof complaining that the law school had rejected his son’s application because of the law school’s “mandatory minority and women quotas.” 57 See P-165 (copy of the letter sent by Elliot’s father). It is undisputed that Elliot’s application was thereafter placed under reconsideration. Hamilton testified that the law school offered Elliot admission in August 1992, shortly before classes were to begin. The law school asserted at the first trial that Elliot lacked the standing to sue because he declined his offer of admission. See Hopwood I, 861 F.Supp. at 565-66. On remand, the Fifth Circuit instructed the Court to reconsider its “contradictory” findings in Hopwood I that (1) Elliot was not notified of his admission to the law school, and (2) Elliot would not have received an offer of admission even under a constitutional system. See Hopwood II, 78 F.3d at 957 n. 57.

As the Court indicated in Hopwood I, Hamilton was the only law school official to testify that Elliot was extended an offer of admission. Unlike other applicants who were admitted late in the process, there was no documentation in Elliot’s file to substantiate Hamilton’s claims, and Elliot testified at the first trial that he was unaware he had been offered either a place on the waiting list or admission to the law school. See Hopwood I, 861 F.Supp. at 565-66. Indeed, Johanson testified at .the first trial that “it was ‘quite unusual’ for someone to be reconsidered and placed on the waiting list without [his] awareness of the decision.” See id. at 566 & n. 44; 1991 trial transcript, Johanson, vol. 5 at 19. Furthermore, the Court found several discrepancies in Hamilton’s trial tes timony and affidavit, which, in addition to the troubling lack of documentation, “weigh[ed] in Elliot’s favor” in terms of determining whether Elliot had standing to sue the law school. See id. at 566. The Court therefore found in Hopwood I that no offer of admission had been communicated to Elliot, that is, that no offer of admission had been made. 58 Moreover, Elliot’s application was only placed under reconsideration because of the letter sent by Elliot’s father. In a race-neutral world, the letter would not have received the special treatment that it did in 1992, and Elliot’s application would not have been reconsidered. See Wellborn, vol. 1 at 80-81.

Wellborn testified that Rogers’s extremely weak college record prevented his admission to the law school despite his relatively high LSAT score. Rogers earned a bachelor’s degree in professional writing, a non-standard major, from the University of Houston-Downtown, a college with one of the lowest LSAT means Wellborn testified he has ever seen. 59 See Wellborn, vol. 1 at 83. Rogers’s cumulative GPA of 3.13 was significantly below the overall 1992 GPA median of 3.52. Although Rogers also received a master’s degree in professional writing from the University of Southern California in 1992, one of the law school admissions committee members who reviewed his file in 1992 described Rogers’s performance there as “decidedly unimpressive.” See Gergen Declaration D-335. Prior to graduating from the University of Houston-Downtown, Rogers attended the University of Texas, where he was placed on scholastic probation once and flunked out twice over a period of three and a half years. 60 Wellborn concluded that Rogers was “clearly” the weakest of all of the applicants with a TI score of 197. Wellborn, vol. 1 at 82. And, Rogers’s assertions that he was a “turnaround” candidate to the contrary, Wellborn testified:

[TJhis was not the picture of someone who just had a shaky start or one bad semester when they were having trouble, but this was extended over a three and a half year period. Three strikes. And that was his experience at a major institution____[H]e did make high grades later on at his degree school, but it’s not as through he transferred to a school that would provide a rigorous testing ground in our opinion.

Wellborn, vol. 1 at 84. In addition, Rogers provided no letters of recommendation and his answers to the questions on the application form were brief and relatively uninformative. See P-171. Given Rogers’s disastrous three and a half years as an undergraduate at the University of Texas and his subsequent failure to distinguish himself academically, the Court is convinced beyond any doubt that Rogers would never be admitted to the law school under any circumstances.

The Court regrets having to discuss publicly and in such great detail the specific weaknesses in each of the plaintiffs’ applications. It goes without saying that their applications do not reflect the sum of their existence as students or individuals. And even though the Court finds the plaintiffs were not denied admission to the law school as a result of the unlawful use of racial preferences, the Court once again acknowledges “the gravity of the noneconomic injury to persons denied equal treatment.” See Hopwood I, 861 F.Supp. at 583. On a final note, being denied admission to the University of Texas School of Law should not be the defining moment in any of these plaintiffs’ lives. At the risk of sounding trite, there are much greater tragedies in life. Each year the law school denies admission to many bright, qualified, and deserving individuals simply because there are not enough spaces available in each entering class. The plaintiffs should not let their denial letters from 1992 — or this lawsuit — continue to determine the course of their lives or prevent them from becoming successful lawyers if that is a goal they still wish to pursue.

IV.

The Court makes the following alternative findings of fact and conclusions of law regarding each of the plaintiffs’ damages to ensure there is no third trial and for the benefit of the circuit court. It should be emphasized, however, that the Court made the following finding in the first trial: “[H]ad the plaintiffs been entitled to damages, none of them established monetary damages as required under the law and rules of this circuit” because the evidence presented merely “consisted of each plaintiff’s testimony and speculation about the value of a law degree.” Hopwood I, 861 F.Supp. at 583. It is beyond the intellectual skills of this Court to comprehend why the plaintiffs have been given a second trial to present evidence regarding damages when they utterly failed to present competent evidence on damages at the first trial. 61 Nevertheless, the Fifth Circuit has instructed the Court to consider new evidence on damages, including damages allegedly accrued since May 1994. See Hopwood II, 78 F.3d at 957 (“[T]he law school’s inability to establish the plaintiff’s non-admission ... opens a panoply of potential relief, depending in part upon what course that plaintiff’s career has taken since trial in mid-1994.”) 62

Each of the plaintiffs, with varying degrees of enthusiasm, has requested on order from this Court directing his or her admission to the law school. 63 See Hopwood, vol. 2 at 177; Rogers, vol. 2 at 86; Elliot, vol. 3 at 268. To the extent the Fifth Circuit concludes the plaintiffs were denied admission as a result of the law school’s unconstitutional admissions procedures, an injunction ordering the law school to admit Hopwood, Elliot, and Rogers would be the most appropriate and equitable remedy the Court could fashion. The remaining issues, therefore, are (1) whether Hopwood, Elliot, and Rogers have proved any economic damages for lost earnings in their second trial; (2) whether Carvell has established any economic damages; and (3) whether Hopwood and Carvell have established any damages for mental anguish. 64 The plaintiffs must prove then-damages by a preponderance of the evidence, and damages which are too speculative, remote, or conjectural cannot be awarded. Collectively, the plaintiffs request over five million dollars in compensatory damages, three million dollars of which is sought for emotional injuries. This figure is exponentially higher than the requests for damages made at the May 1994 trial. See 199b trial transcript, T. Smith, vol. 26 at 40 (counsel for plaintiffs explaining to the Court in closing arguments that the plaintiffs’ damages “are not great”). 65

Hopwood, Elliot, and Rogers request damages for loss of future income resulting from the delay in their legal education. The defendants contend these so-called “front pay” damages are not available in nonemployment eases, in part because they are not foreseeable by educational institutions. 66 They argue that the law school, in rejecting thousands of applicants each year, does not foresee that denied applicants will forego their law school plans entirely but merely that they will have to attend law school elsewhere. While the foreseeability of a plaintiff’s injury is irrelevant in the context of intentional torts, 67 the argument raises two related points. The first point relates to the remote and speculative nature of the plaintiffs’ injuries. The Court emphatically rejeets the notion that the defendants’ actions prevented Hopwood, Elliot, and Rogers from becoming lawyers. At most, the law school prevented the plaintiffs from obtaining a law degree from the University of Texas. Many rejected applicants — Carvell, for instance — successfully attend other law schools. The second point pertains to a plaintiff’s ability to prove front pay damages in a Title VI case. As a practical matter, front pay damages may simply be too speculative or attenuated given the injury in this case — the denial of a seat in law school. Hopwood, Elliot, and Rogers have the burden of establishing they would have successfully graduated from the University of Texas and attained full-time employment as lawyers. To the extent this case is remanded for damages, the Fifth Circuit implicitly rejected the argument that front pay damages are always improper under Title VI or too speculative as a matter of law. But in determining whether a front pay award is appropriate in this case, the Court must grapple with the inherently speculative nature of the plaintiffs’ damages — particularly when, as in the case of Hopwood and Carvell, the requests for damages extend forty years into the future. With these two thoughts in mind, the Court now turns to each plaintiff’s request for damages.

Hopwood contends she suffered economic damages in the amount of $1,360,000, which she claims is the projected career earnings differential between an accountant, her current profession, and an attorney. See P-477. Hopwood additionally seeks $1,500,000 in emotional distress damages she claims re- suited from her denial of admission to the law school. See P-459.

(1) Damages for Loss of Future Income

Hopwood is not entitled to any monetary damages for the alleged loss of future income as an attorney. The Court fails to find from a preponderance of the evidence that Hopwood would have completed her law degree by January 1996: she would not have graduated from law school due to a series of personal tragedies occurring from 1992 to 1996, the time period during which she would have been attending law school. Hopwood contemplated limited enrollment at the law school before she applied for admission in 1992. In a letter attached to her application to the law school, Hopwood sought information regarding reduced participation so that she could continue to care for her severely handicapped daughter Tara while attending law school. See Hopwood, vol. 1 at 132-33. Hopwood testified at trial that Tara “required a substantial amount of energy and time, and had a large number of doctors appointments and hospital visits [and] surgeries.” See Hopwood, vol. 2 at 239. Furr thermore, Hopwood’s husband was in the military and traveled extensively, see Hop-wood, vol. 2 at 239, during which time Hop-wood was Tara’s primary caretaker. In addition, Hopwood was living in San Antonio, Texas at the time and would have been commuting to the law school seventy miles each way for three years.

• Hopwood became pregnant with her second daughter Erica in August 1992, immediately before the start of the law school Fall semester. See Hopwood, vol. 2 at 221. During what would have been the end of Hop-wood’s first year in law school, Erica was born. Erica passed away on May 18, 1993, the day following her birth. See Hopwood, vol. 2 at 221. Hopwood continued as Tara’s primary caretaker and began experiencing marital difficulties due to the stress and depression related to the impending death of Tara and to the demands Hopwood placed on her husband to move away from Texas. See Hopwood, vol. 2 at 170-71, 222-28. In June 1995, Hopwood and her husband separated, presumably leaving Hopwood as Tara’s sole caretaker. On November 23, 1995, Tara passed away. Hopwood, vol. 2 at 176. In February 1996, Hopwood relocated from San Antonio, Texas to Columbia, Maryland, and her husband moved to Korea. See Hopwood, vol. 2 at 222-23.

Hopwood’s limited employment during this time reflects the toll these events took on her professional life. Hopwood was unemployed from June 1991 to January 1993. See Hopwood, vol. 2 at 229. From January 1993 to May 1995, Hopwood worked on a part-time basis ás a CPA for a sole proprietorship in San Antonio; although she worked full-time during tax season, she was able to do so only because she frequently worked out of her home. See Hopwood, vol. 2 at 230-31, 240. In anticipation of her move to Maryland, Hopwood was not employed from May 1995 to February 1996. See Hopwood, vol. 2 at 232. From February 1996 to May 1996, Hopwood worked as an accountant for a temporary agency, and for most of the remainder of the year she was unemployed. In January 1997, Hopwood received her first full-time employment as a CPA since mid-1991. See Hopwood, vol. 2 at 228, 233-34. Hopwood testified that she could have worked full time as a CPA during' these years, but she choose not to do so in part because she knew she had a limited amount of time with her daughter Tara. See Hopwood, vol. 2 at 240.

The Court is not suggesting that women with families or severely handicapped children are incapable of successfully pursuing a law degree or that individuals who suffer tremendous personal tragedies lack the will or ability to accomplish their goals. However, in this ease the facts are clear: Hopwood was facing many serious obstacles, as well as enormous time constraints, that would have undermined her ability to earn a law degree during the 1992 to 1996 time frame. It would have been virtually impossible for anyone — female or male — to complete a law degree traveling 150 miles per school day to a highly competitive environment such as the University of Texas while undergoing similar strains in his or her personal life. 68 Supporting the Court’s finding that Hopwood would not have graduated from law school is her own testimony. See Hopwood, vol. 2 at 172 (testifying that “Tara died unexpectedly ... I didn’t feel I was mentally or emotionally ready to handle what I know is a very difficult coarse [sic ] load in the law program”); Hopwood, vol. 2 at 175-76 (testifying that she has had a “sporadic work history” since Tara’s death). Hopwood’s previous academic performance supports the Court’s conclusion as well. It took Hopwood eight years (instead of four) to earn her undergraduate degree because difficult financial circumstances forced her to spend much of her time working. In addition, although Hopwood repeatedly refers to earning a law degree as a lifelong dream, she has not demonstrated her commitment to that goal by seeking admission to any law school following her rejection from the University of Texas in 1992. See Hopwood, vol. 2 at 172.

Even assuming Hopwood would have completed her law degree, the evidence she proffered to prove economic damages did not reasonably predict any loss in earnings she may have suffered as a result of her denial of admission to the law school. Hopwood’s first expert witness regarding her economic damages was economist Wayne E. Ruhter, Ph.D. Ruhter performed two analyses, “Analysis I” and “Analysis III,” to determine the career earnings differential between attorneys and accountants. See P-477. In Analysis I, Ruhter compared attorney salaries in the private sector in Texas and the mid-Atlantic region with accountant salaries in Philadelphia. 69 See P-477, Exhibit I, Tab 3, Analysis 1-Table 3. Ruhter assumes Hopwood would progress from associate to partnership level in both career paths. In Analysis III, he utilized 1990 Census data, which did not reflect any particular career path or geographic location, to determine the difference in career earnings between female attorneys and female accountants. See P-477, Exhibit III, Tab 3, Analysis 3-Table 3. In both Analysis I and III, Ruhter assumed Hopwood would have graduated from law school, passed the bar exam, attained full-time employment by January 1996, and continued working until she reached age 75 in the year 2037. Subtracting the costs of attending law school, Ruhter concluded the present value of Hopwood’s economic damages to be $424,604 in Analysis I and between $400,364 and $416,876 in Analysis III. See Ruhter, vol. 2 at 258, 265-66. Finally, Ruhter concluded in Analysis II that Hopwood could not have mitigated her damages by enrolling in law school in 1997. 70 See Ruhter, vol. 2 at 272.

In the Court’s opinion, Ruhter’s expert report was riddled with so many untenable premises, logical inconsistencies, and unreliable and inappropriate sources that it cannot support an award of economic damages. By Ruhter’s own admission, his expert report is merely “an evaluation of two career paths” that examines the potential career earnings that are “available” to Hop-wood as an attorney and an accountant. Ruhter, vol. 2 at 253, vol. 3 at 48, 53. Ruhter unequivocally stated that he did not specifically attempt to “forecast” Hopwood’s losses. Ruhter, vol. 3 at 71. Furthermore, prior to preparing his expert report, Ruhter did not interview Hopwood or her current employer or obtain enough information about Hopwood to predict her future earnings capacity to any reasonable degree of certainty. See Ruhter, vol. 3 at 45, 48-49. It is from this initial, misguided premise that all other errors in his report flow. Some, but by no means all, of the more glaring problems are described below.

First consider Analysis I. The accountant career path does not reasonably predict Hop-wood’s future earnings. Ruhter uses salary information from Philadelphia to predict future earnings (even for 1997) instead of reasonably predicting earnings from her current employment in Columbia, Maryland. See Ruhter, vol. 3 at 53. Similarly, the attorney career path is implausibly optimistic, resulting in wildly inflated future earnings. Ruhter assumes Hopwood would (1) become a partner (2) in eight years (3) in a private law firm (4) in a large city in Texas or the mid-Atlantic region. That particular career path is one of the most lucrative a lawyer can take and is extremely difficult and rare. See D-513, Report of Stephen Mims; Ruhter, vol. 3 at 48-49 (recognizing that salaries for private law firm practice are generally higher than for government, corporate, or public interest practice); Mims, vol. 3 at 236 (stating that approximately less than 9% of University of Texas law graduates from the class of 1987 are partners at large law firms by the year 1996). Furthermore, because empirical information related to partnership buy-ins was unavailable, those costs are not reflected in Analysis I. 71 See Ruhter, vol. 2 at 268-69. The Court also questions the use of the David J. White Survey as an appropriate source to project the growth rate of attprney salaries. See D-684, Tab 17 (indicating the survey does not ensure that the same law firms respond to the survey each year it is given).

The most compelling reason to discount Analysis I, however, is its logical inconsistency. Hopwood graduated with an accounting degree in 1988 and earned her CPA license in 1991. See Hopwood, vol. 2 at 140-41. Ruhter does not begin to measure Hopwood’s earnings capacity as an accountant, however, until 1997. Prior to 1997, Ruhter uses Hop-wood’s actual earnings, which, due to her personal circumstances, were low. See P-483. Had Ruhter properly measured Hop-wood’s earnings capacity as a CPA starting in 1992, his own methodology reflects that she would have earned more money over her lifetime by continuing to work as an accountant as opposed to enrolling in law school in 1992. See D-508, Report of James R. Vinson, Ph.D., Table I (indicating that Hop-wood’s career earnings as a CPA would have exceeded her attorney career earnings by $31,412); see also Ruhter, vol. 2 at 271 (acknowledging that there is an opportunity cost associated with a delay in education). If, as Ruhter testified, Analysis I was intended to be a hypothetical comparison of the lifetime earnings available to Hopwood as an accountant and as a lawyer, the analysis should have accounted for Hopwood’s ability to earn as an accountant since the time she earned her degree in 1988 or, at the very latest, since she obtained her CPA license in 1991.

The chief problem with Analysis III is the unreliability and inappropriateness of the Census data which serves as the source for Ruhter’s conclusions. The salary information reflected in the data is self-reported by employees. Self-reporting does not, in and of itself, necessarily render the data inaccurate, but in this case there are other indicia of unreliability. First, some of the data is implausible on its face. For instance, the Census includes salary information on individuals with less than a high school education who claim to be lawyers. See P-477, Exhibit III, Tab 11. Second, the Census significantly undervalues Hopwood’s future earnings potential as an accountant because the data pertaining to accountant salaries includes information for bookkeepers and accountants who, unlike Hopwood, have not obtained a CPA license. See Ruhter, vol 3 at 44-45, 94-95. Unrelated to the problem of self-reporting is the fact that salary information is categorized by age and, consequently, years of experience. While the average age of a female law graduate is twenty-six, Hopwood would have been thirty-four had she graduated from the law school in 1995. See Ruhter, vol. 2 at 266. Ruhter overstates Hopwood’s earnings potential as a lawyer because the Census data he uses in Analysis III compares Hopwood to females who, as a result of their additional eight years of experience, would undoubtedly have higher earnings than Hopwood. See P-477, Exhibit III, Tab 11. Ruhter belatedly recognized this problem and performed Alternative Analysis III, in which he replaces the Census data related to lawyer salaries for the years 1995 to 2000 with data from the David J. White Survey. This approach is perhaps even more flawed because it improperly compares two sources of data for attorney salaries (five years of data from the David J. White Survey and thirty-five years of data from the Census) and only one source of data for accountant salaries (forty years of data from the Census). As Ruhter himself acknowledged, as a general rule it is inappropriate to use different sources of information- in a comparative study because doing so can create a bias in the results. See Ruhter, vol. 3 at 56.

The remaining $900,000 Hopwood claims in economic damages is based on the testimony of Bradford W. Hildebrandt. Hildebrandt is chairman of Hildebrandt, Inc., a management consultant and placement firm related to the legal profession. See D-501 (Hildebrandt’s Report). Hildebrandt is not an economist and did not perform an independent study of Hopwood’s economic losses. See Hildebrandt, vol. 3 at 106. Instead, he reviewed the accuracy of the attorney compensation figures in the private sector as reflected in Ruhter’s Analysis I. Hildebrandt testified that Ruhter’s figures, which project an average annual partner salary of $219,699, undervalue partnership compensation for University of Texas graduates by approximately $100,000 per year. See Hildebrandt, vol. 3 at 107. The Court rejects Hildebrandt’s testimony for two reasons. First, Hildebrandt’s testimony was wholly dependent upon the accuracy of Ruhter’s Analysis I, which the Court has already found unpersuasive. Second, his testimony was inapposite to Hop-wood’s case. It is sheer speculation whether Hopwood would have completed law school, much less whether she would have had an academic record sufficient to result in her recruitment by a private law firm, whether she would have remained at that law firm for eight years, and whether she would have been selected as a partner in that law firm. In short, Hopwood has failed to establish by a preponderance of the evidence that she has suffered the economic damages alleged as a result of being denied admission to the law school.

(2) Mental Anguish Damages

Neither the Supreme Court nor the Fifth Circuit has specifically held that mental anguish damages are recoverable in a suit brought pursuant to Title VI. Although a majority of the Supreme Court in Guardians Association v. Civil Service Commission of City of New York, 463 U.S. 582, 103 S.Ct. 3221, 77 L.Ed.2d 866 (1983), would allow a plaintiff to recover compensatory damages for intentional violations of Title VI, the Supreme Court did not definitively describe the scope of such remedies under the statute. See Shinault v. American Airlines, Inc., 936 F.2d 796, 803 (5th Cir.1991) (quoting Guardians for the proposition that damage remedies under Spending Clause statutes such as Title VI may be limited). The lower federal courts have been grappling with the issue since Guardians, and the law in this area is therefore somewhat muddled. See, e.g., Eastman v. Virginia Polytechnic Institute, 939 F.2d 204, 208 (4th Cir.1991) (concluding that compensatory damages for pain and suffering are not available under Title VT and Section 504 of the Rehabilitation Act of 1973, as amended, 29 U.S.C. § 794 (West 1994), which forbids discrimination on the basis of disability).

In Franklin v. Gwinnett County Public Schools, 503 U.S. 60, 71, 112 S.Ct. 1028, 1035, 117 L.Ed.2d 208 (1992), the Supreme Court held that “federal courts have the power to award any appropriate relief’ under Title IX, a statute that was patterned after Title VI. 72 Since Franklin, the strong trend among federal courts is to allow plaintiffs to recover for mental injuries under Title VI and similar federal anti-discrimination statutes. See Rodgers v. Magnet Cove Public Schools, 34 F.3d 642, 644 (8th Cir.1994) (holding that “the full spectrum” of legal and equitable remedies are available under Title VI and Section 504 of the Rehabilitation Act of 1973); Waldrop v. Southern Co. Serv., Inc., 24 F.3d 152, 156-157 (11th Cir.1994) (citing Franklin for the holding that the full panoply of legal remedies are available under Section 504 of the Rehabilitation Act); Pandazides v. Virginia Bd. of Educ., 13 F.3d 823, 830 (4th Cir.1994) (overruling Eastman in light of Franklin). In any event, this issue is a question of law that is better left to the Fifth Circuit should the panel deem it proper to reach the issue of damages. The Court merely assumes for purposes of this order that mental anguish damages are recoverable under Title VI.

To the extent the Court fails to find by a preponderance of the evidence that Hopwood would have completed her law degree, the defendants are not liable for-any mental anguish damages because there is no causal connection between the harm and the violation. See, e.g., Patterson v. P.H.P Healthcare Corp., 90 F.3d 927, 938 (5th Cir. 1996) (asserting the general principle that there must a causal connection between the injury and any award for damages). Even assuming Hopwood would have earned a degree from the law school, several of Hop-wood’s claims for emotional injury are not compensable because they are not the result of the defendants’ conduct. For example, claims for damages related to Hopwood’s allegation that the defendants falsely maligned her qualifications in the media pertain to the litigation she initiated and the justifiable defense of this case and therefore are not recoverable. 73 See Hopwood, vol. 2 at 187-90 (describing her voluntary media appearances and interviews). Other claims for damages, such as her contention that the law school’s discrimination was a factor that caused her to begin smoking, are frivolous and do not merit discussion.

Hopwood testified she experienced frustration, depression, disappointment, diminished self-confidence, and anger as a result of being denied admission to the law school. See Hopwood, vol. 2 at 170. In her interrogatory responses, Hopwood also claims feelings of hopelessness, confusion, anxiety, embarrassment, inability to concentrate, indecisiveness, inadequacy, and stigmatization. See P-459; P-464. Finally, she contends her rejection by the law school indirectly strained her marriage. See Hopwood, vol. 2 at 170-71; P — 470 (deposition testimony of friend confirming these feelings). Hopwood testified she did not experience any physical symptoms or pain, nor did she seek any psychiatric or medical care, as a result of being denied admission, although she did occasionally discuss the situation with a therapist from whom she had sought grief counseling following Erica’s death. See Hopwood, vol. 2 at 217-18. '

In addition to Hopwood’s testimony, the Court heard corroborating evidence regarding the nature and extent of her emotional injuries from Paul Lees-Haley, Ph.D. Lees-Haley is a forensic psychologist, not a mental health provider, and has never treated Hop-wood for any mental injury. Rather, Lees-Haley evaluated Hopwood based upon her self-reported complaints to determine whether Hopwood suffered any emotional injury as a result of the defendants’ actions. See P~ 526 (Expert Report); Lees-Haley, vol. 3 at 153. Lees-Haley testified that Hopwood suffered some emotional distress from three independent sources: (1) the law school’s discrimination against her', (2) the - law school’s rejection of her application for admission, and (3) public comments made by the law school and her interactions with lawyers in the case. See Lees-Haley, vol. 3 at 152.

Lees-Haley diagnosed Hopwood with chronic “adjustment disorder with mixed anxiety and depressed mood,” which he believes occurred partly in response to the “stressor” 74 of being denied admission and of being a victim of discrimination. See Lees-Haley, vol. 3 at 155-56. He states in his report that adjustment disorder “typically resolves within six months unless the symptoms are in response to ... a stressor that has enduring consequences,” such as Hop-wood’s rejection from the law school. P-526. Lees-Haley also testified that the law school’s actions aggravated other mental injuries, such as those resulting from the deaths of her children, the break-up of her marriage, and the loss of a job. See Lees-Haley, vol. 3 at 152; P-526 (“[Hopwood] plausibly suggests that had she been accepted to law school, some of the other problems and their effects might have been mitigated to some degree ... because at least she would not have had the feeling she had lost all of her dreams at once.”). On the other hand, Lees-Haley testified that the law school’s actions “might have caused some mild disability in the sense of making it harder for her to concentrate,” but that they have not caused Hopwood any “permanent” problems that will cause her “pain and suffering and disability ... for the rest of her life.” Lees-Haley, vol. 3 at 154; vol. 3 at 209-11 (indicating that Hopwood is not suffering from any disability and is fully functional). Moreover, he opined that Hopwood would not have had a diagnosable mental condition were it not for other negative “stressors” in her life. See Lees-Haley, vol. 3 at 220-21.

Hopwood has had more than her share of tragedies, both before and after she was denied admission to the law school, all of which have caused her severe emotional distress. Without doubt, the evidence establishes the least traumatic “stressor” in her life was the law school’s denial of her application. However, because of the unusual circumstances occurring in Hopwood’s personal life from 1992 through 1996 and her mental condition, the Court would find from a preponderance of the evidence that Hopwood has suffered mental anguish damages as a result of the law school’s rejection of her application. The denial of admission aggravated Hopwood’s preexisting emotional injuries beyond the mere hurt feelings, frustration, and anger that are a part of everyday life. See Patterson, 90 F.3d at 940. Accordingly, the Court would find from a preponderance of the evidence that Hopwood has sustained mental anguish damages in the amount of $6,000 as a result of the law school’s rejection of her application for admission.

Carvell is the only plaintiff who pursued a law degree in 1992 after his rejection from the law school. Carvell graduated with a joint degree in law and a master’s in business administration from Southern Methodist University (“SMU”) in 1996. See Carvell, vol. 4 at 101. Since August 1996, he has been employed as an attorney with the private law firm of Bickel & Brewer in Dallas, Texas, 75 where he was earning a starting salary of $50,000 plus benefits at the time of trial (Carvell has since earned a pay increase). See Carvell, vol. 4 at 105-06,114-15. Carvell seeks total economic damages in the amount of $705,886, consisting of $40,036 in increased tuition for attending SMU. and $668,850 in damages resulting from the diminished value of an SMU law degree. 76 See P-478. Additionally, Carvell seeks $1,500,000 in mental anguish damages resulting from the law school’s discrimination against him and the denial of his application for admission. See P-460.

(1) Tuition Differential

Carvell would have attended the University of Texas over SMU Law School had he been admitted to the law school in 1992. See Carvell, vol. 4 at 98; 19% trial transcript, Carvell, vol. 10 at 9. Accordingly, the Court finds from a preponderance of the evidence the difference in the cost of tuition between the two schools was $40,036. See P-478, Exhibit 2, Table 2; Ruhter, vol. 3 at 4.

(2) Damages for Loss of Future Income

Ruhter calculated the diminished value of Carvell’s degree by comparing the difference in earnings hypothetically available to University of Texas and SMU law graduates based upon the average starting associate salaries of all private law firms that recruited at those two schools in 1995, regardless of whether those firms actually hired graduates of .those two law schools that year. See. Ruhter, vol. 3 at 3; P-478, Exhibit 3, Table 3. Ruhter did not identify any salary differential at the partnership level because he was unaware of any empirical information available to calculate it; therefore, Ruhter analyzed Carvell’s alleged lost earnings only as an associate until the year 2003. See Ruhter, vol. 3 at 80-81; P-478, Exhibit 3, Table 3. Ruhter testified that the present value of the earnings differential was $11,850. Ruhter, vol. 3 at 4. Hildebrandt reviewed the accuracy of the figures contained in Ruhter’s report. He testified that Ruhter underestimated Carvell’s losses by not carefully analyzing “the weight of the law. schools” and by not including any damages for the differential in partnership compensation between graduates of SMU and the law school. Hildebrandt testified these damages were approximately $50,000 per year, for a total of $654,000. Hildebrandt, vol. 3 at 110-11.

The Court fails to find from a preponderance of the evidence that Carvell is entitled to any damages for loss of future income; indeed, his requests for economic damages are counterfactual. First, Carvell did not use the Career Services office at SMU to obtain any employment. See Carvell, vol. 4 at 122. Consequently, a comparative analysis of the average starting salaries of law firms that recruit at the University of Texas and SMU provides an inappropriate basis for measuring damages in Carvell’s case. Second, Ruhter calculates Carvell’s lost earnings beginning in mid-1995 even though Carvell did not graduate from law school until May 1996. See Carvell, vol. 2 at 122. Third, the “opportunity” differential between SMU and the University of Texas is due in large part to the high starting salaries of national firms and regional boutiques which tend to recruit more heavily at the University of Texas. In general, these firms only hire top-tier students with very strong academic records. See Jones, vol. 4 at 82-84. Carvell, however, graduated in the third quarter of his class at SMU; therefore, these higher earnings probably would not have been available to Carvell because he would not have had an academic record at the University of Texas sufficient to result in recruitment by those firms. See Carvell, vol. 4 at 114. Fourth, in requesting $654,000 for the differential in partnership compensation, Carvell apparently overlooked the salient fact that he is not on partnership track at Bickel & Brewer. 77 See Carvell, vol. 4 at 106.

Carvell proffered speculative evidence that he has suffered lost earnings as a result of not obtaining his degree from the law school. In fact, the evidence suggests Carvell has suffered no loss in earnings: Carvell’s starting salary is higher than the average starting salary for graduates of the University of Texas in 1995. See D-685, Tab 4 (indicating that the average starting salary for 1995 University of Texas graduates was $49,761); D-685, Tab 8 (indicating that the average starting salary for 1995 SMU graduates was $51,464). Even assuming an earnings differential exists and that it can be quantified to a reasonable degree of certainty — a notion the Court seriously doubts — the statistics regarding average starting salaries in private, government, corporate, and public interest practice suggest the differential actually favors SMU graduates. See D-685, Tab 4 & 8. Moreover, Carvell proffered no evidence that private law firms distinguish among lawyers, in terms of salary and/or benefits, on the basis of where they obtained their law degree. True, graduates of highly regarded universities, tend to have enhanced career opportunities over graduates of universities that are not as highly regarded. But, in the Court’s opinion, the ability of an individual to succeed in his or her profession depends much more upon the value of that individual rather than the value of his or her degree. 78

(3) Mental Anguish Damages

Carvell testified he experienced disappointment, anxiety, stress, shock, frustration, and anger and that he became withdrawn from others as a result of his denial of admission to the law school. See Carvell, vol. 4 at 98-101; P-472 (deposition testimony of friend confirming these feelings). In his interrogatory responses, Carvell also indicated he suffered feelings of confusion, depression, embarrassment, inability to concentrate, lack of motivation, indecisiveness, diminished confidence, inadequacy, stigmatization, and diminished self-esteem. See P-460; P-465. On the other hand, he testified he was not “shattered or crippled” by the experience and that he is functioning well at his current employment. See Carvell, vol. 4 at 101. Moreover, Carvell never sought psychiatric or medical treatment for any emotional injuries caused by the defendants’ conduct, even though he has suffered from “very serious and substantial mental health problems in the past.” Carvell, vol. 4 at 101, 119-20; Carvell Stipulation 1.

As with Hopwood, Lees-Haley evaluated Carvell’s self-reported complaints to determine the nature and extent of his emotional injuries resulting from the defendants’ actions. Lees-Haley stated in his expert report that “Carvell appears to have a chronic, mild depression that pre-existed this injury and continues to the present, which was temporarily mildly aggravated by not being admitted to the University of Texas School of Law and by discovering that he was the subject of race discrimination.” See P-527 (emphasis added). The Court does not find from a preponderance of the evidence that the denial of admission caused Carvell any compensable emotional injuries, either by themselves or resulting from the alleged temporary aggravation of his preexisting depression. The denial, at most, resulted in the mere hurt feelings, frustration, and anger that are simply a part of everyday life.

C.

Elliot seeks $56,021 in lost earnings he contends resulted from his delayed entry into the workforce as a lawyer during the pendency of this lawsuit. He arrived at this figure by calculating the five-year differential, adjusted for inflation and discounted to present value, between his current salary as a CPA and the average starting salary of 1995 graduates of the law school. The Court does not find from a preponderance of the evidence that Elliot has suffered any loss in earnings as an attorney that can be reasonably attributed to his rejection by the law school. Elliot declined an offer of admission to attend Texas Tech Law School in 1992. See Elliot, vol. 3 at 273. Therefore, the delay in his legal education resulted not from the law school’s admissions procedures but from Elliot’s deliberate choice not to attend Texas Tech. Furthermore, separate from the issue of the remoteness of the injury is the issue of whether Elliot has mitigated his damages. Elliot could have attended Texas Tech in 1992 or he could have applied to other law schools within the last five years. The damages Elliot seeks are also speculative and remote.

D.

Rogers seeks compensatory damages of $332,867.94 plus one third of any amount awarded in attorneys’ fees. 79 See D-660, Exhibit A. Rogers claims approximately $200,000 in lost wages as a lawyer during the pendency of the lawsuit; roughly half the lost earnings he claims are losses he sustained in unpaid loans he made to his business. 80 In addition, Rogers seeks $200 for the cost of reapplying to the law school, $795 to attend a Princeton Review LSAT course, $950 in costs to move back to Austin in the event the Court orders an injunction, a minority scholarship in the amount of $20,000 plus tuition for three years, and $909.24 in expenses for voluntarily attending a conference on affirmative action in California. The Court does not find from a preponderance of the evidence that Rogers has sustained any economic damages as a result of his rejection by the law school. The reasons are essentially the same for which the Court deified Elliot’s request for damages — the damages alleged are not attributable to the defendants’ conduct and are speculative and remote. Rogers declined an offer to attend the University of Houston School of Law in 1992. See Rogers, vol. 2 at 100. The remaining requests for damages (those that do not relate to Rogers’s request for lost earnings) are frivolous and do not merit discussion.

V.

Title 42 of the United States Code, Section 1988, provides that a court “in its discretion, may allow the prevailing party, other than the United States, a reasonable attorney’s fee as part of the costs.” The Court denied the plaintiffs’ original requests for fees and expenses following the May 1994 trial, concluding that the plaintiffs, although prevailing parties under statute, only attained de minimis relief. The plaintiffs separately appealed that judgment on January 9, 1995. On July 26, 1996, the Fifth Circuit granted the plaintiffs’ motion to vacate the Court’s judgment on attorneys’ fees and remanded the ease with an instruction to reconsider whether the fees and expenses requested at the trial and appellate levels were reasonable and necessary for the prosecution of this case. The plaintiffs’ attorneys purportedly spent 7,241.41 hours litigating this case, and the fees and expenses sought exceed $1.5 million.

At the May 1994 trial, the plaintiffs were represented by (1) Steven Smith; (2) Terral Smith; (3) Joseph Wallace, Paul Harris, R. Kenneth Wheeler, and Scott Wilson of the private law firm of Wallace, Harris & Sims (‘WHS”), located in West Virginia; and (4) Michael McDonald, Michael Rosman, Vincent Mulloy, Michael Troy, and Joseph Shea of the Center for Individual Rights (“CIR”), a not-for-profit, public interest law firm located in Washington, D.C. Steven Smith is an Austin sole practitioner who initiated the suit. He was the only attorney representing the plaintiffs during the first eight months of the ease. Terral Smith, also a sole practitioner in Austin, conducted the majority of the trial presentation. Along with Terral Smith (although to a lesser degree), Wallace participated in the trial presentation, while the other attorneys of WHS assisted in trial preparation. The CIR’s main involvement in the case was to conduct most of the legal research, write all legal briefs, and bankroll the case. Following the 1994 trial, the plaintiffs sought fees and expenses of $853,847.69 for 4,840.56 hours of work (“trial hours”). 81 After the successful appeal of this case, the plaintiffs increased their requests to $941,-391.24 to reflect current hourly billing rates. 82

Beginning in November 1994, the plaintiffs (for reasons unknown to the Court) obtained separate representation on appeal to the Fifth Circuit and the Supreme Court. Hop-wood and Carvell were represented by (1) Theodore Olson, Douglas Cox, Thomas Hun-gar, Walter J.. Scott, and, Daniel Nelson of the Washington, D.C. office of Gibson, Dunn & Crutcher (“GDC”); and (2) McDonald, Rosman, Troy, and Hans Bader of CIR. 83 Terra! Smith and Steven Smith remained counsel of record for Elhot and Rogers. GDC primarily handled the appeal relating to the constitutionality of the law school’s use of racial preferences, including briefing the issues to both appellate courts and arguing the case to the Fifth Circuit. CIR worked on the appeals relating to the proposed interventions by several minority groups and this Court’s denial of attorneys’ fees. In addition, CIR had primary responsibility for coordinating activity among the lawyers representing Hopwood and Carvell. It appears Terral Smith and Steven Smith assisted GDC with the trial record only — their billing records demonstrate they did little independent appellate work and instead relied on GDC to argue the merits of the case to the Fifth Circuit.and the Supreme Court. The plaintiffs seek fees and expenses of $614,138.56 for 2,400.85 hours of work during the appellate phase of this case (“appellate hours”). 84

To determine the appropriate award, the Court must calculate the “lodestar” by multiplying the number the number of hours reasonably spent on the litigation by a reasonable hourly billing rate. See Hensley v. Eckerhart, 461 U.S. 424, 434, 103 S.Ct. 1933, 1939, 76 L.Ed.2d 40 (1983). The fee applicant bears the burden of establishing that the hours expended and the biking rate are reasonable. Id. at 437, 103 S.Ct. at 1941. The Court should consider the factors set forth in Johnson v. Georgia Highway Express, Inc., 488 F.2d 714, 717-19 (5th Cir. 1974), to evaluate the reasonableness of the fee requests. See Walker v. United States Dep’t of Housing & Urban Dev., 99 F.3d 761, 771 (5th Cir.1996) (indicating that most of the Johnson factors are reflected in the lodestar amount and cannot be used to compute any upward or downward adjustment of the lodestar); Watkins v. Fordice, 7 F.3d 453, 457 (5th Cir.1993). The Johnson factors are (1) the time and labor required; (2) the novelty and difficulty of the questions; (3) the skkl required to perform the legal services properly; (4) the preclusion of other employment by the attorney due to acceptance of the case; (5) the customary fee; (6) whether the fee is fixed or contingent; (7) time limitations imposed by the client or the circumstances; (8) the amount involved and the results obtained; (9) the experience, reputation, and ability of the attorneys; (10) the “undesirability” of the case; (11) the nature and length of the professional relationship with the client; and (12) awards in simkar cases. Watkins, 7 F.3d at 457 n. 4. As discussed below, the defendants challenge the fees sought for certain categories of work performed by counsel, as well as fees sought for duplicative, unnecessary, and excessive hours. Moreover, the Court made certain discretionary reductions in accordance with the Johnson factors.

(1) Public and Media Relations

The defendants argue the plaintiffs are not entitled to reimbursement for time spent dealing with the media. As the defendants correctly contend, the Fifth Circuit routinely denies such requests because they are not related to the litigation of the case. See id. at 458; Associated Builders & Contractors of La., Inc. v. Orleans Parish Sch. Bd., 919 F.2d 374, 380 (5th Cir.1990) (affirming the district court’s decision to discount award for time spent preparing press releases); see also Rum Creek Coal Sales, Inc. v. Caperton, 31 F.3d 169, 176 (4th Cir. 1994) (“The legitimate goals of litigation are almost always attained in the courtroom, not in the media.”). The plaintiffs urge the Court to apply the standard of recovery announced in Davis v. City and County of San Francisco, 976 F.2d 1536, 1545 (9th Cir. 1992), vacated in part on other grounds, 984 F.2d 345 (9th Cir.1993), in which the Ninth Circuit held that fees incurred in public relations efforts are recoverable as long as such efforts “contributed, directly and substantially, to the attainment of [the party’s] litigation goals.” Watkins, 7 F.3d at 458 (noting the existence of the Davis rule but refusing either to accept it or to reject it). Even applying the Davis standard to this case, the Court finds the plaintiffs have not shown that their efforts in the media assisted in the litigation. 85 Instead, the public relations efforts in this case clearly relate to the divisive political — as opposed to legal — issue of affirmative action. Cf. Rum Creek, 31 F.3d at 176 (refusing to allow recovery for time spent dealing with the media because those efforts were aimed at rehabilitating the plaintiff’s public image). Accordingly, the Court deducts 5.50 trial hours and 68 appellate hours from its award for time spent by counsel in press and television interviews, as well as “reviewing clips” and other actions related to the media, because they were not reasonable and necessary for the prosecution of this ease. 86

(2) Proposed Interventions

The defendants argue the plaintiffs are not entitled to attorneys’ fees for work related to the failed interventions of the Thurgood Marshall Legal Society, the Black Pre-Law Association, the NAACP Legal Defense Fund, and the Mexican-Ameriean Legal Defense and Educational Fund. The Supreme Court has held that a prevailing party in a civil rights lawsuit cannot recover attorney’s fees against an intervenor unless the intervention is “frivolous, unreasonable or without foundation.” Independent Fed. of Flight Attendants v. Zipes, 491 U.S. 754, 761, 109 S.Ct. 2732, 2737, 105 L.Ed.2d 639 (1989). Several circuit courts have read Zipes to imply that plaintiffs should bear the risk of incurring intervention-related costs as a result of filing a lawsuit and have therefore extended Zipes’s holding to a prevailing party’s claims for intervention-related attorneys’ fees against a losing defendant. See Rum Creek, 31 F.3d at 176-78 (“Zipes instructs us not to shift intervention-related expenses to the losing defendant.”); Bigby v. City of Chicago, 927 F.2d 1426, 1428-29 (7th Cir. 1991) (“Zipes indicates that a ... defendant’s fee liability does not presumptively extend to cover fees incurred by plaintiffs in litigating third-party interests ____”). In this case, the defendants did not participate in the proposed third-party interventions. 87 Be cause the plaintiffs are not prevailing parties as against the defendants on the issue of intervention, they are not entitled to attorneys’ fees for time spent litigating that issue, and the Court deducts 184.90 trial hours and 338.47 appellate hours from its award of fees. 88 See Bigby, 927 F.2d at 1429 (holding that the plaintiff could not recover against a defendant that opposed intervention); Reeves v. Harrell, 791 F.2d 1481, 1483-84 (11th Cir. 1986) (holding that a plaintiff is not a prevailing party vis-a-vis á defendant who remains neutral on issues raised by an intervenor’s complaint).

(3) Travel Time

The Court is of the opinion that time spent for travel should not be billed at the same hourly rate sought for active legal work. Accordingly, the number of hours sought for travel time shall be reduced by fifty percent. See Watkins, 7 F.3d at 459 (affirming the district court’s decision to completely disallow compensation for travel time); Sun Publ’g Co., Inc. v. Mecklenburg News, Inc., 594 F.Supp. 1512, 1520 (E.D.Va.1984) (reducing fees for travel time by eighty percent) (citing Younger v. Glamorgan Pipe & Foundry Co., 418 F.Supp. 743, 794 (W.D.Va. 1976), vacated and remanded on other grounds, 561 F.2d 563 (4th Cir.1977)). The attorneys spent 155.50 trial hours and 36.05 appellate hours traveling, resulting in a discount of 77.75 trial hours and 18.03 appellate hours. 89

(I) Issues on Remand

The plaintiffs seek compensation for time spent preparing for the April 1997 trial. These time entries begin in mid-July 1996. Because the plaintiffs are not prevailing parties on the issues submitted by the Fifth Circuit on remand, they are not entitled to compensation for those failed efforts. The Court has identified 249.10 appellate hours spent in this endeavor. 90

(5) Duplicative Work Product and Billing Judgment

In addition to disallowing hours for certain kinds of work altogether, the Court must also consider whether the plaintiffs, have adequately eliminated fees sought for duplicative, unproductive, and unnecessary work product. See Hensley, 461 U.S. at 432-34, 103 S.Ct. at 1939; League of United Latin American Citizens v. Roscoe Indep. Sch. Dist., 119 F.3d 1228, 1232-1233 (5th Cir.1997) (“LULAC”). Moreover, the plaintiffs must exercise proper billing judgment when submitting fee applications. 91 See Walker, 99 F.3d at 769. The Court has very carefully perused the attorney time entries for instances of redundant and wasteful hours. Given the sheer number of attorneys on the case (there were a total of sixteen attorneys involved — eleven at trial and twelve on appeal), there was significant duplication of effort for which the plaintiffs are not entitled to compensation. See Leroy v. City of Houston, 906 F.2d 1068, 1079 (5th Cir.1990) (indicating that hours which result from the ease being overstaffed are not hours reasonably expended and should be excluded from the lodestar calculation). Indeed, the lawyers in this case exceeded the plaintiffs by three-to-one even though the legal issues for each plaintiff were identical. Reductions for duplicative work are particularly appropriate in this case because only two attorneys, Terral Smith and Wallace, basically performed the trial presentation, while the bulk of the compensable appellate work was performed by GDC. See id. (“[T]here should be no compensation for hours spent in duplicative activity or spent in the passive role of an observer while other attorneys performed.”) In addition, although it is the plaintiffs’ prerogative to obtain additional representation on appeal, doing so in this case required attorneys for GDC to familiarize themselves with the record and legal issues already known by trial counsel. Finally, the amount of billing in unproductive attorney and client conferences at the trial and appellate levels is somewhat astonishing, especially given the fact that each attorney present at the conference usually billed this time at his full hourly rate. For example, it appears that more than half of the $38,500 in fees Harris submitted relate exclusively to co-counsel conferences.

The following is a non-exhaustive list of additional examples of redundant and unnecessary work product, as well as a lack of billing judgment: (1) thousands of dollars were spent repeatedly reading and reviewing the same Supreme Court cases on affirmative action; (2) compensation is sought for nonlegal work such as reading a national best seller on affirmative action, attending a panel discussion regarding the impact of the case, and reviewing co-counsel agreements; (3) an excessive amount of time was spent reading the defendants’ pleadings and reviewing the documents in the case; 92 (4) virtually without exception, the attorneys listed hours spent learning the rules and procedures of the Western District of Texas, the Fifth Circuit, and the Supreme Court; (5) several research projects involved basic civil rights doctrines which should already have been known by counsel, such as the standards of review in equal protection cases; (6) there was excessive research and drafting regarding the issue of Elliot’s standing to sue; (7) the attorneys spent hours exhaustively reading, reviewing, and summarizing the deposition testimony of Professor Johanson and Dean Yüdof, such as one attorney from CIR who spent 45 hours and $5,625 summarizing Johanson’s deposition; (8) there was excessive and inappropriate billing for routine clerical work, organization of files, and data entry which should have been accomplished by non-attorneys; (9) many time entries were inadequately documented under the rules and standards of the Fifth Circuit, particularly by attorneys for GDC, WHS, and Terral Smith, and some of CIR’s time entries did not even have a description of the work performed; 93 (10) Steven Smith had entries for time, spent in relation to three plaintiffs who withdrew from the lawsuit prior to trial; and (11) CIR seeks compensation for Dr. Michael Greve, the founder of CIR and a non-lawyer, for consulting work performed during the trial phase of the case. These charges are inappropriate, and they will not count against the defendants.

It was impractical for the Court to wade through literally hundreds of time entries in order to assess a specific number of duplicative and excessive hours for each attorney. Moreover, “[t]he proper remedy when there is no evidence of billing judgment is to reduce the hours awarded by a percentage intended to substitute for the exercise of billing judgment.” Walker, 99 F.3d at 770. Therefore, after careful consideration, the Court concludes that a twenty-five percent reduction in the amount of trial hours for Steven Smith, Terral Smith, and WHS is equitable and warranted. The fee application CIR. submitted for trial work, however, is another story. The “padding” of attorney hours was quite extensive, not only in the amount of time it took attorneys to complete a particular task but also in the types of tasks the attorneys performed. In the Court’s opinion, a reduction of thirty-five percent would more accurately reflect the number of hours CIR reasonably spent on litigation.

The percentage reduction for work at the appellate level is a more complex question because, unlike the work performed for trial, the appeal was less of a collective endeavor. Compared to the work performed by GDC, the work performed by CIR, Steven Smith, Terral Smith, and Wallace was of considerably less value to the' plaintiffs. 94 In fact, their work on appeal primarily consisted of performing duplicative research, assisting GDC with the trial record, filing notices of appearance, reviewing court decisions, and revising the plaintiffs supplemental applications for attorneys’ fees. 95 Therefore, the Court concludes the following reductions in appellate hours are fair and appropriate: a twenty-five percent reduction for GDC and a thirty-five percent reduction for CIR, Steven Smith, and Terral Smith.

(6) Degree of Success

Without question, the plaintiffs attained extraordinary success in the appellate courts. The plaintiffs accomplished the principal goal of the lawsuit — to dismantle all forms of racial preferences in public higher education in Texas. Thus, the Court does not question the success of the suit at the appellate levels through July 1, 1996, the date the Supreme Court denied the defendants’ petition for writ of certiorari. See Hensley, 461 U.S. at 435, 103 S.Ct. at 1940 (“[T]he fee should not be reduced simply because the plaintiff failed to prevail on . every contention raised in the lawsuit.”). However, the plaintiffs did not succeed in one important aspect of their lawsuit — that they were denied admission because of the law school’s affirmative action program. The failure of the plaintiffs to obtain specific injunctive and monetary relief is not, in the Court’s opinion, a trivial or insignificant matter. See Albright v. Good Shepherd Hosp., 901 F.2d 438, 441 (5th Cir. 1990) (asserting that fees, must be apportioned in light of the results obtained). Accordingly, the final fee award for trial level work shall be reduced for each attorney and/or law firm by fifteen percent.

Steven Smith, Terral Smith, and Wallace submitted current hourly rates of $125, $225, and $250, respectively. The current hourly rates submitted by CIR are $280 for McDonald, the president of CIR; $260 for Rosman, CIR’s general counsel; $240 for Mulloy; $210 for Shea; $180 for Troy; and $150 for Bader. The current rates for paralegals (CIR used three at trial and six on appeal) is $70 per hour. GDC submitted the following hourly rates for its lawyers: $450 for Olson, a former United States Assistant Attorney General, a partner, and head of GDC’s appellate and constitutional law practice group; 96 $275 for Cox, an attorney “of counsel” to firm; $275 for Hungar and Scott, two senior associates with the firm; and $215 for Nelson, a junior associate with the firm. The current billing rates for the four GDC paralegals who worked on the appeal range from $75 to $100 per hour. The Court does not have current hourly rates for Harris, Wheeler, Wilson, and the one WHS legal assistant who worked on the ease. Their historic hourly rates are $200 for Harris and Wheeler, both partners at WHS; $85 per hour for Wilson, an associate; and $60 per hour for the legal assistant. 97

(1) Relevant Community

The reasonable hourly billing rate is based on the prevailing market rates in the relevant community — in this instance, Austin, Texas, where the case was tried. See Blum v. Stenson, 465 U.S. 886, 895, 104 S.Ct. 1541, 1547, 79 L.Ed.2d 891 (1984). GDC, CIR, WHS, and Wallace argue, however, that they are entitled to the prevailing market rate for comparable services performed in Washington, D.C. (where GDC and CIR are located) because local counsel was unavailable to take the ease. See National Wildlife Fed’n v. Hanson, 859 F.2d 313, 317-18 (4th Cir.1988); Chrapliwy v. Uniroyal, Inc., 670 F.2d 760, 768 (7th Cir.1982) (both holding that outside counsel may be retained when local counsel is unavailable and the plaintiffs have acted reasonably in choosing outside counsel). Several attorneys submitted affidavits with the applications for attorneys’ fees indicating that they attempted the locate pro bono assistance from members of the private bar in Texas but, other than Terral Smith, none could be obtained. See, e.g., Affidavit of Michael McDonald, August 29, 1994. The plaintiffs’ attorneys also assert the complexity of the case and the superior resources of the defendants required the involvement of outside counsel. The Court rejects those arguments and the affidavits. From the inception of this case, there has been a shortage of clients, not lawyers. Steven Smith spent months soliciting clients through the mail, and only seven individuals responded. See D-650 (sample solicitation letter). Once Smith was retained, he sought the assistance of CIR because he apparently lacked both the expertise and the funds to try the lawsuit. See id. The Court will not assess against the defendants the higher rates of Washington, D.C. because local counsel got in over his head. In addition, Terral Smith provided competent and skilled legal representation in the trial court, and the Court has no doubt he was capable of the same degree of competence regarding the appeal had it been determined he should have remained the plaintiffs’ primary lawyer.

(2) Prevailing Market Rate

When the attorney’s customary rate fails within the range of hourly fees in the prevailing market for attorneys of similar skill and experience, the Court should consider that rate in setting a reasonable hourly rate; if the Court deviates from the customary rate, the Court must specifically articulate its reasons for doing so. See LULAC, 119 F.3d at 1234; Watkins, 7 F.3d at 459. Although the hourly fee awarded must be supported by the record and cannot merely reflect the experiences of the Court, see LULAC, 119 F.3d at 1234, the Court may consider evidence outside the record, such as the hourly rate given in similar cases in the Western District of Texas, see Walker, 99 F.3d at 770. Terral Smith was the only attorney who timely presented evidence regarding the prevailing market rates for comparable services per formed in Austin, Texas. 98 He indicates that the prevailing market rate for federal court practice in the Western District of Texas is between $220 to $300 per hour (these rates were charged in a case involving the Endangered Species Act). For federal civil rights litigation, Terral Smith indicates that a rate of $225 per hour is reasonable for an attorney of his skills and abilities. See Affidavit in Support of Motion for Attorney’s Fees, Plaintiffs’ Motion for Attorneys’ Fees, August 29, 1994. In the Court’s experience, most recent fee awards in civil rights cases in the Western District of Texas generally have billing rates ranging from $150 to $200 per hour. 99 The record in this ease does not support a billing rate higher than $225 per hour. See LULAC, 119 F.3d at 1234.

(3) Assessing the Rates

The hourly rate of $125 requested by Steven Smith appears reasonable given both his limited trial and appellate experience and his role in the case. He was primarily involved in discovery, including document review, responding to the defendants’ discovery requests, and participating in discovery disputes. He also performed legal research related to tangential issues in the case, such as Hopwood’s residency status. As discussed above, he performed little substantive work on the appeal. Furthermore, this rate is higher than Smith’s normal billing rate and is the highest rate Smith has ever received in any federal civil rights litigation he has undertaken. Terral Smith’s hourly rate of $225 is also reasonable. Given his competence and years of experience, the rate is in line with prevailing market rates for comparable work.

The rates requested by Wallace, CIR, and GDC, however, are excessive. Wallace’s rate of $250 runs afoul of the current prevailing market rates in Austin. Wallace’s rate for trial level work is therefore reduced to $225, a rate that reflects his role in the trial and his years of experience. As for the time Wallace spent on appeal, the Court concludes that his rate should be reduced to $200 per hour because all of Wallace’s compensable post-trial time relates to his application for attorney’s fees. As the Fifth Circuit has indicated, an attorney should not command his or her normal billing rate for such work. See Leroy, 906 F.2d at 1079 (“Certainly, the base rate for time spent on computing and enforcing attorneys’ fees should be less than that allowed for professional services rendered primarily on the merits.”). As for Harris and Wheeler, the Court assesses an hourly rate of $200. According to their billing records, their main function was to consult with co-counsel about the case; they performed very little independent work and did not actively participate at trial. The Court concludes an hourly rate $100 for Wilson for the legal research and document review he performed is fair and reasonable. 100

With respect to CIR, its normal billing rates are excessive given its function at trial and on appeal. Considering the evidence of prevailing market rates in the record, as well as the Court’s experience in these kinds of cases, a rate higher than $200 per hour for any attorney who did not actively participate at the trial (in terms of presenting witness or opening and closing statements) or on appeal (in terms of oral argument) would be unreasonable. Accordingly, an hourly rate of $200 for McDonald and Rosman, the two most experienced CIR lawyers, is appropriate. A rate of $175 per hour is appropriate for Mulloy, Shea, and Troy, the mid-level lawyers, and a rate of $150 per hour is reasonable for Bader, the least experienced lawyer at CIR.

GDC’s billing rates are extreme even with respect to the evidence GDC itself offered. For instance, GDC’s evidence establishes that no firm in Washington, D.C. charges an hourly rate for associate work over $230, and yet GDC requests that such work be compensated at a rate of $275 per hour. See Leroy, 906 F.2d at 1079 (“[T]he hourly rates are to be adequate to attract competent counsel, ... not the rates which lions at the bar may command.”) (internal quotations omitted). Indeed, such exorbitant rates are particularly inappropriate when “the experience and special skill of the attorney does not result in the expenditure of fewer hours than counsel would normally be expected to spend.” Id. The Court assesses a rate of $225 per hour for Olson and Cox, the two most experienced and distinguished lawyers from GDC and the two attorneys who controlled the appeal. The Court finds an hourly rate of $175 for Hungar and Scott and an hourly rate of $150 for Nelson is appropriate in light of their contribution to the case (primarily legal research and assisting with the trial record and the briefs) and their levels of experience.

Finally, rather than assign a specific rate for each paralegal and law clerk from WHS, CIR and GDC who worked on the case, the Court has aggregated the hours for each firm. The Court concludes a reasonable rate for work performed by paralegals and law clerks from GDC is $80 per hour; for WHS and CIR paralegals, a reasonable rate is $70 per hour.

(L) Compensation for Delay in Payment

In assessing a reasonable rate, the Court considered the attorneys’ current hourly rates (as opposed to the historic rates submitted with the 1994 fee applications) in order to compensate for the delay in payment. See Missouri v. Jenkins, 491 U.S. 274, 282-84, 109 S.Ct. 2463, 2468-69, 105 L.Ed.2d 229 (1989) (“[A]n appropriate adjustment for delay in payment — whether by the application of current rather than historic hourly rates or otherwise — is within the contemplation of [§ 1988].”); In re Washington Public Power Supply Sys. Sec. Litig., 19 F.3d 1291, 1305 (9th Cir.1994) (“Full compensation requires charging current rates for all work done during the litigation, or by using historical rates enhanced by an interest factor.”). Where appropriate (as with Wallace, Shea, Troy, Wilson, and paralegal work), the Court increased the original hourly rate to compensate for the delay in payment. In this case, however, the Court found that many of the requested billing rates were excessive. In those instances, the Court assessed hourly rates which, although lower than the rates requested, are intended to reflect current hourly rates in the prevailing market.

(1) Trial Hours

Table I represents the total hours for which the plaintiffs’ are entitled to compensation for hours spent at the trial level, with the appropriate deductions identified in Part V(A) for time spent on public and media relations, the proposed interventions, travel, duplicative work product, and lack of billing judgment. Table II multiplies the hours allowed by the billing rate previously determined by the Court in Part V(B) and takes into account the fifteen percent deduction for the plaintiffs’ lack of success in attaining specific injunctive and monetary relief.

[[Image here]]

[[Image here]]

Aggregating the individual lodestar amounts of Wallace, Harris, Wheeler, Wilson, and the WHS Clerk results in a total lodestar amount for WHS of $105,909.84. Aggregating the individual amounts of McDonald, Rosman, Mulloy, Troy, Shea, and the CIR Clerks results in a total lodestar amount for CIR of $122,843.70.

(2) Appellate Hours

Table III represents the total hours for which the plaintiffs’ are entitled to compensation for time spent on appeal, with the appropriate deductions identified in Part V(A) for hours spent on public and media relations, the proposed interventions, travel, the issues on remand, duplicative work product, and lack of billing judgment. Table IV multiplies the appropriate number of hours by the billing rate previously determined by the Court in Part V(B).

[[Image here]]

Aggregating the individual lodestar amounts of McDonald, Rosman, Troy, Bader, and the CIR Clerks results in a total lodestar amount for CIR of $60,861.00. Aggregating the individual amounts of Olson, Cox, Hun-gar, Scott, Nelson, and the GDC Clerks results in a total lodestar amount for GDC of $136,217.00.

The plaintiffs do not request a multiplier of the lodestar, see Plaintiffs’ Motion for Attorneys’ Fees, August 29, 1994, and the Court does not believe adjustment of the lodestar is warranted. This case involved extensive discovery and has spanned over five years. The legal issues, however, were neither novel nor extraordinarily difficult, as there have been many significant challenges to affirmative action since Bakke, both in and out of the context of higher education. Perhaps the most difficult aspect of this ease was the comprehensive marshaling and analysis of equal protection case law in the context of affirmative action. The Court is confident, however, that the difficulties of the case are accurately reflected in the lodestar amount.

The Court realizes that, at first blush, the reductions may seem excessive. Frankly, that impression is due to the sloppy state of most of the fee applications and the considerable overstaffing of this case, both of which necessitated significant reductions by the Court. In fact, the fee applications submitted by GDC, WHS, and Terral Smith were so vague and uninformative the Court seriously considered not awarding attorneys’ fees to those individuals at all. In any event, the Court conducted its own calculations regarding the number of hours it should have taken to prepare, try, and appeal this case. Based on an evaluation of the work performed by counsel, the undersigned’s thirty years of trial experience, and the Johnson factors, the Court concludes that the lodestar is a fan-estimate of an appropriate fee award. A single law firm spending an average of fifty hours per week for an entire year could have easily prepared the case and tried it. Had the same law firm handled the appeal, the case could have been prepared and argued to the Fifth Circuit and the Supreme Court in less than six hundred hours (an average of forty hours per week for fifteen weeks). Assuming an average rate of $200 per hour at trial and $225 per hour on appeal, the case should have cost around $650,000. The lodestar exceeds that amount by over $50,000.

E.

The defendants object to the expenses sought in this ease associated with the plaintiffs’ opposition to the proposed inter ventions and the costs relating to the involvement of outside counsel to try the case. 101 These expenses include travel expenses, lodging, phone calls, mailing expenses, and Fifth Circuit fees relating to the appeal of the motions to intervene. The Court agrees these charges are inappropriate under § 1988 and makes a resulting deduction of $19,475.09 in CIR’s first fee application. As for the post-appeal applications, the Court was able to identify $246.25 spent by GDC in relation to NEXIS searches; presumably, these expenses relate to GDC’s efforts with the media and therefore are not recoverable. CIR’s post-appeal records specifically label $955.70 spent on “public relations” which the Court will likewise' exclude. However, given the general nature of the research, printing, mailing, and traveling expenses submitted by CIR, it was impossible for the Court to assess a specific dollar amount to be excluded from CIR’s post-appeal fee application. Therefore, the Court reduces all such expenses by fifty-five percent ás a reasonable approximation of the costs related to opposing the intervention attempts. The Court also deducts $18.72 from Steven Smith’s application for charges incurred after July 1996, which charges presumably relate to the issues on remand.

Tables V and VI represent the amount of expenses requested by the plaintiffs in the trial and appellate levels of this case, respectively, with the appropriate deductions made in accordance with the paragraph above. The 'interest is calculated using the prime rate to compensate the plaintiffs for the delay in payment. See Alberti v. Klevenhagen, 896 F.2d 927, 938 (5th Cir.) (“[T]he appropriate rate of interest to be used in compensating a delay in payment adjustment is the cost of borrowing money, the prime rate.”), vacated in part on other grounds, 903 F.2d 352 (5th Cir.1990). The interest in Table V is calculated from August 29,1994, the date the plaintiffs filed their initial applications for attorneys’ fees, until the date of this order. The interest in Table VI is calculated from October 1, 1996, the date the plaintiffs filed their supplemental applications for attorneys fees, until the date of this order. 102

F.

Table VII represents the final calculation of total fees, expenses, and interest the Court awards in this case for each lawyer and/or law firm.

[[Image here]]

The plaintiffs shall have post-judgment interest of 5.407% to be calculated from the date of the judgment to the date of the payment of the judgment. The rate of interest is based on the average investment rate (equivalent coupon issue yield) according to the auction of the 52-week Treasury Bills by the United States Department of the Treasury.

VI.

The outrageous requests for damages made by three of the plaintiffs in this ease illustrate the significant financial risks federally funded state universities face when routine admissions decisions are challenged. To the extent minority and nonminority applicants share similar test scores — in which case subjective judgments will always inform the admissions process — there is the risk that the rejected applicant will blame his or her nonadmission on race and sue for damages in court. In the end, the determination of who deserves an offer of admission and who does not is left in the less-than-capable hands of people outside the academic arena— judges. Moreover, the specter of large compensatory damage awards could threaten to compromise the integrity of the admissions process itself. Public universities deserve the freedom to make the necessarily difficult choices regarding admission, and part of that endeavor entails not only consideration of the individual applicant but also the needs of higher education in general and the educational institution and class in particular. This case certainly does not make these serious responsibilities any easier.

JUDGMENT

BE IT REMEMBERED on the 20th day of March 1998 the Court entered its memorandum opinion consisting of its findings of fact and conclusions of law in the above-captioned matter and, consistent with those findings and conclusions, enters the following judgment:

IT IS ORDERED, ADJUDGED, and DECREED that the plaintiffs Cheryl J. Hop-wood, Douglas W. Carvell, Kenneth R. Elliott, and David A. Rogers have judgments of and against the University of Texas at Austin in the amount of One Dollar ($1.00) each, and further that the University of Texas at Austin as the University of Texas School of Law and its officers in their official capacities are hereby enjoined from taking into consideration racial preferences in the selection of those individuals to be admitted as students at the University of Texas School of Law.

IT IS FURTHER ORDERED, ADJUDGED, and DECREED that the plaintiffs Cheryl J. Hopwood, Douglas W. Carvell, Kenneth R. Elliott, and David A Rogers TAKE NOTHING further in their causes of action against the defendants the State of Texas; the University of Texas Board of Regents; Bernard Rapopart, Ellen C. Temple, Lowell H. Leberman, Jr., Robert Cruiksharik, Thomas O. Hicks, Zan W. Holmes, Jr., Tom Loeffler, Mario E. Ramirez, and Martha E. Smiley, as members of the Board of Regents in their official capacities; and Robert M. Berdahl, in his official capacity as President of the University of Texas at Austin.

IT IS FURTHER ORDERED, ADJUDGED, and DECREED that the plaintiffs shall have and recover against the University of Texas at Austin the sum of $703,992.29 in attorneys’ fees; costs in the amount of $58,-135.50 incurred as of August 29, 1994, which includes an interest rate of 8.47 percent per annum (simple) from August 29, Í994, to the date of this judgment; and costs in the amount of $13,632.52 incurred as of October 1, 1996, which includes interest of 8.38 percent per annum (simple) from October 1, 1996, to the date of this judgment, for which let execution issue.

IT IS FURTHER ORDERED, ADJUDGED, and DECREED that this judgment shall bear. interest of 5.407 percent .from the.date of this judgment until paid.

IT IS FINALLY ORDERED, ADJUDGED, and DECREED that all other relief requested by any party herein is DENIED as this is a final judgment.

1 . Sections 1981 and 1983 are thé vehicles by which the plaintiffs assert violations of the Equal Protection Clause of the Fourteenth Amendment. Title VI provides in relevant part:

No person in the United states shall, on the ground of race, color, or national origin, be excluded from participation in, be denied the benefits of, or be subjected to discrimination under any program or activity receiving Federal financial assistance.

42 U.S.C. § 2000d.

2 . Many of the University of Texas officials sued by the plaintiffs in 1992 are no longer associated with the University of Texas system and/or the law school. Defendants Bernard Rapopart, Ellen C. Temple, Lowell H. Leberman, Jr., Robert Cruikshank, Thomas O. Hicks, Zan W. Holmes, Jr., Tom Loeffler, Mario E. Ramirez, and Martha E. Smiley were sued in their official capacities as members of the University of Texas Board of Regents. Defendant Robert M. Berdahl was sued in his official capacity as President of the University of Texas at Austin; the current interim President is Peter T. Flawn. Defendant Mark Yudof was sued in his official capacity as the Dean of the University of Texas School of Law; M. Michael Sharlot, an Assistant Dean of the law school in 1992, has since replaced Yudof as Dean. Defendant Stanley M. Johanson was sued in his official capacity as Chair of the University of Texas School of Law Admissions Committee and remains in that capacity to this date.

3 . The 1992 admissions procedures favored only black and Mexican-American applicants. Other minority groups, including Hispanics not of Mexican-American descent, were not granted preferences in admission. In this opinion, the Court loosely uses the term "minority” to refer to blacks and Mexican-Americans; the term "non-minority” is used to refer to whites and non-preferred minorities.

4 . In the Title VII arena, the Supreme Court has developed a burden-shifting scheme to be used when there is no direct evidence of an employer’s racial animus. See, e.g., Texas Dep't of Community Affairs v. Burdine, 450 U.S. 248, 101 S.Ct. 1089, 67 L.Ed.2d 207 (1981); McDonnell Douglas Corp. v. Green, 411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973). The employee bears the initial burden of establishing a prima facie case of discrimination. See Burdine, 450 U.S. at 253-54, 101 S.Ct. at 1094. If successful, the burden of production shifts to the defendant to establish legitimate, nondiscriminatoiy grounds for the adverse employment action. See id. at 254-55, 101 S.Ct. at 1094-95. If such grounds are shown, the burden shifts back to the employee to establish that the reasons proffered by the defendant are merely pretext for unlawful discrimination. See id. at 255-56, 101 S.Ct. at 1095. The burden of persuasion remains on the employee at all times. See id. at 253, 101 S.Ct. at 1093. In this case, the Court held that the plaintiffs established a prima facie case of discrimination by proving that the 1992 admissions process was constitutionally flawed. See Hopwood I, 861 F.Supp. at 580. The plaintiffs failed to persuade the Court, however, that the law school's proffered reasons for. the plaintiffs’ non-admission were pretextual. See id. at 581-82.

5 . In view of the unequal treatment to which the plaintiffs were subjected, however, the Court granted nominal damages in the amount of one dollar to each plaintiff and allowed the plaintiffs to reapply to the law school in 1995 without incurring any administrative costs or fees. See id. at 582-83.

6 . Over the strenuous and critical objections of Chief Judge Politz and Circuit Judges King, Weiner, Benavides, Stewart, Parker, and Dennis, the Fifth Circuit declined to reconsider the case en banc. See Hopwood v. State of Texas, 84 F.3d 720 (5th Cir.1996).

7 . Finding no compelling state interest, the Fifth Circuit did not reach the second inquiry of whether the affirmative action program was narrowly tailored. See id. at 955.

8 . The Court need not reconsider the issue of punitive damages, as the Fifth Circuit upheld the Court's determination that punitive damages were inappropriate in this case. See id. at 959.

9 . The 1992 admissions procedures and the evolution of the law school’s admissions procedures since the 1960s are explained in greater detail in the Court's first opinion, see Hopwood I, 861 F.Supp. at 557-563, as are the plaintiffs’ qualifications for admission to the law school, see id. at 564-567.

10 . Slightly less than half of the applications the law school received were submitted by Texas residents. See Hopwood I, 861 F.Supp. at 561 n. 22. In 1992, the University of Texas Board of Regents required an entering law school class of at least 500 students, and Texas law mandated that at least 85% of the entering class be Texas residents (the latter figure has since decreased to 80%). See Hopwood I, 861 F.Supp. at 563; Hopwood II, 78 F.3d at 935 n. 2.

11 . Although members of the minority subcommittee were supposed to review nonminority files as well, testimony from the first trial established that Alemán did not review any nonminority files in 1992. See Hopwood I, 861 F.Supp. at 562.

12 . Because of the limited number of seats available for nonresidents, Johanson set the presumptive admission and denial lines at a higher level for nonresidents than for residents. See Hop-wood I, 861 F.Supp. at 561 n. 22.

13 . "Johanson's setting of these scores was a process that evolved over the course.of the admissions process based on the pool of applicants, the number of offers, and the number of acceptances. Initially, the numbers were set high and lowered as the yield from offers and composition of the entering class began to develop.” Hopwood I, 861 F.Supp. at 561 n. 24.

14 . During the preliminary review process in January 1992, however, Johanson reviewed minority and nonminority files together. See Hopwood I, 861 F.Supp. at 561 n. 23.

15 . The law school recently discarded the piles-of-thirty approach in favor of a smaller admissions committee comprising three members. All three members of the admissions committee review each application. See Wellborn, vol. 1 at 51, 246-47. (All references to the trial transcript in this opinion are from the second trial unless otherwise noted.)

16 . There were 18 stacks in the nonminority discretionary zone. Seventeen stacks contained 30 application files;. the remaining stack comprised only 16 files. Each committee member was required to screen 5 stacks. See Hopwood I, 861 F.Supp. at 562 n. 26 & 27.

17 . In 1992, LSDAS was in the process of changing the LSAT scoring system from a two-digit system to a three-digit system. See Hopwood I, 861 F.Supp. at 561 n. 25; Hopwood II, 78 F.3d at 936 n. 5. For sake of consistency with the Fifth' Circuit decision, the TI scores reported in this opinion reflect only the three-digit scoring system. See Hopwood II, 78 F.3d at 936 n. 5. With respect to LSAT scores, the Court refers to the applicants’ percentile rank on the exam rather than to the applicant’s two- or three-digit score in order to draw a more accurate comparison among the applicants.

18 . Carvell took the LSAT twice, once under the two-digit system and again under the three-digit system. He scored in the 61st percentile on his first exam and in the 91st percentile on his second exam. P-151. Carvell’s TI score of 197 only reflects his higher LSAT score in the 91st percentile. See Wellborn, vol. 1 at 71. The Court’s first opinion, which indicates that Carvell’s TI score of 197 reflects his averaged LSAT score, see Hopwood I, 861 F.Supp. at 566 n. 47, is in error.

19 . As explained in further detail in Part 111(B) infra, the law school reconsidered Elliot’s application for admission following Elliot’s initial rejection in April 1992.

20 . In his concurring opinion in Price Waterhouse v. Hopkins, 490 U.S. 228, 109 S.Ct. 1775, 1795, 104 L.Ed.2d 268 (1989), a case in which the Supreme Court applied the Mt. Healthy burden-shifting scheme to a Title VII sex discrimination case, Justice White explained the distinction between a "mixed-motives" and "pretext” case:

The Court'has made it clear that ‘mixed-motives’ cases, such as the present one, are different from pretext cases such as McDonnell Douglas and Burdine. In pretext cases, the issue is whether either illegal or legal motives, but not both, were the true motives behind the decision. In mixed-motives cases, however, there is no one ‘true’ motive behind the decision. Instead, the decision is a result of multiple factors, at least one of which is legitimate____ [T]he showing required by Mt. Healthy does not improperly shift from the plaintiff the ultimate burden of persuasion on whether the defendant intentionally discriminated against him or her.

Id. at 260, 109 S.Ct. at 1796 (citations and quotations omitted); see also Holo-Krome Co. v. N.L.R.B., 954 F.2d 108, 110 (2nd Cir.1992) ("[P]retext analysis determines what the true motivation actually was; dual motivation analysis determines what the employer's conduct would have been if the improper motivation had not been present.”).

21 . The Supreme Court in Adarand implicitly recognized that an individual's inability to compete on an equal footing as a result of discriminatory classifications is a harm that is related to the issue of standing. In holding that the plaintiff had standing to challenge the use of racial preferences in the awarding of subcontractor bids in federal agency contracts, the Supreme Court stated:

We note that ... [the plaintiff] Adarand need not demonstrate that it has been, or will be, the low bidder on a government contract. The injury in cases of this kind is that a discriminatory classification prevents] the plaintiff from competing on an equal footing. The aggrieved party need not allege that he would have obtained the benefit but for the barrier in order to establish standing.

Adarand, 515 U.S. at 211, 115 S.Ct. at 2105 (citations and quotations omitted).

22 . For that reason, International Brotherhood of Teamsters v. United States, 431 U.S. 324, 97 S.Ct. 1843, 52 L.Ed.2d 396 (1977) is inapplicable as well. Teamsters was a Title VII disparate impact case in which the Supreme Court held that once the employee establishes a system-wide pattern or practice of discrimination, the burden shifts to the employer to prove that individual hiring decisions were not made in pursuit of that policy. See id. at 359-60, 97 S.Ct. at 1866-67. In that case, however, the Supreme Court specifically supported the burden-shifting scheme on the ground that an employee's proof of broad-based employment discrimination creates “a greater likelihood that any single [employment] decision was a component of the overall pattern,'' id. at 361 n. 45, 97 S.Ct. at 1867 n. 45, and, consequently, there are "reasonable grounds to infer that individual hiring decisions were made in pursuit of the discriminatory policy,” id. at 359, 97 S.Ct. at 1867. No similar inference can or should be made in this case: a policy that affects 7% of a given population can hardly be called broad-based or widespread, nor would it cause one to infer that an individual decision was infected by the discriminatory policy.

23 . Furthermore, as a matter of public policy, it may be imprudent to apply the Mt. Healthy burden-shifting framework to the admissions procedures of public universities. In making its admissions decisions every year, the law school is faced with the enormous responsibility and discretion to select an entering class among applicants who share similar academic qualifications (in terms of GPAs and LSAT scores), but who have varying personal backgrounds and vastly different educational circumstances. Public universities are put in the unfortunate, perhaps even untenable, position of making such decisions with the threat of liability hanging in the background.

24 . The plaintiffs did not proffer their own expert to testify as to what a constitutional admissions system would look like or whether any of the plaintiffs would have been admitted under such a system.

25 . The statistical breakdown for the 1992 resident admissions occurred as follows:

TI score Nonpreferred Residents Admitted Nonpreferred Residents Denied ) Nonpreferred Residents Admitted Minorities Admitted

203 up 169 1 -100% 4

202 27 3 90% 2

201 48 5 90% 1

200 51 4 92% 1

199 57 7 89% 4

198 34 15 69% 2

197 44 27 59% 1

196 29 44 2

195 24 40 38% 3

194 15 64 19% 7

193 13 59 18% 5

192 10 57 15% 3

191 4 72 5% 5

190 5 71 7% 11

189 2 63 3% 3

188 1 57 2% 4

187 3 55 5% 9

186 2 50 4% 10

185 3 393 0% 19

down

26 . The percentages in Wellborn's report, however, are not hard-and-fast figures, as there were small variations in the percentages from year to year. Furthermore, the 1992 percentages varied from other years because LSDAS was in the process of changing the LSAT scoring system from a two-digit score to a three-digit score. Under the two-digit system, applicants were more compressed along the Texas Index. Because this methodology relies too heavily on unreliable percentages of admission, the Court finds the methodology employed in Wellborn’s supplemental report preferable.

27 . It was the practice of the law school to lower the presumptive admission and/or denial scores in order to generate more offers. See Hopwood I, 861 F.Supp. at 561-62; Hopwood II, 78 F.3d at 937 n. 10. Wellborn did not alter the number of votes per pile for discretionary zone applicants because doing so, he testified, would have yielded too many offers.

28 . In 1992, the law school extended 637 resident offers to achieve 425 matriculants. Therefore, the overall resident yield rate, i.e., the rate at which candidates accepted their offers of admission, was approximately 67%. See Johanson Declaration, D-332; Hopwood I, 861 F.Supp. at 561 n. 22. The 1992 yield rate was approximately 80% for African-Americans and 73% for Mexican-Americans. Wellborn assumed the yield rate for marginal nonminority candidates in a color-blind admissions process would be higher than 67% (and probably as high as 75%) because the additional offers of admission to nonminorities would be made to less qualified (and therefore less desirable) applicants for whom the law school would probably be their first choice. See Wellborn, vol. 1 at 152-53.

29 . Wellborn’s projected statistical breakdown in a race-blind admissions procedure:

TIscore % Nonpreferred Admitted — Actual Total Applicants % Admitted-Projected Projected Number Admitted

203 up -100% 174 100% 174

202 90% 32 100% 32

201 90% 54 90% 49 + 1

200 92% 56 90% 50 + 1

199 89% 68 90% 61 + 1

198 69% 51 90% 46 + 1

197 59% 72 70% 50 + 1

196 40% 75 60% 45 + 1

195 38% 67 40% 27 + 1

194 19% 86 40% 34 + 1

193 18% 77 20% 15

192 15% 71 20% 14

191 5% 81 15% 12

190 7% 87 5% 4

189 3% 68 5% 3

188 2% 63 5% 3

187 5% 71 5% 4

186 4% 65 5% 3

185 - 0% 63 5% 3

184 down 0% 658 0% 0

Adjusting the presumptive admit and deny lines would not have accounted for all of the offers that would have been made assuming the same yield rate. Eight offers would still have to be made, and Wellborn accounted for them by assuming that one candidate from each TI score between 201 and 194 would have been chosen from a waiting list.

30 . Statistically, Wellborn estimated that six additional offers would have been made at this index. He concluded, however, that the minority admit-tee might not have been admitted at a TI score of 197. Wellborn therefore projected seven additional offers.

31 . For purposes of stacking application files into piles of thirty, the law school classified Carvell as a TI 197 applicant. Carvell’s "true” TI score, using his combined LSAT score in the 76th percentile, was either 191 or 192. See Wellborn, vol. 1 at 70. Although it was and is the normal practice of LSDAS to average multiple scores, LSDAS did not do so in 1992 for candidates who received LSAT scores under both systems. See Wellborn, vol. 1 at 70-71. For purposes of reviewing Carvell's application, Wellborn testified that reviewers would have mentally discounted Carvell’s TI score and would do so again in a color-blind process. See Wellborn, vol. 1 at 73; see also Declaration of Steven Goode D-336 (stating that he would discount Carvell's attempts to explain his poor performance on his first LSAT since applicants have the option of canceling their scores). At trial, Carvell objected to Wellbom’s circular argument in his first report that Carvell compares unfavorably to other TI 197 applicants in part because "Carvell does not really belong at this index.” See D-519. For purposes of his first report, Wellborn classified Carvell with other TI 197 applicants because it was Carvell’s actual classification in 1992. In his supplemental report, however. Wellborn compares Carvell to the entire pool of discretionary zone applicants, including applicants with TI scores of 191 and 192.

32 . As Wellborn testified, the fact that he only evaluated LSDAS sheets with TI scores of 190 or above was an omission that probably benefitted the plaintiffs. See Wellborn, vol. 1 at 50. Well-born acknowledged that "one or two” candidates with TI scores below 190 might have been "extraordinary” enough to compensate for their low index. Id. To the extent that Wellborn did not consider those applicants — or, for that matter, the 250 applicants within the hypothetical discretionary zone whose entire application files might have revealed extraordinary qualities meriting the applicant’s admission — Wellborn eliminated from the applicant pool a few potential competitors who probably would have been admitted.

33 . The TI scores calculated by LSDAS in 1992, on the other hand, gave slightly greater weight to an applicants’s LSAT score. See Hopwood II, 78 F.3d at 935 n. 1 (indicating that LSDAS assigned a 60% weight to a candidate’s LSAT score and a 40% weight to his or her GPA in 1992). Of course, the law school is not bound by LSDAS's determination that the LSAT score should be weighed more heavily in determining admission.

34 . The college LSAT mean is the average LSAT score of students who took the LSAT from that particular college or university. See Wellborn, vol. 1 at 44. In Wellborn’s opinion, "the college LSAT mean is the best way of comparing colleges in terms of the caliber of the student body.” See D-520. The numbers bear out Wellborn’s assessment. The 1992 LSAT means of the following universities illustrate the difference: Harvard University (40); Duke University (39); Rice University (38); Trinity University (36); the University of Texas (between 35 and 34); Texas A & M University (33); Hendrix College (32); California State University-Sacramento (28); and the University of Houston-Downtown (26).

35 . Wellborn initially identified 18 minorities who he believed would have been admitted in a race-neutral system. The file of one minority applicant was inadvertently omitted from the original report; Wellborn reviewed that file and concluded that the applicant would have been admitted. See Wellborn, vol. 1 at 89-90 and D-669 (indicating that the applicant graduated in the 98th percentile at Texas A & M University and received an LSAT score in the 96th percentile). Therefore, the original calculation of 78 additional offers is erroneous.

36 . Wellborn segregated the applicants by race into Groups A, B, C, and D for administrative convenience only. When evaluating the application files, he did not review the minority and nonminority files separately.

37 . Although the plaintiffs openly questioned Wellborn’s integrity and ability to render an unbiased opinion at trial, they did not contend that Wellborn was unqualified to render an expert opinion regarding the plaintiffs’ qualifications for admission vis-a-vis other denied nonminority applicants. And, other then the nonsensical objection that Wellborn is not a mathematician or statistician (discussed infra ), the plaintiffs do not argue that Wellborn is incapable of assessing the plaintiffs’ chances of admission under a constitutional admissions system.

38 . Johanson recognized this difficulty as well. When asked in his 1993 deposition whether Hop-wood’s chances of admission would have been improved in the absence of the law school’s affirmative action program, Johanson testified:

I haven’t got a clue____I can’t say, if she was in another stack, whether she would have gotten two votes to admit or no votes to admit .... What was it in that file that impressed one screener and didn’t impress two? Was it the makeup of the overall competition? What was it? There is no way that — it would be not only sheer speculation, but I think it would be improper to speculate because it’s just based on the misunderstanding of the relativity of every decision.

Johanson, vol.l at 256-57.

The plaintiffs have attempted to use Johanson’s testimony to demonstrate that the law school cannot establish by a preponderance of the evidence that the plaintiffs would not have been admitted under a constitutional system. Well-born testified, however, that the validity of his predictions is not dependent on whether the admissions procedure uses stacks of thirty or whether one reviewer or group of reviewers evaluates all of the application files. See Wellborn, vol. 1 at 251-52. Likewise, Sharlot testified that, although the piles-of-thirty approach had a "certain degree of artificiality” (given the limited number of choices each reviewer could make), the voting patterns of the faculty members on the admissions committee was not aberrational or unpredictable:

I think that as the committee, which maintained its personnel to a very large extent [for a very long time], as the committee worked over many years, there came to be a fairly high degree ... of shared agreement as to which criteria were more pertinent and the weights that should be given to them. And that is why I believe it is possible to undertake a review of hundreds of files and to make a reasonable prediction as to the likelihood, given the make up [sic ] of the committee, who was most likely to receive an offer of admission. Sharlot, vol. 2 at 22-23. Using the piles of thirty was, in the Court’s opinion, a flawed approach. However, the approach was intended to produce the most qualified candidates for admission, and that was the same goal Wellborn attempted to achieve in his expert reports.

39 . In 1992, the application files in the discretionary zone were separated according to whether the applicant had a two-digit or three-digit LSAT score. See Wellborn, vol. 1 at 161. Although two-digit and three-digit candidates were, for the most part, compared separately in 1992, Well-born did not separate the two groups in preparing his reports. Wellborn, vol. 1 at 159. The plaintiffs complain that Wellborn’s failure to segregate applicants by their LSAT score is another fundamental flaw in his methodology. In view of the fact that Wellborn did not attempt to replicate the 1992 admissions system, the Court finds that Wellborn’s failure to evaluate two-digit and three-digit candidates separately is inconsequential.

40 . Wellborn testified that 123 wait list letters were sent to applicants and that the "vast majority” of the waiting list offers were accepted. See Wellborn, vol. 1 at 55-56. Therefore, the actual number of individuals on the waiting list was somewhat smaller than 123.

41 . Wellborn explained the fact that committee members were required to cast a certain number of votes per pile meant that in some instances the last vote was "forced.” In Wellborn's words, it would have been “crazy” to construct a process in advance in which anyone who received a forced vote in the screening process would automatically be admitted. See Wellborn, vol. 1 at 54.

42 . At the remand trial, Hamilton testified as follows:

And the Hopwood file I remember in particular having read the file and agonizing over her personal situation with respect to her child. And then I returned my files to the admissions office, I checked to — I did not vote for Ms. Hopwood at all. I checked to see if other members had voted for her, and they did not. And out of sympathy for her case, again, which I had some concern about, I changed my vote to put her on the waiting list and took a vote from someone else who had already gained admission.

Hamilton, vol.- 2 at 27. Although Professor Johanson had instructed committee members to conduct blind voting, Hamilton did not follow this instruction in 1992. She testified she was concerned that her voting had been “particularly aberrational” compared to other members of the admissions committee; she therefore reviewed the voting patterns of other faculty members "to try to get a sense of what was valuable.” Hamilton, vol. 2 at 42-44.

43 . Waiting list candidates were moderately overrepresented in Wellborn's C and D Groups, suggesting some correlation between placement on the waiting list and an increased probability of admission. See 'Wellborn, vol. 1 at 211. Of the 98 Group C and D selections, 44 received one vote in discretionary zone screening in 1992. In the Court’s opinion, this correlation is not strong enough to support the inference that all individuals on the waiting list would be more likely to receive offers of admission in a race-neutral world.

44 . According to Wellborn’s supplemental report, these individuals had TI scores of 202 and 198, respectively, and thus would probably have been presumptively admitted under a constitutional system. SeeD-521.

45 . In Group A, one applicant graduated in the 95th percentile of her class from Washington University, a college with an LSAT mean score of 36, see D-5-33; another applicant graduated from the University of Texas in the 90th percentile of her class, see D-539. In Group C, two applicants graduated from Trinity University where they were in the 90th and 86th percentiles of their classes, see D-567 and D-583; another applicant graduated in the 94th percentile of her class at George Washington University, a 33 LSAT mean school, see D-581.

46 . In addition to the Berkeley and Plan II applicants described supra (D-524 and D-529), one applicant from Group C graduated from Rice with a GPA of 3.40 and had an LSAT score in the 83rd percentile. That applicant had also completed one semester at the LBJ School of Public Affairs at the University of Texas with a GPA of 3.77. See D-547. Another Group C applicant had a GPA of 2.99 from Rice with a degree in architecture and an LSAT score in the 86th percentile; the applicant also earned a master’s degree from Princeton and had been working as an architect for many years. See D-605. The final Group C applicant was an engineering major from Southern Methodist University with a 3.07 GPA and had an LSAT score in the 90th percentile. See D-594.

47 . A small number of applicants attended undergraduate institutions that were too small to provide the statistical sampling needed for LSDAS to compile the college LSAT mean. In those instances, Wellborn looked at a "ribbon" showing the distribution (in terms of percentile rank) of LSAT scores by students who took the LSAT from that institution. See Wellborn, vol. 1 at 44, 103-04.

48 . This criticism, in the Court's opinion, is puzzling coming from the plaintiffs. At the first trial, the plaintiffs repeatedly emphasized the relative strength of their TI scores — a purely objective factor that did not account for the quality of the applicant's undergraduate institution or curriculum — as compared to minority admittees. See, e.g., Hopwood, vol. 2 at 195.

49 . Carvell contends that improvement in grades is an “important” criteria simply because the defendants listed this factor first in their answers to interrogatories regarding subjective criteria to be considered in admissions. This argument is absurd. Likewise, the Court rejects the plaintiffs’ argument that reviewers must grant preferential consideration to subjective factors such as age or work experience because the admissions committee lists such factors as criteria to consider in determining whether a particular applicant should be admitted. For instance, Johanson (unlike Hopwood's other reviewers) reacted negatively to the fact that Hopwood had worked substantially during college; he viewed it as evidence she earned her undergraduate degree on "a slow track.” See 1994 trial transcript, Johanson, vol. 5 at 15-16. The Court declines Hop-wood’s invitation to find that Johanson abused his discretion in failing to consider her work experience a "plus” factor.

50 . Although this applicant, like Hopwood, had a significant number of hours from a junior college, he completed his associate’s degree in two and a half years, had a distinguished military career in the Marine Corps, had a strong personal statement in which he recounted overcoming an abusive childhood, and, importantly, transferred to a challenging undergraduate institution (the University of Texas) following junior college. He also had several strong faculty letters of recommendation. See D-538; Wellborn, vol. 1 at 99.

51 . Hopwood and Carvell also identify three flies (D-605, D-619, and D-632) they contend should not be considered under a hypothetical admissions system because they were untimely applications. In fact, only two applications (D-605 and D-632) were considered untimely; the other application (D-619) lacked a transcript, an error that was attributed to the undergraduate institution, not the applicant. The evidence establishes the law school nevertheless considered all three applications on their merits in 1992, and the Court therefore sees no reason to exclude them from consideration in this case. Even if the Court were to exclude these files from consideration, the Court’s conclusions regarding the plaintiffs’ chances for admission would not be disturbed.

52 . Indeed, at trial the plaintiffs identified only 4 such applicants out of the 936 applicants who were offered admission in 1992 — less than 0.5%. See Sharlot, vol. 2 at 3-6; Wellborn, vol. 1 at 221.

53 . It might be the case, for instance, that one or more of these individuals were admitted because their parents were influential alumni or generous benefactors of the University of Texas, a practice specifically condoned by the Fifth Circuit. See Hopwood II, 78 F.3d at 946 ("An admissions process may also consider an applicant's home state or relationship to school alumni.”).

54 . In 1992, Johanson reviewed Hopwood’s application file twice — first when he pulled her application from the presumptive admit zone and a second time as a part of the initial waiting list screening. See Johanson Declaration D — 332.

55 . Hopwood contends the law school should he equitably estopped from arguing that the absence of a personal statement and letters of recommendation in her file weakened her application because of written and oral statements allegedly made by representatives and employees of the law school. Hopwood claims that in May 1991 and January 1992 she spoke to unnamed individuals in the law school’s admissions office at length about the information required to complete her application. Hopwood claims these individuals told her that, based on her GPA and LSAT score, she would be within the pool of applicants who would be "automatically admitted.” Hopwood claims that these representations were confirmed by the law school's 1992 bulletin, which indicated that "approximately 55% of the available admissions are automatically granted” to candidates reflecting "outstanding” GPAs and LSAT scores, see P-172. Based on these statements, Hopwood declined to “embellish" her application. This contention is so frivolous it barely merits discussion. Even assuming the statements were made and that the law school and the State of Texas should be bound by them, the Court finds that any reliance on those statements was unreasonable. Hopwood did not know what the presumptive admission line would be prior to submitting her application; in fact, the line was not lowered to include Hop-wood’s application until March 1992, well into the admissions process.

56 . Although Elliot provided two professional letters of recommendation, Johanson described them as "not overwhelmingly impressive.” See Johanson Declaration D-332. Additionally, Well-born testified that members of the admissions committee generally tend to focus more heavily on letters of recommendation from faculty members as opposed to employers because faculty members "are usually commenting on students in an academic environment, which is what we are looking for.” Wellborn, vol. 1 at 74.

57 . For the record, the law school’s 1992 admissions procedures did not impose quotas or grant admissions preferences on the basis of sex. Mr. Elliot was mistaken.

58 . Admittedly, the Court used a poor choice of words when it wrote, "Elliot was not notified of his admission to law school.” See Hopwood I, 861 F.Supp. at 566. The Court recognizes that the sentence could lead one to the implicit conclusion that an offer of admission had been contemplated by the law school but simply not communicated to Elliot. The Cotut did not intend for readers to reach that conclusion and regrets the error.

59 . The University of Houston-Downtown is a separate campus from the University of Houston-Main Campus, and the two campuses therefore have different LSAT college means (26 and 30, respectively).

60 . Rogers argues the fact that he was dismissed from the University of Texas as an undergraduate is not a valid reason to deny him admission to the law school. Rogers identifies one individual who also was dismissed from the University of Texas as an undergraduate but who was nevertheless admitted in 1992. See P-304. If anything, the individual identified by Rogers is the exception that proves the rule. See Gergen Declaration, D-335 ("[I]t would take a great deal of redemption to overcome the disastrous performance as an undergraduate at [the University of Texas].”). That applicant had a TI score of 204, received an LSAT score in the 99th percentile, and returned to the University of Texas at Austin where he participated in the history honors program. See P-304. The applicant graduated with a cumulative GPA of 3.10 and received almost all A’s after his return from academic dismissal. In addition, the applicant provided 3 exceptional faculty letters of recommendation. One professor describes the applicant as one of the 10 best of the 10,000 students he had ever taught. See P-304. Another describes him as better than the 2,500 students he had ever taught; the professor observed, "[o]n reflection, I wish I had the nerve to ask him to grade the course rather than take it.” See P-304. Frankly, Rogers's application does not compare.

61 . Because the Fifth Circuit gave no weight to the Court’s more conclusory finding in its first opinion that the plaintiffs did not prove their damages by a preponderance of the evidence, the Court finds it necessary in this opinion to provide very detailed and extensive findings regarding damages.

62 . The defendants likewise object to any damage award in light of the Court's factual finding at the first trial that the plaintiffs failed to prove any damages resulting from their rejections by the law school. Additionally, the defendants argue they cannot be held liable for monetary damages because the State of Texas enacted its admissions program in response to a court-ordered investigation of Texas's public higher education system by the United States Office of Civil Rights ("OCR”) and the Department of Education. In the late 1970s, the OCR found that Texas was in violation of Title VI for its failure to eliminate the vestiges of its former de jure segregation of public higher education. In an effort to comply with directives from the OCR, Texas instituted a plan to increase enrollment of blacks and Hispanics in public higher education. In June 1983, - after years of negotiations between the OCR and the State of Texas, the OCR conditionally found Texas in compliance with Title VI, but the OCR required Texas to continue to implement plans to increase minority enrollment. These plans paved the way for the continued receipt of federal funds for. public higher education in Texas. See Hopwood I, 861 F.Supp. at 555-57 (discussing the negotiations between the OCR and the State of Texas regarding Texas’s failure to comply with Title VI). Notwithstanding the factual validity of these two arguments, the Court complied with the instructions of the Fifth Circuit.

63 . As discussed more fully in Part IV(B) infra, Carvell was the only plaintiff who attended law school following his rejection by the University of Texas.

64 . Elliot and Rogers did not seek any mental anguish damages. See Elliot, vol. 3 at 269-71; P-322; Rogers, vol. 2 at 96-99, 101-11.

65 . This statement by the plaintiffs' lawyer is virtually an admission that the damages after the first trial result from the lawsuit itself, which damages, of course, are not compensable. The scope and measure of both economic and emotional damages should have been the same after the May 1994 trial and post-appeal.

66 . The defendants argue that the evidence Hop-wood, Carvell, and Rogers presented on damages is legally insufficient because they did not compute their future income stream using the below market discount rate method set forth in Culver v. Slater Boat Co., 722 F.2d 114 (5th Cir.1983), cert. denied, Reederei v. Byrd, 467 U.S. 1252, 104 S.Ct. 3537, 82 L.Ed.2d 842 (1984); the Court did not consider this argument because it was not factually persuaded that these plaintiffs suffered any economic damages.

67 . The Fifth Circuit overruled the Court’s determination that the defendants did not intentionally discriminate under Title VI. See Hopwood II, 78 F.3d at 957.

68 . The Court is not finding Hopwood incapable of passing the courses; to the contrary, to the degree Hopwood would have been able to take and complete her courses, the Court believes she would have received a passing grade.

69 . Ruhter relied on the David J. White Survey for attorney salaries in private law firms in Dallas, Houston, Philadelphia, Baltimore, and Washington, D.C. Ruhter gave special consideration to attorney salaries in Texas because many graduates of the law school remain in Texas following graduation. See Ruhter, vol. 2 at 254. He also considered attorney salaries from the mid-Atlantic region because that is where Hopwood is currently living. See Ruhter, vol. 2 at 253-54. For data on accountant salaries, Ruhter relied on the ROMAC Report.

70 . Ruhter did not examine whether Hopwood could have mitigated her damages by enrolling in law school in 1993, 1994, 1995, or 1996. See Ruhter, vol. 3 at 65.

71 . Even more troubling to the Court is Ruhter’s failure to account for the benefits of not being a lawyer. Presumably one of the reasons for the salary differential is that, on average, lawyers tend to work longer, harder hours than accountants and require more years of education. It seems inherently unfair to award Hopwood the lifetime salary differential between accountant and attorney earnings without even attempting to take into account such quality of life issues.

72 . Title IX of the Education Amendments of 1972, 20 U.S.C. §§ 1681-1688 (West 1994) ("Title IX”). Title IX prevents discrimination on the basis of sex by any education program or activity receiving federal funds. See 20 U.S.C. § 1681(a).

73 . Carvell was also "upset” and "bothered” by public statements made by the defendants and other state officials during the pendency of this case. See Carvell, vol. 4 at 108. These damages, if they exist, are not compensable.

74 . Lees-Haley defines a "stressor” as an experience, either positive or negative, "that usually involves change in the patient's life that’s great enough that it requires some coping or adjustment.” Lees-Haley, vol. 3 at 192.

75 . Employment in Dallas by private law firms (with their attendant high starting salaries) is one of the positive benefits of graduating from SMU Law School, as the school purposefully seeks to place its graduates in the Dallas law market.

76 . When specifically questioned by his counsel at the May 1994 trial whether he was claiming any loss in earnings as a result of his rejection by the law school, Carvell explicitly stated he was not. See 1994 trial transcript, Carvell, vol. 10 at 9-10.

77 . To the extent Carvell implies he might have been considered for partnership track at Bickel & Brewer had he graduated from the University of Texas, that proposition is ridiculous. One of the name partners of that firm is a graduate of SMU Law School, so it is highly unlikely that members of the firm regard SMU law alumni unfavorably. See Carvell, vol. 4 at 124; D-653.

78 . In any prestigious area of the law one examines, from top practitioners to Supreme Court Justices, few people know which law school the individual attended, and, presumably, not all such individuals are graduates of the University of Texas.

79 . At trial, Rogers testified he sustained losses of approximately $382,000. See Rogers, vol. 2 at 95-96; P-325. Because Rogers failed to supplement his interrogatory responses regarding the extent of his damages as required during discovery, the Court considered only his original request.

80 . This last contention — that the State of Texas should pay for Rogers’s poor business judgment — sounds like a bad joke and, perhaps, betrays a lack of analytical skills Rogers will sorely need if he intends to succeed in law school.

81 . The first set of attorneys’ fee applications were filed on August 29, 1994. One application was submitted on behalf of Steven Smith, Terral Smith, and WHS. Steven Smith sought compensation for 1,552 hours at $125 per hour plus $164.21 in expenses for a total of $194,164.21. Terral Smith sought compensation for 837.80 hours at $225 per hour for a total of $188,-505.00. WHS sought compensation for 1,005.82 hours plus $151.09 in expenses for a total of $171,104.34. The billing rates for WHS ranged from $95 to $200 per hour. The second application, filed on behalf of CIR, sought $235,461.95 in fees for 1,444.94 hours of work and $64,-526.19 in expenses, for total compensation of $299,988.14. CIR’s rates ranged from $125 to $230 per hour.

82 . In their supplemental requests for attorneys fees filed on October J, 1996, Hopwood and Carvell argue that the original request for fees should be adjusted in order to compensate for the delay in payment. The adjusted figures are $362,250.19 for CIR (representing current rates of $150 to $280 per hour) and $196,471.84 for WHS (representing a current rate for Wallace at $250 per hour).

83 -. WHS did not represent any of the plaintiffs on appeal, but Joseph Wallace (now with the law firm of Harris & Bush) consulted with GDC, CIR, and Terral Smith during the appeal on a limited basis.

84 . The post-appeal applications were filed beginning in October 1996. One application for attorneys’ fees was filed on behalf of CIR, GDC, and Wallace. CIR seeks total compensation of $243,-421.68, consisting of $195,897.50 in fees for 909.8 hours, $17,638.56 in expenses, and $29,-885.62 in interest. GDC seeks compensation of $287,925.00 in fees for 1,010.5 hours of work and $4,189.87 in expenses for a total of $292,-114.87. GDC’s hourly rates range from $450 to $275 per hour. Wallace seeks compensation for 18 hours at $250 per hour, plus expenses of $70.30 and interest of $25.50, for a total of $4,595.50. Terra! Smith submitted a post-appeal application seeking compensation for 155.8 hours at $225 per hour for a total of $35,055 in fees. Steven Smith’s post-appeal application seeks compensation for 306.75 hours at $125 per hour plus $607.76 in expenses for a total of $38,951.51.

85 . For instance, the plaintiffs argue they are entitled to such fees because they relied on and cited to press coverage in their motion for attorneys’ fees to demonstrate the impact of the results the plaintiffs achieved in the Fifth Circuit. The significance of the Fifth Circuit opinion did not need to he explained to this Court by the media.

86 . The following is the breakdown, by firm and/or attorney, of time spent dealing with the media during the trial phase of the case: Steven Smith (4.50 hours) and Terra! Smith (1 hour). The following is the breakdown of the appellate hours: GCD (48.25 hours), CIR (5.90 hours), Terra! Smith (7.10 hours) and Steven Smith (6.75 hours).

87 . The extent of the defendants’ involvement in the intervention was as follows: On January 10, 1994, the defendants filed a one-sentence response to the proposed intervention in this Court, stating that they "do not object to the intervention.” On February 16, 1994, the defendants supported a motion filed by the intervenors in the Fifth Circuit to expedite the appeal of the proposed intervention, stating "Defendants believe that the intervention would be beneficial to an appropriate resolution of the litigation and believe it to be important to resolve the intervention issue expeditiously.” On May 3, 1994, the defendants submitted a letter to the Fifth Circuit case manager indicating that they did not oppose intervention and that "the adverse parties in this appeal are the plaintiffs and the potential interve nors.” See Exhibits 1, 2, and 3, Plaintiffs’ Second Supplemental Memorandum in Further Support of Attorneys’ Fees, October 11, 1994. On February 1, 1995, the defendants submitted another letter, this time to the Clerk of the Fifth Circuit, indicating that they did not intend to file a brief on the issue of intervention but reiterating that intervention “would be beneficial to the final resolution” of the case. See Exhibit K, Supplemental Appendix, Supplemental Memorandum of Plaintiffs Hopwood and Carvell in support of Attorneys’ Fees, October 1, 1996. These actions are insufficient to support the plaintiffs’ contention that the defendants participated in the intervention.

88 . The breakdown of trial hours: CIR (127.60 hours), Terral Smith (27.80 hours), WHS (18.75 •hours), and Steven Smith (4.75 hours). The breakdown of appellate hours: CIR (302.57 hours), GDC (23 hours), and Terra! Smith (12.90 hours).

89 . Travel time during the trial phase of the case is 153.25 hours for WHS, 15.75 hours for Steven Smith, 12 hours for Terral Smith, and 10 hours for CIR. There are 23.50 appellate hours for GDC and 12 .55 appellate hours for CIR relating to travel. The numbers reflected here do not reflect the 50% reduction.

90 . Given the time sequence, these hours were listed in the plaintiffs’ supplemental applications for attorneys’ fees. These hours will therefore he subtracted from the "appellate hours” calculation. The breakdown for each attorney and/or firm is CIR (151 .10 hours), Steven Smith (52.50 hours), and GDC (45.50 hours).

91 . As evidence of billing judgment, the CIR claims it excluded from its fees application time spent on fundraising, publicity, efforts to obtain additional counsel, expenses related to the turnover of attorneys assigned to the case, compensation on behalf of a pro bono attorney who conducted a deposition, and fees related to the submission of the fee application. Except for the latter two items, these hours are not compensable. Therefore, the Court does not consider the majority of these exclusions to be evidence of billing judgment. See Walker, 99 F.3d at 769-70.

92 . CIR, for instance, submitted one entry labeled "Read Defendant-Appellee’s brief” which the attorney claims took 4.50 hours at a cost of $810.

93 . For instance, GDC’s records are replete with such time entries as "REVIEW AND REVISE BRIEF,” "LEGAL RESEARCH RE 5TH CIRCUIT ISSUES," "REVIEW OF MATERIALS,” and "TC W/MR ROSMAN,” all of which are simply too vague and brief to inform the Court precisely what work was done. CIR had similar entries, such as "Review Court Filings,” as did WHS, which had the following imaginative and informative entries: "Research” and "Trial Preparation.” Furthermore, GDC, Terral Smith, CIR, and WHS did not separate the activities for each day and instead lumped all work together under one heading. The Fifth Circuit has specifically and repeatedly denounced such billing practices. See, e.g., Walker, 99 F.3d at 773 (calling identical billing practices "woefully inadequate to support any fee application ... [because] ... no responsible client would accept [such] records as capable of supporting a bill”).

94 . Wallace submitted only 18 post-trial hours which primarily consisted of reviewing motions and court orders. These hours were of minimal or no value to the plaintiffs and will be reduced to 3.50 hours, the time he reasonably spent on the plaintiffs’ application for attorneys’ fees.

95 . Other than the pleadings related to their requests for attorneys’ fees, a brief requested by the Fifth Circuit regarding the Adarand decision, and two briefs related to the defendants’ failed petition for writ of certiorari, all of CIR’s post-trial filings relate to the intervention attempts. See Exhibit J, Supplemental Memorandum in Support of Motion for Attorneys’ Fees, Supplemental Appendix (Part I), October 1, 1996.

96 . The undersigned has been privileged to try lawsuits against some of the most competent trial lawyers in the United States from 1965 to 1991 before taking a seat on the federal bench. The undersigned has never seen a lawyer worth $450 per hour or $4,500 per day — only successful promoters.

97 . Terra! Smith, Wallace, Wheeler, and Olson have over 20 years experience; McDonald, Rosman, and Cox have over 10 years experience; and Steven Smith, Troy, Bader, Shea, Hungar, Scott, and Nelson have less than 10 years experience. The Court was provided with no information regarding the years of experience of Harris, Mulloy, or Wilson.

98 . Steven Smith untimely submitted evidence that the prevailing market rates in Austin range from $123 to $150 per hour for an attorney of his skill and experience. The remaining attorneys only submitted evidence of the prevailing market rate in Washington, D.C. and Virginia. WHS's evidence establishes a prevailing market rate for civil rights litigation at $200 per hour. CIR submitted evidence that the prevailing market rates range from $90 to $190 per hour for associates and from $195 to $350 per hour for partners. GDC submitted evidence'that hourly rates range from $90 to $230 for associates and $195 to $525 for partners. The rates submitted by CIR and GDC only reflect the general hourly rates charged to clients in Washington, D.C. and do not purport to reflect the market rate for work performed in federal civil rights litigation.

99 . The Court has never granted an application for attorney's fees with a billing rate higher than $200 per hour without the agreement of the parties so stipulating.

100 . Three other WHS associates (all of whom have rates lower than $100 per hour) performed a total of one hour of work on the case. For the sake of simplicity, the Court will assign this hour of work to Wilson in calculating the lodestar.

101 . The defendants also object to over $7,000 for the deposition costs and travel expenses related to the plaintiffs' expert witness at trial. Under § 1988(c), it is within the Court's discretion to include such expenses as part of a- reasonable attorney's fee for cases brought pursuant to § 1981. 42 U.S.C. § 1988(c). The Court declines to deduct this expense.

102 . The Court derived the prime rate from a booklet entitled "Selected Interest Rates” published by the Board of Governors of the Federal Reserve System. For the sake of simplicity, the Court applied the average prime rate per annum (simple) for the relevant time period rather than •applying a different prime rate for each quarter.