Pouncy v. Prudential Insurance Co. of America

499 F. Supp. 427 (1980)

Riley D. POUNCY, Plaintiff,
v.
PRUDENTIAL INSURANCE COMPANY OF AMERICA, Defendant.

Civ. A. No. 75-H-1877.

United States District Court, S. D. Texas, Houston Division.

July 9, 1980.

Gordon R. Cooper, II, Houston, Tex., for plaintiff.

Richard R. Brann and Tony P. Rosenstein, Baker & Botts, Houston, Tex., for defendant.

FINDINGS OF FACT AND CONCLUSIONS OF LAW

CARL O. BUE, Jr., District Judge.

I. Introduction

Plaintiff Riley D. Pouncy, a black male, brought this employment discrimination action pursuant to Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e et seq. (1978), and the Civil Rights Act of 1866, 42 U.S.C. § 1981 (1974), on behalf of himself and all other Blacks who have been discriminated against by the defendant Prudential Insurance Company of America, pursuant to Rule 23(a) and 23(b)(2), Fed.R.Civ.P. On September 29, 1978, the Court provisionally certified a class consisting of all present and future black employees in the Administrative Section of defendant's Southwestern Home Office in Houston, Texas, plus all former black employees in the Administrative Section at the Southwestern Home Office in Houston who were in defendant's employ on or after October 31, 1973. By order of August 6, 1979, the Court modified the class previously certified to exclude part-time and commissary workers and wage band employees and granted the parties' motion to bifurcate the trial.

At the conclusion of the evidence on liability, the Court requested additional briefing by the parties and took the case under advisement. Pursuant to Rule 52, Fed.R. Civ.P., the Court hereby enters its Findings of Fact and Conclusions of Law detailing the reasons for its conclusion that plaintiff has failed to sustain his burden of proving that he or the plaintiff class has been discriminated against by defendant on the basis of race and that as a consequence the defendant should prevail. Because of the multiple legal issues involved as well as the complex statistical and other proof discussed as to each, the case authorities relied upon are examined in the Court's Findings of Fact, rather than set forth separately in the Conclusions of Law.

II. Findings of Fact

A. Discriminatory Treatment

1. Plaintiff Riley D. Pouncy

1. Plaintiff Riley D. Pouncy is a black male citizen of the United States and a resident of Houston, Harris County, Texas.

2. Defendant Prudential Insurance Company of America is a duly incorporated organization, authorized to do business within the State of Texas, and is an employer within the meaning of 42 U.S.C. § 2000e et seq. (1978).

3. Plaintiff Pouncy was initially hired by defendant on or about September 27, 1965, as a chauffeur-driver, which is a level 4 position^[1] within the office services division in the administrative section of defendant's Southwestern Home Office. Plaintiff's duties as a chauffeur-driver included distribution of incoming mail, preparation of outgoing mail for delivery to the post office, and other miscellaneous functions akin to those of a shipping and receiving clerk.

4. On or about April 15, 1971, plaintiff was transferred to the position of multilith operator, also a level 4 position, with the duplicating section of the office services division. Plaintiff was first assigned to operate a 360 A. B. Dick printer, but within several months he was moved to a 1250 multilith, which is a more complex printing machine. Plaintiff was trained for the operation of these machines by E. J. Wilkins, who was his supervisor in the duplicating section from 1971 to 1975.

5. On May 21, 1973, some two years after plaintiff had commenced working in the duplicating section, one Olga Aschenbeck, a white female, began work in that section as a copyprint operator, which is a level 2 position. Ms. Aschenbeck had previously been employed as a commissary worker in defendant's Southwestern Home Office from March 12, 1969, to May 18, 1973, when she resigned in order that she might be reassigned to office services. Ms. Aschenbeck filled a vacancy created by the promotion of Mary Harris, a black woman, to a level 4 position elsewhere in office services.

On November 5, 1973, Ms. Aschenbeck was placed in training for a multilith operator's job and in 20 days was promoted to that position, as a level 4. At this time, the duplicating section of the office services division consisted of Mr. E. J. Wilkins, supervisor; Ms. Wanda Simms, assistant supervisor; plaintiff, Ms. Aschenbeck and another Xerox copy machine operator.

Approximately fifteen months later, in March of 1975, Mr. Wilkins was transferred to the auditing division within the company, and Ms. Simms was promoted to the supervisor's position. At the same time it was announced by Mr. Wilkins and Mr. Hansberry, the manager of office services, that Ms. Aschenbeck had been promoted to assistant section supervisor of the duplicating section, a level 5 position.

6. Plaintiff, believing that he had been discriminatorily denied the promotion which Ms. Aschenbeck received, filed a charge of discrimination with the Equal Employment Opportunity Commission (hereinafter "EEOC") on April 10, 1975. Plaintiff's Exhibit 12. On August 28, 1975, the EEOC issued its determination in the matter, concluding that there was reasonable cause to believe that plaintiff had been denied the promotion on the basis of his race. Plaintiff's Exhibit 1. On October 14, 1975, the EEOC issued its "Notice of Right to Sue Within 90 Days," Plaintiff's Exhibit 2, and on October 31, 1975, plaintiff filed his original complaint in this Court.

7. In March of 1976, plaintiff was transferred into the dental claims department of defendant's Southwestern Home Office as a claims examiner. Upon his transfer into the department, plaintiff was a level 4 in training for a level 5 position. After three months in the department, plaintiff received his promotion to level 5. Plaintiff's duties as a claims examiner included the examination and payment of dental claims. In early 1977, plaintiff was promoted to a claims quality reviewer, a level 6 position, in which he examined the more complex dental claims prior to payment and advised other claims examiners how and whether to pay dental claims. During the year in which plaintiff worked in the dental claims department there were five or six other Blacks in his section.

On May 3, 1977, plaintiff was terminated. Plaintiff has never filed an EEOC charge based on his discharge by defendant and has never sought to amend his complaint in this proceeding to allege that his discharge was retaliatory or even discriminatory.

(a) Defendant's Failure to Promote Plaintiff

8. The decision to promote Ms. Aschenbeck rather than the plaintiff to assistant section supervisor was made by Mr. Wilkins, the section supervisor, and Mr. Hansberry, who was manager of the office services division. Based on personnel evaluations and their observations of plaintiff and Ms. Aschenbeck during their tenure in the section, Wilkins and Hansberry concluded that although plaintiff had more education and seniority in the duplicating section, plaintiff lacked other qualities necessary to be a supervisor.

For example, in a March 18, 1974, a personnel appraisal report on plaintiff, Mr. Wilkins, plaintiff's supervisor, indicated that plaintiff "seems to have difficulty in communicating with others in the section, and for this reason I feel it would be very difficult for him to train other employees." Wilkins further recommended that it "would be helpful for him if he would learn to communicate with others and to enroll in some of the study courses offered by the Company." Defendant's Exhibit 21, page 24.

On the other hand, in a September 24, 1974 evaluation of Ms. Aschenbeck, Mr. Wilkins described her as the "type of employee who is a real pleasure to work with," as well as "very cooperative" and someone who "gets along well with almost everyone." Defendant's Exhibit 22. In addition, Mr. Wilkins found Ms. Aschenbeck to have a pleasant and businesslike attitude, as well as an ability to reschedule projects and to explain tactfully to the various division heads why their work had to be rescheduled.

9. Further testimony established that plaintiff was frequently sullen and sharp with others in his division, was inflexible with regard to the rescheduling of work that was often necessary in the section, and was uncooperative about working overtime.

In addition, other incidents such as plaintiff's confrontations with Marian Tipton and Norman Schnatterer in two separate incidents in the commissary, see Defendant's Exhibit 21 at p. 23, as well as a subsequent clash with Mr. Robbins over plaintiff's parking in unauthorized areas, are consistent with the Court's finding that although plaintiff was an able worker, he was also a defensive, uncommunicative individual who believed that company rules did not apply to him. In short, he lacked the ability to work as part of a team effort.

Thus, although plaintiff had performed and did perform well in his subsequent period of employment for defendant in other positions, at the time he was passed over for promotion, his supervisors reasonably concluded that he lacked the necessary skills in communication and training ability to be a supervisor. Therefore, the Court finds that the selection of Ms. Aschenbeck over plaintiff was not based on racial discrimination.^[2] Rather, the Court concludes that this decision was based on a good-faith, rational evaluation of the relative qualifications and abilities of the two individuals in question.

10. In order to establish a prima facie^[3] case of racial discrimination according to a theory of disparate treatment, a Title VII complainant must prove: "(1) that he belongs to a [protected] minority; (2) that he applied for and was qualified for a job for which the employer was seeking applicants; (3) that, despite his qualifications, he was rejected; and (4) that, after his rejection, the position remained open and the employer continued to seek applications from persons of complainant's qualifications." McDonnell Douglas Corp. v. Green, 411 U.S. 792, at 802, 93 S. Ct. 1817, at 1824, 36 L. Ed. 2d 668 (1973). With regard to his claim of denial of promotion, plaintiff clearly satisfies the first McDonnell Douglas requirement, and the Court believes that he would have made formal application for the assistant supervisor's position had he known of the vacancy, thus fulfilling the second prong. The Court finds, however, that plaintiff cannot satisfy the third requirement. Although he had achieved technical proficiency in his performance in the duplicating section, he lacked the necessary communication, leadership and training skills to perform as a supervisor. Thus, the Court finds that plaintiff has failed to establish a prima facie case of discriminatory treatment on the basis of his race.

11. In the alternative, even if plaintiff had established a prima facie case according to the McDonnell Douglas standard, the Court finds that defendant has adequately rebutted plaintiff's proof inasmuch as defendant has established that the decision to promote Ms. Aschenbeck rather than plaintiff was made on the basis of factors other than plaintiff's race. See Findings of Fact Nos. 8 and 9. Numerous courts have indicated that an employee's "ability to work amicably with fellow employees" is a valid criterion, and therefore not an impermissible consideration, in selecting someone for a supervisory position. Nance v. Union Carbide Corp., 540 F.2d 718, 727 (4th Cir. 1976); see also Ayon v. Sampson, 547 F.2d 446, 451 (9th Cir. 1976); Levens v. General Services Administration, 391 F. Supp. 35, 36 (W.D. Mo.1975), aff'd, 538 F.2d 332 (8th Cir. 1976); Burns v. Thiokol Chemical Corp., 6 FEP 261, at 268 (N.D.Ala.1971), rev'd on other grounds, 6 FEP 269 (5th Cir. 1973).

12. Since the Court finds that defendant has met its burden of proving a legitimate, non-discriminatory reason for its actions, the burden thus shifts to plaintiff to prove by a preponderance of the evidence that the articulated reason for plaintiff's non-selection was merely a pretext^[4] for racial discrimination. Turner v. Texas Instruments, 555 F.2d 1251 (5th Cir. 1977). The Court finds no evidence that the reasons articulated for the challenged actions were pretextual; accordingly, plaintiff has failed to satisfy its ultimate burden of establishing that he was discriminated against on the basis of his race when he was denied a promotion to assistant section supervisor of the duplicating section.

(b) Plaintiff's Discharge

13. Defendant has raised the question whether the issues surrounding plaintiff's termination are properly before this Court, in view of plaintiff's failure to file an EEOC complaint based on his termination, to amend his complaint in this case to cover his discharge, or even to include or refer to his discharge in the Pretrial Order in this case. See Defendant's Post Trial Brief at pp. 14-15. However, while a court is ordinarily without jurisdiction to consider a discriminatory act which is not made the subject of a timely EEOC charge, United Air Lines v. Evans, 431 U.S. 553, 558, 97 S. Ct. 1885, 1889, 52 L. Ed. 2d 571 (1977), the rule is not without exception. For example, jurisdiction may be extended to claims of discrimination which are like or reasonably related to allegations contained in the charge and which grow out of such allegations during the pendency of the charge before the EEOC. Ramirez v. National Distillers & Chemical Corp., 586 F.2d 1315, 1320 (9th Cir. 1978); Sanchez v. Standard Brands, 431 F.2d 455, 466 (5th Cir. 1970); Haggerty v. Exxon Corp., 17 FEP 1322 (S.D.Tex.1978); but see Oubichon v. North American Rockwell Corp., 482 F.2d 569, 571 (9th Cir. 1973); Hubbard v. Rubbermaid, Inc., 436 F. Supp. 1184, 1194 (D.Md.1977) (even if claim is like or related to original charge, it must arise during pendency of original charge before the EEOC in order for the district court to have jurisdiction). However, in the case at bar, plaintiff's termination on May 3, 1977, did not occur during the pendency of the EEOC investigation, which concluded on October 14, 1975, with the issuance to plaintiff of a notice of right to sue. See Finding of Fact No. 6. Accordingly, plaintiff's termination would not constitute a claim reasonably related to and growing out of the allegations of his EEOC charge during its pendency before the Commission.

In addition, the district court has jurisdiction over claims of discriminatory retaliation for previously filed EEOC charges. Pettway v. American Cast Iron Pipe Co., 411 F.2d 998, 1002-04 (5th Cir. 1969) (district court held to have ancillary jurisdiction of plaintiff's claim of retaliatory discharge during pendency of appeal of district court's dismissal of plaintiff's original Title VII complaint, even though retaliation claim had not been made the subject of an EEOC charge); Curran v. Portland Superintending School Committee, 435 F. Supp. 1063, 1075-76 (D.Me.1977); Held v. Missouri Pacific Railroad, 373 F. Supp. 996, 1001 (S.D.Tex.1974). While it is not clear at all from plaintiff's briefs whether he is contending that he was discharged for pursuing his charge before the EEOC, if such a contention were urged, the Court would have jurisdiction of plaintiff's termination claims. Therefore, out of an abundance of caution to ensure that plaintiff's claims are fully examined, the Court feels compelled to give a liberal reading to the pleadings before it. Accordingly, the Court finds on such a basis that it has jurisdiction of plaintiff's discharge claims as claims of retaliatory action in violation of Section 704(a), 42 U.S.C. § 2000e-3(a).

However, as stated hereinabove, plaintiff never amended his complaint in this cause to include claims relative to his discharge; neither was the subject of plaintiff's discharge mentioned in the pre-trial order in this cause. Nevertheless, inasmuch as testimony and evidence relating to plaintiff's termination were admitted without defendant's objection and, further, since defendant was afforded the opportunity to defend fully on these issues, the Court views these matters as being tried by implied consent within the scope of Rule 15(b), Fed.R.Civ.P. See Norris v. Bovina Feeders, Inc., 492 F.2d 502, 505-06 (5th Cir. 1974).

Finally, even if the Court lacked jurisdiction over this claim because of plaintiff's failure to make it the subject of a timely charge before the EEOC, such a discriminatory act will nevertheless constitute relevant background evidence to the Court's consideration of other claims of discrimination. United Air Lines v. Evans, supra, 431 U.S. at 558, 97 S.Ct. at 1889, 52 L.Ed.2d at 578. Therefore, in order to give full vent to all of plaintiff's claims, the Court will examine the circumstances surrounding plaintiff's discharge by defendant.

14. The facts surrounding plaintiff's discharge are relatively simple; plaintiff was discharged on May 3, 1977, for his repeated refusals to perform a legitimate assignment that was part of his job. While plaintiff maintains that he was terminated for his refusal to teach a class in dental claims unless he was given a level promotion commensurate with the increased responsibility, the more credible testimony and documentary evidence indicate that plaintiff refused to simplify a corporate memorandum which related to claim payment guidelines.

Plaintiff was first requested by Diana Wong, his immediate supervisor, to rewrite the memo, but plaintiff refused. Defendant's Exhibit 50. Plaintiff also refused the same request by Janis Blount, who was associate manager of the group claims division, and Dave Streilein, who was the division manager. Defendant's Exhibits 48 and 49. Ultimately, after plaintiff told Streilein that he would not revise the memo and that if he had to continue to work with Diana Wong, he would "cause a disruption," Mr. Streilein recommended to Ronald Claiborne, who was the associate general manager of the group insurance area, that plaintiff be terminated; this recommendation was concurred in by Fran Early, the personnel director of defendant's Southwestern Home Office. See also Defendant's Exhibits 38 and 79.

15. In view of the defendant's written policy that refusal to perform a legitimate assignment necessitates "termination with no warning or notice and no option to resign," Defendant's Exhibit 6, the Court views plaintiff's termination as clearly warranted by company policy. In addition, it is clear that plaintiff was given repeated warnings and the opportunity to perform the assignment by Diana Wong, Janis Blount and Dave Streilein. Finally, the facts surrounding the incident and the proposed termination were reviewed by the associate manager over the group insurance area and the personnel director who, after conducting an interview with the plaintiff, concurred in the decision to discharge him. Upon consideration of the witnesses' testimony and documentary evidence concerning this incident, the Court finds that there is no evidence of racial discrimination attendant to the occasion of plaintiff's termination.

16. While the Court hereinabove felt obligated to examine the issue of retaliatory discharge, the record of this trial reflects no evidence remotely suggestive of plaintiff's having been fired in retaliation for filing his EEOC charge some two years earlier. In fact, plaintiff had received a transfer to the dental claims department and two promotions since the filing of his EEOC complaint. See Finding of Fact No. 7.

Therefore, the Court concludes that there is no evidence that plaintiff was discharged because of his race or in retaliation for exercising his rights under Title VII.

17. The McDonnell Douglas standard of order and allocation of proof is also applied in individual cases based on discrimination in discharge. See McDonald v. Santa Fe Trail Transportation Co., 427 U.S. 273, 281-285, 96 S. Ct. 2574, 2579-2581, 49 L.Ed.2d 493 (1976); Powell v. Syracuse University, 580 F.2d 1150, 1154-55 (2nd Cir. 1978), cert. denied, 439 U.S. 984, 99 S. Ct. 576, 58 L. Ed. 2d 656 (1979); Flowers v. Crouch Walker Corp., 552 F.2d 1277, 1281-82 (7th Cir. 1977); Barnes v. St. Catherine's Hospital, 563 F.2d 324, 329 (7th Cir. 1977); Turner v. Texas Instruments, 555 F.2d 1251, 1254-55 (5th Cir. 1977); Henry v. Ford Motor Co., 553 F.2d 46, 48 at n. 3 (8th Cir. 1977). But see Stevens v. Junior College District of St. Louis-St. Louis County, 548 F.2d 779, at 781 (8th Cir. 1977); King v. Yellow Freight System, Inc., 523 F.2d 879, 882 (8th Cir. 1975).

Thus, in order to establish a prima facie case, plaintiff "must meet the initial burden of proving that he was a member of the protected class, that he was discharged, and produce evidence of disparate treatment from which the Court may infer a causal connection between the basis and the discharge." Citing McDonnell Douglas Corp. v. Green, supra 411 U.S. at 802, 93 S.Ct. at 1824. 36 L.Ed.2d at 677. B. L. Schlei and P. Grossman, Employment Discrimination Law, at 511 (1976).

Although plaintiff meets the first two criteria of the applicable test, plaintiff has produced no evidence that he was terminated for an infraction for which non-minorities were not. Accordingly, the Court finds that plaintiff has failed to establish a prima facie case of racial discrimination based on disparate treatment in his discharge.

18. Even if plaintiff had satisfied the McDonnell Douglas criteria or any standard that would be a likely substitute therefor, the Court finds that defendant has adequately carried its burden^[5] of establishing that the decision to terminate plaintiff was prompted by a legitimate consideration—the failure to perform a valid assignment. See Findings of Fact Nos. 14 and 15. Finally, plaintiff has wholly failed to prove that the reason on which defendant's decision was allegedly based was pretextual.^[6]

19. However, regardless of whether the McDonnell Douglas test is utilized, the ultimate issue in an individual discharge case is whether there has been disparate treatment between a member of a racial minority group and others not of that group. Henry v. Ford Motor Co., supra, at 48. It is therefore incumbent upon the Court to consider the full merits of plaintiff's proof and the company's defense in order to arrive at a conclusion as to whether plaintiff has been the victim of employment discrimination based on his race. See Peters v. Jefferson Chemical Co., 516 F.2d 447, 450 (5th Cir. 1975).

Since there is rarely direct evidence of disparate treatment in the individual discharge case, courts look to other factors that might sustain the inference of discrimination. One of the factors often considered is whether the employee was given repeated warnings and the opportunity to improve. Bolton v. Murray Envelope Corp., 493 F.2d 191, 194 (5th Cir. 1974); Francis v. Allied Service Company of Texas, Inc., 486 F.2d 597, 598 (5th Cir. 1973) (per curiam). In the case of plaintiff's termination, however, plaintiff's supervisor and her immediate superiors made repeated efforts to persuade plaintiff to perform the assigned task and to warn him that his refusal was grounds for termination. See Finding of Fact No. 14.

20. Another commonly relied upon factor is whether the complainant was afforded a hearing or the benefit of an investigation prior to termination. See, e. g., Turner v. Texas Instruments, 555 F.2d 1251, 1254 (5th Cir. 1977); Martin v. Chrysler Corp., 10 FEP 329, 332 (E.D.Mich.1974); Goodloe v. Martin Marietta Corp., 7 FEP 964, 966 (D.Colo.1972), aff'd mem., 10 FEP 1176 (10th Cir. 1974). In the instant case, prior to the decision to terminate plaintiff, the situation was reviewed by Ron Claiborne, director of the group insurance division, and by Fran Early, the personnel director, who interviewed the plaintiff. Findings of Fact Nos. 14 and 15.

21. Finally, other courts have looked to whether the employer deviated from its normal patterns and practices in effecting a complainant's termination. See, Lowry v. Whitaker Cable Corp., 348 F. Supp. 202, 210 (W.D.Mo.1972), aff'd per curiam, 472 F.2d 1210 (8th Cir. 1973). In the instant case, there is no evidence that defendant's employees followed anything other than standard operating procedure in making the decision and effecting the termination of the plaintiff.

Therefore, the Court finds that plaintiff has failed to sustain the inference that his discharge was the result of racial discrimination.

2. Plaintiff's Witnesses

In order to bolster his individual and class claims, plaintiff offered the testimony of several other black former employees in the Administrative Section of defendant's Southwestern Home Office.

(a) Esther DuPont

22. Esther DuPont is a black female who was originally hired into a level 2 position in defendant's commissary division in 1971. On October 8, 1973, Ms. DuPont was transferred to the office services division where she assumed the position vacated by Olga Aschenbeck upon her promotion to assistant section supervisor. See Finding of Fact No. 5. Approximately nine months later, on July 9, 1974, Ms. DuPont was transferred to the debit insurance services division where she was placed in training for a level 3 position. On September 26, 1974, Ms. DuPont resigned.

23. Ms. DuPont complains about two separate instances in which discrimination was allegedly manifested against her. First, she complains of the discriminatory manner in which commissary workers were chosen for transfer to other divisions and asserts that while Blacks were tested prior to transfer, Whites were not.

Three black commissary workers, Esther DuPont, Emma Hagger and Mariba Harrison, were interviewed for the vacancy which occurred in office services in October of 1973. Although Ms. DuPont was the least senior of the three, she received the transfer. While Ms. DuPont was given a clerical test prior to her transfer, as the Court concludes hereinafter, see Finding of Fact No. 53, testing of transferring commissary employees was utilized in a haphazard, but not discriminatory, manner. Further, while Ms. DuPont complains that Whites with less seniority than she were given priority in transferring to clerical jobs, the Court finds that there is no evidence of discrimination manifested by the pattern of transfers of Whites and Blacks from the commissary division. See Finding of Fact No. 52.

24. Secondly, Ms. DuPont claims that after her transfer to the debit insurance services division, she was wrongfully and discriminatorily denied her promised promotion to level 3. However, Ms. DuPont's supervisors, Cile Cleghorn, who is white, and Lela Raven, who is black, found Ms. DuPont to be belligerent in her attitude and deficient in mastering her job, and therefore undeserving of the promotion. On September 17, 1974, Ms. Cleghorn and Ms. Raven jointly prepared a personnel appraisal report on Ms. DuPont. Defendant's Exhibit 56. In the appraisal, Ms. Cleghorn described Ms. DuPont as showing "all evidence of a negative, defensive attitude toward her supervisors, co-workers and her present job assignment." Ms. Cleghorn further indicated that on several occasions it had been called to the attention of her supervisors that Ms. DuPont had "been rude and discourteous, almost to the point of being belligerent, causing tension in the section." Ms. Raven, the black assistant supervisor, indicated that Ms. DuPont had failed to respond to additional help and training she had received and, therefore, "[i]t now looks questionable whether Esther will qualify for Level 03 due primarily to her unwillingness to learn or perform as needed." Id.

25. On September 26, 1974, a dispute arose between Ms. DuPont and Ms. Cleghorn, over Ms. Cleghorn's instructions to Ms. DuPont to re-do a job that she had done improperly. Ms. Cleghorn contacted Al Espenship, manager of the debit insurance services division, who requested that Ron Claiborne, then the personnel manager, discuss the problem with the parties concerned. Defendant's Exhibit 39. Mr. Claiborne talked with Ms. DuPont for two hours, discussed the matter with her supervisors, and then told Ms. DuPont that the company would work with her. Ms. DuPont refused to return to work and asked Mr. Claiborne to fire her, but he refused and Ms. DuPont resigned. Id.

Ms. DuPont filed a charge with the EEOC on September 23, 1974, based on her failure to receive her promotion. Plaintiff's Exhibit 10. The Commission ultimately determined that there was no reasonable cause to believe that Title VII had been violated. Id.

26. The Court finds no evidence of racial discrimination attendant to defendant's failure to give Esther DuPont her level 3 promotion. Rather it was the considered judgment of Ms. DuPont's supervisors, both black and white, that her performance and behavior precluded her advancement.

(b) Emma Hagger

27. Emma Hagger, a Black, was hired in December, 1970, into defendant's commissary division, where she remained until she was hospitalized in October of 1973. She returned to work for a brief period in February of 1974, but she was then physically unable to perform the duties of her former job. Ms. Hagger was terminated effective April 17, 1974, when her company disability benefits expired. Defendant's Exhibit 23.

28. Ms. Hagger complains of the defendant's failure to transfer her from the commissary to a position in the clerical division and her subsequent termination.

Ms. Hagger testified that she worked part-time in the commissary and substituted in various departments in office services when she was needed. Evidently Ms. Hagger requested on several occasions to be transferred to a clerical position, and she was eventually tested. She also was interviewed by Norma Hightower for the vacancy in the duplicating section created by Olga Aschenbeck's promotion, see Findings of Fact Nos. 5 and 23, but she was not chosen because of her impending surgery.

29. Following her convalescence, she returned to work in the commissary, but she was advised by her physician that she could not continue in her old job because of her back problems. Ms. Hagger then met with Ron Claiborne to request a transfer to a clerical job because of her physical disability. Mr. Claiborne twice requested that Ms. Hagger's physician consult with the company physician about her condition, but the Prudential doctors were never contacted by her private physician. See Defendant's Exhibit 40. Neither did Ms. Hagger renew her contact with Mr. Claiborne about returning to her old job or locating another. Accordingly, in accordance with defendant's policy, Ms. Hagger was terminated upon the expiration of her company disability. Id.

On March 27, 1974, Ms. Hagger filed a charge of discrimination with the EEOC, Plaintiff's Exhibit 11, which resulted in a determination by the Commission that there was no reasonable cause to believe that Title VII had been violated. Defendant's Exhibit 24.

30. The Court finds no evidence of discrimination in the events surrounding Ms. Hagger's requests for transfer or her termination.

(c) Murreba Harrison

31. Murreba Harrison, a black female, was hired as a food service worker in defendant's commissary division in 1964 and remained in this division until the commissary was closed in September, 1977, whereupon her employment was terminated. During her tenure, Ms. Harrison received two promotions, first, to senior food service worker and finally, in September of 1976, to assistant supervisor.

32. Ms. Harrison complains generally of the failure of her supervisors in the commissary to recommend her for transfer to a clerical position in another division of the company. She claims that she was discouraged by her supervisors in her quest for advancement and rebuffed by the personnel administrator when she sought a transfer on her own.

The evidence indicates only one clerical position for which Ms. Harrison made specific application and was considered—the copyprint operator job that was awarded to Esther DuPont, a Black. See Finding of Fact No. 23.

33. Further, Ms. Harrison was sent to company-sponsored equal employment opportunity training courses and to a course related to cafeteria operations conducted at the University of Houston.

The Court finds that there is no evidence of racial discrimination by the defendant against Murreba Harrison.

(d) Luretta Moses

34. Luretta Moses, a black female, was hired by Prudential in 1967 into the insurance services ordinary division as a file clerk, a level 2 position. During her tenure at Prudential, Ms. Moses worked her way up in the level system in positions as a clerk-typist, policy picker, service clerk, commissions calculator, and finally, to the position of senior commissions calculator, a level 6 position in which she remained until her resignation in 1976.

35. Ms. Moses complains that she sought promotions from level 6 without success numerous times, but the specific position which she claims was discriminatorily denied her was a commissions technician's job that was awarded to Robbie Kerr, a White.

The vacancy in the commissions technician position, which was a level 8 job, was caused by the departure of Mary Carnes on maternity leave in October, 1974. Ann Benton, the department supervisor, knew that Robbie Kerr, who had also taken maternity leave from her job at Prudential, was considering returning to work. According to Ms. Benton and Jack Clark, the department manager, Ms. Kerr was highly qualified in all phases of agents' commission work and had more experience and expertise in the area of commissions than anyone else in the department. Ultimately, the decision was made to offer Ms. Kerr this opportunity, and she accepted. Ms. Kerr returned to Prudential at the same job level in which she had left, a level 8; consequently, the new position did not constitute a promotion for her.

It is not unusual for Prudential to attempt to recall an employee on leave to fill a position for which he or she is known to be qualified.

36. On March 6, 1975, Luretta Moses filed a charge of discrimination with the EEOC. Defendant's Exhibit 25. On May 23, 1975, the Commission issued its determination that the evidence did not support Ms. Moses' charge.

37. Because there is no evidence that Ms. Moses was as qualified as Ms. Kerr for the commissions technician's job, the Court finds that the selection of Ms. Kerr over Ms. Moses was not based on racial discrimination but rather on Ms. Kerr's superior qualifications. In addition, there was evidence of attendance problems on the part of Ms. Moses that would justify her supervisors' reluctance to promote her. Thus, the Court concludes that Ms. Moses' testimony is not probative of discrimination in this case.

(e) Mary Lee Byrd Browning

38. Mary Byrd, a black female, was hired in June, 1966, into defendant's ordinary policy division in a level 1 position and was subsequently transferred into the insurance service ordinary division. Like Luretta Moses, Mary Byrd progressed through the level system during her tenure with the Company and was serving as a senior premium reconciliation clerk, a level 6 position, when she resigned in November, 1976.

39. Although Ms. Byrd complains generally about her failure to advance rapidly enough in the company, she raises two specific instances in which she was passed over for promotion. In the first instance, the job to which Ms. Byrd aspired was awarded to Linda Wilmes, who transferred to the Southwestern Home Office from Prudential's North Central Home Office in June, 1973. At the time of her promotion in September, 1973, to the level 7 assistant supervisor's position sought by Mary Byrd, Linda Wilmes had been in a level 6 position since August, 1972, while Ms. Byrd had been a level 5 premium reconciliation clerk since March, 1972. At the same time Ms. Wilmes received the desired promotion, Ms. Byrd receive a promotion to level 6.

40. In addition, it is clear that Ms. Wilmes was superior in qualifications, performance, and attitude to Ms. Byrd and, therefore, that her selection over Ms. Byrd was not racially motivated.

41. The second promotion about which Ms. Byrd complains is that which was awarded to Marty Freeman in April, 1975, when Linda Wilmes went on maternity leave. The evidence reflects in this instance also that the decision not to promote Ms. Byrd to this position was based on her performance and not on racial grounds.

Some four months before, in November, 1976, upon the recommendation of the associate manager James Paul, who is black, Ms. Byrd was placed on "increase not recommended" status, the equivalent of probation. Defendant's Exhibit 51. Mr. Paul recommended that Ms. Byrd be placed on "INR" because of several complaints from field offices and policyholders revealing that she had not been performing her duties satisfactorily. Mr. Paul reviewed these complaints with Ms. Byrd and advised her that her work was "not up to standard, especially in areas requiring judgment." The manager of ISO, Jack Clark, also discussed Ms. Byrd's job performance with her on December 2, 1974, at which time she "agreed that her performance had not been satisfactory and that her discussions with her supervisor and associate manager [Mr. Paul] had made this clear." Defendant's Exhibit 52. Mr. Clark also told her that her supervisor, Ms. Wilmes, and Mr. Paul had recommended that her annual increase be held up. Clark approved their recommendation and Ms. Byrd was placed on increase not recommended status. Mr. Clark further warned Ms. Byrd that her job would be in jeopardy if her performance did not improve.

42. On February 5, 1975, Mr. Clark again spoke with Ms. Byrd about her progress in work performance since the December 2, 1974 probationary status and the quality reviews which had been performed on her by Ms. Wilmes. The result of the quality reports was that Ms. Byrd was then doing "what could best be described as an average satisfactory job," but still had "some areas where improvement [was] needed. . . ." Defendant's Exhibit 53. Based on his recognition that she was then doing an acceptable job, Mr. Clark notified the personnel division to grant her deferred salary increase effective February 10, 1975.

43. The position to which Ms. Byrd desired a promotion became available within two months of her removal from probationary status, at a time when her work was considered merely satisfactory. The decision by Mr. Paul, Mr. Clark, and the lower level supervisors to promote Ms. Freeman was based on quality reviews and the performance and ability of the candidates. See Defendant's Exhibit 33.

In view of testimony concerning several serious errors made by Ms. Byrd as well as her irresponsible attitude toward her work that was readily apparent during her testimony, the Court concurs with Mr. Paul's assessment indicating his surprise that Ms. Byrd would have expectations of receiving the promotion. Thus, the Court finds that there is no evidence that this denial of promotion to Ms. Byrd was based on racial grounds.

44. Ms. Byrd filed two charges with the EEOC, one on March 6, 1975, and another on April 22, 1975. Defendant's Exhibits 28 and 29. After investigating these charges, the EEOC determined that the evidence did not support the charges. Defendant's Exhibit 30.

(f) Vernel Armstrong

45. Plaintiff's rebuttal witness, Vernel Armstrong, has been employed in defendant's group health claims division since 1973. Ms. Armstrong complains that although she had progressed from level 2 to a level 5 job by 1974, she has progressed no further since that time, and others with less experience have achieved a level 8 or better. Although Ms. Armstrong named several Whites who had been promoted to higher levels, she admitted on cross-examination that numerous Blacks in her division have achieved promotions as high as level 11.

Thus, the Court concludes that Ms. Armstrong's brief testimony is not probative of racial discrimination within the defendant company.

B. Disparate Impact-Class Claims

Plaintiff's claims of class-wide discrimination are based on his theory that defendant's personnel policies, albeit facially and intentionally neutral, have an adverse impact on members of the protected class. See Griggs v. Duke Power Co., 401 U.S. 424, 91 S. Ct. 849, 28 L. Ed. 2d 158 (1971). The practices specifically challenged by plaintiff are numerous, but interrelated. First, plaintiff claims that during the relevant time period, defendant had no experience or educational prerequisites for hires into any of its entry level positions; yet Blacks were concentrated at the very lowest jobs in defendant's level system and were conspicuously absent at the upper levels. Further, plaintiff maintains that although defendant's policy of promoting from within is facially neutral, it has a disparate impact on Blacks. This is so because vacancies are not posted, there are no clearly defined criteria to guide supervisors in awarding promotions, and defendant's management personnel decide which employees will occupy what level and therefore be available for promotion to higher levels.

It is axiomatic that in order to establish liability pursuant to this theory, plaintiff must demonstrate that the challenged policies did, in fact, have a disparate impact on the protected class. Therefore, in support of his contentions, plaintiff offered various exhibits in order to illustrate the impact of these policies on defendant's black employees. Plaintiff's proof concerned promotions and two closely related areas, salary and utilization. In addition, plaintiff offered some evidence relating to terminations.

However, in order to properly determine the validity of plaintiff's class claims, it is first necessary to set forth some of the relevant practices of defendant, as well as the parameters of the class and the relevant time frame for liability and statistical proof.

1. Parameters of the Class—Commissary Employees

46. On September 29, 1979, the Court provisionally certified the class herein to include all black present and future employees in the Administrative Section of the defendant Prudential Insurance Co. at its Southwestern Home Office building in Houston, plus all black former employees in the Administrative Section at the Southwestern Home Office building in Houston who were in defendant's employ on or after October 31, 1973, that being two years prior to the date this suit was filed. On August 6, 1979, the Court modified the class previously certified to exclude part-time, commissary and wage band employees. However, the treatment of commissary workers was explored in depth at trial by means of the testimony of several witnesses called by both plaintiff and defendant and by the record of all transfers from that division. Defendant's Exhibit 55. As a consequence, defendant has requested that the Court redefine the class to include commissary workers in order that defendant may obtain a decision binding upon this group. See Defendant's Post Trial Brief at p. 1-2. For the reasons set forth hereinafter, the class will remain as redefined and commissary workers will not be included within it.

47. Defendant's commissary or cafeteria was staffed and operated by defendant during the period that the Southwestern Home Office was located at its Holcombe Boulevard facility. A noon meal was served at the commissary for all Prudential employees; breakfast was provided for some full-time commissary workers. Defendant closed its commissary division on September 1, 1977, in anticipation of the company's move to its new facility on the West Loop South.

48. During defendant's operation of the commissary, its staff consisted of approximately 22 persons. Employees in this division fell into several categories, see Plaintiff's Exhibit 16 at page 11, but included generally food preparers, food service workers, and a small number of supervisory and management personnel. The majority of commissary employees worked a twenty-hour work week and were paid overtime if they worked more than twenty hours; the supervisors and managers worked forty hours a week. However, commissary workers were paid on a salaried basis as opposed to an hourly basis and received the same benefits as other Prudential employees.

49. On occasion, twenty-hour commissary workers were requested to fill in part-time in other divisions within the company. In these instances, the commissary employee would work four hours in the commissary and the balance of the day in the other division of the company.

50. Although some commissary workers were transferred to clerical positions in other divisions of the company, the commissary was not utilized as a training area for other divisions. In fact, during the eight year period from 1969 to 1977, only seventeen people were transferred from the commissary to other divisions. Of these seventeen, nine (or 52.9% of the total transfers) occurred when the commissary was closed in September, 1977. See Defendant's Exhibit 55.

51. In view of the limited opportunities for advancement within the commissary division and the fact that commissary workers were not generally subject to promotion into other divisions, the Court has serious doubts that the claims of the plaintiff with regard to promotion are typical of the claims of commissary employees. In addition, the vast difference in procedures for transferring from the commissary division to clerical positions in other divisions from the procedures employed in awarding inter-level promotions in other divisions make it clear that plaintiff and the commissary employees whom he seeks to represent are not similarly situated. Accordingly, the Court declines to redefine the class previously certified to include commissary workers.

However, because discrimination within this division would be probative of discrimination in other divisions of the administrative section, the Court has examined the evidence concerning testing and transfer procedures in the commissary division.

52. From the evidence before it, the Court is unable to find that transfers from the commissary to other divisions of office services were awarded in a racially discriminatory manner. Five of these transfers occurred prior to the relevant time period of this case. See Findings of Fact Nos. 54 and 55 and Defendant's Exhibit 55. If the transfers which occurred when the commissary closed are also excluded, there occurred during the relevant time frame only three "promotions" from the commissary to other divisions, one of which was awarded to a White, one to a Black, and one to a Mexican-American. Even considering the evidence of all seventeen transfers, no discrimination is manifested inasmuch as seven of the total transferred were Blacks, eight were Whites and two were Mexican-Americans.

53. The Court further finds that the determination to test an employee prior to transfer from the commissary was not made in a racially discriminatory manner. Of the seven Blacks transferred, four were subjected to a written test and three were not. Of the eight Whites transferred, two were tested and six were not;^[7] of the two Mexican-Americans, one was tested and one was not. Thus, it is clear that testing was utilized as the particular instance, rather than the race of the transferee, demanded.

Therefore, the Court concludes that the evidence concerning transfers from and testing in the commissary division is not probative of racial discrimination within that division or elsewhere in the Administrative Section of Defendant's Southwestern Home Office.

2. Relevant Time Periods for Class Claims and Statistics

54. With respect to liability under Title VII, the class consists only of those employees or former employees who could have filed valid charges of discrimination with the EEOC at the time the class representative, Riley Pouncy, filed his charge. Wetzel v. Liberty Mutual Insurance Co., 508 F.2d 239 (3rd Cir. 1975). Plaintiff filed his EEOC charge on April 10, 1975; accordingly, liability under Title VII may be predicated only on actions by defendant on or after October 13, 1974, i. e., 180 days prior to the date plaintiff's EEOC charge was filed. See Crawford v. Western Electric Co., supra, at 1309.

55. Under 42 U.S.C. Section 1981, the federal courts borrow state statutes of limitations. See, e. g., Johnson v. Railway Express Agency, Inc., 421 U.S. 454, 462, 95 S. Ct. 1716, 1721, 44 L. Ed. 2d 295 (1975); Page v. U. S. Industries, 556 F.2d 346, 351 (5th Cir. 1977). The applicable Texas statute of limitations for claims under Sec. 1981 is two years. Dupree v. Hutchins Bros., 521 F.2d 236, 237-38 (5th Cir. 1975); Prophet v. Armco Steel, Inc., 575 F.2d 579, 580 (5th Cir. 1978) (per curiam). Since plaintiff's complaint based on Sec. 1981 claims was filed with this Court on October 31, 1975, liability under Sec. 1981 extends only to actions by defendant on or after October 31, 1973. Accordingly, any claim in existence between October 31, 1973, and October 31, 1974, the earliest date on which liability under Title VII may be found, must be analyzed under the standards applicable to 1981 claims.

With respect to those claims covered only by Sec. 1981, the plaintiff and the class must make a showing of "purposeful discrimination" before casting the burden on defendant to rebut the claim. Moreover, in the context of alleged employment discrimination, a violation of Sec. 1981 cannot be established by statistical evidence alone. Williams v. DeKalb County, 577 F.2d 248, mod. on reh'g, 582 F.2d 2 (5th Cir. 1978); Grigsby v. North Mississippi Medical Center, Inc., 586 F.2d 457, 460-61 (5th Cir. 1978).

56. Inasmuch as liability for disparate treatment must be based on claims that are thus actionable, it therefore follows that statistical evidence must reflect personnel decisions that would also be actionable in order to constitute the basis of a prima facie case. The Supreme Court in Hazelwood School District v. United States, 433 U.S. 299, 97 S. Ct. 2736, 53 L. Ed. 2d 768 (1977), gave substantial guidance in the area of time frame analysis, i. e., the period in which statistics must be analyzed. The plaintiffs in Hazelwood established that there was a substantial disparity between the percentage of black teachers employed by the school district in the school years 1972-1973, and 1973-1974, and the availability of black teachers in the relevant geographical area. However, the Supreme Court stated that:

"The Court of Appeals totally disregarded the possibility that this prima facie statistical proof in the record might at the trial court level be rebutted by statistics dealing with Hazelwood's hiring after it became subject to Title VII. . . .

A public employer who from that date forward made all its employment decisions in a wholly nondiscriminatory way would not violate Title VII even if it had formerly maintained an all white work-force by purposefully excluding Negroes." (Footnote omitted.)

433 U.S. at 309, 97 S.Ct. at 2742, 53 L. Ed. 2d 788. See also EEOC v. United Virginia Bank, 615 F.2d 147, 150 (4th Cir. 1980).

Further, in United Air Lines v. Evans, 431 U.S. 553, 97 S. Ct. 1885, 52 L. Ed. 2d 571 (1977), the Supreme Court held that:

"A discriminatory act which is not made the basis of a timely charge is the legal equivalent of a discriminatory act which occurred before the statute was passed."

431 U.S. 558, 97 S. Ct. 1889, 52 L. Ed. 2d 588. Thus, the relevant statistics for purposes of analyzing the disparate impact claims of the plaintiff class are those statistics which reflect defendant's personnel actions taken during the relevant time period covered by this law suit, i. e., for claims under 42 U.S.C. Sec. 1981, on or after October 31, 1973, and for Title VII claims, on or after October 13, 1974.^[8]See Neloms v. Southwestern Electric Power Company, 440 F. Supp. 1353 (W.D.La.1977); Schlei and Grossman, supra, (1976) Supp. at 305, 326-327.

3. Inclusion or Exclusion of Pre-Act Hires

57. Likewise, in order to properly analyze certain statistics, employees hired by defendant prior to the effective date of Title VII must be omitted. Exclusion of data relating to pre-Act hires is necessary in some instances to prevent tainting the statistics with data concerning personnel decisions made by defendant prior to the time it was subject to the dictates of Title VII. For example, in EEOC v. United Virginia Bank, supra, an action challenging defendant's hiring practices, the Court of Appeals determined that in making a comparison of the racial makeup of defendant's workforce with the workforce as a whole, pre-Act hires should be factored out. 615 F.2d at 150-151. The Fourth Circuit reasoned that if employees hired prior to the effective date of the Act were not excluded, the statistics would reflect hiring decisions made by defendant prior to the time it was subject to Title VII.

Similarly, in the case at hand, inclusion of pre-Act hires in plaintiff's utilization statistics taints these analyses inasmuch as they reflect hiring decisions made by defendant prior to the time in which defendant was subject to Title VII. See Utilization Analysis, Section II.B.9., infra. On the other hand, statistics concerning promotions during the relevant years and service years prior to promotion should include all pre-Act and post-Act hires inasmuch as all persons within defendant's employ who were otherwise eligible competed for the available promotions during the years in question; further, these statistics reflect promotional actions taken by defendant during the period relevant to this lawsuit. Finally, in analyzing salary statistics, pre-Act hires should also be excluded. It is undisputed that no Blacks were employed by defendant prior to 1965. Therefore, in comparing employees with the same tenure, there would be no Blacks with whom to compare the salaries of Whites hired before 1965.

4. Defendant's EEO Training and Affirmative Action Program

58. Prudential has provided its supervisors and other employees with Company-sponsored training programs designed to educate them in areas of equal employment obligations and to impress upon them the Company's commitment toward equal employment goals. Managers, as well as lower-level supervisors, have been encouraged to attend these courses since at least 1973. Moreover, the evidence demonstrated that when the performance of a supervisor, associate manager or manager is evaluated, his or her success in fulfilling the Company's equal employment opportunity obligations is considered.

59. In addition, defendant maintains an aggressive affirmative action program by means of which it attempts to increase its minority employees through college recruiting at minority colleges. In addition, each division has goals as to minority representation and promotions. Further, the personnel department reviews disciplinary actions taken by supervisors and managers, and evaluates all terminations as a part of the company's affirmative action program.

The testimony and evidence demonstrated that quality reviews are performed on defendant's employment interviewers to ensure that their recommendations are supported by legitimate considerations. Lists of promotable employees, including qualified promotable black employees, are maintained by the personnel division, and promotional decisions are policed to be sure that persons on those lists are given fair consideration.

The defendant's written policies on promotion, compensation and termination are carefully designed with these safeguards to ensure that decisions on these matters are based on legitimate factors, not on an employee's race. See Defendant's Exhibits 4, 5 and 6. More importantly, the internal monitoring systems implemented by the personnel division are designed to ensure that the written policies are, in fact, followed.

5. Defendant's Level System

60. Defendant maintains a system of classifying and grading positions in its various departments by use of a level system. Within the level system, there are 300 to 350 job titles. In order to assign a level to each job, personnel in the applicable division write a job description for each position. The personnel department then evaluates the position according to a point system and thereafter classifies the job according to level.

Within each level there is a minimum and maximum salary range. Thus, an employee's salary will depend not only on the level in which his job is classified but also on the number of merit, general or periodic increases that he himself has received.

61. In each of the departments in the administrative section, levels 2^[9] through 12 are classified as clerical positions. Within this group of clerical employees, supervisors occupy levels 7, 8 and above positions, and assistant section supervisors occupy level 5 positions. The level following level 12 is level 20; persons occupying jobs in this and higher levels are classified as management employees. The level steps following level 20 are 28, 60, 64, 77-78, and 84. Level 20 is the associate manager level, while level 60 is the manager level. Level 84, the highest level in defendant's Southwestern Home Office, is occupied by the company president.

Several levels contain only a very few employees who occupy positions that are unique to that level. For example, level 12 consists of field office planning representatives and underwriters; there are only ten to twelve level 12 employees in defendant's Southwestern Home Office. Similarly, systems analysts constitute level 28, and attorneys and physicians make up levels 77 and 78.

62. Although levels 2 and 3 are entry levels, most college hires enter defendant's employ in level 8 or 9. However, a college degree is not a requirement for promotion into level 8 or 9, but in order to be hired in at level 8 or 9, one must have a college degree or equivalent training and experience.

63. Within the lower levels, the ordinary line of progression is from one level into the level immediately higher. However, because of the level classification of several jobs, the lines of progression may vary. For example, persons selected for promotion to level 12 are usually drawn from level 10. Similarly, level 11 promotions are generally chosen from level 9, and level 20 promotions from level 11. Associate managers (level 20) are most frequently promoted to level 60 manager positions, thus bypassing level 28.

Promotions to levels above 28 require approval from corporate headquarters, while promotions to lower levels are approved through the local office.

6. Defendant's Promotion Policies

64. The department manager and the supervisor will often have the first information concerning a vacancy in their department. Because it is the stated policy of Prudential to promote from within, managers and supervisors will first look within their own division to ascertain the eligible candidates for promotion, or, if necessary, to transfer from another division. If the position cannot be filled by these methods, the personnel department is consulted about hiring from outside. Thus, if it is necessary to go outside the company to fill the vacancy, personnel will accomplish the initial screening and present to the manager and supervisors the names of persons who are qualified to fill the position.

If, on the other hand, the selection is to be made within the department or section, selection will be made from the pool of available candidates as determined by the managers, supervisors and personnel; defendant has no system for posting job vacancies to solicit applications for promotion from within.

65. In order to select the employee who is to be promoted, each department will maintain an inventory of persons eligible for promotion, which is compiled according to individual developmental guidelines. Supervisors in the department are also asked to suggest candidates for promotion. By means of a consultation between the manager and supervisors, the candidates are reviewed on the basis of their latest appraisal reports, which are maintained in the various departments, as well as the employees' disability records and quarterly quality reviews.^[10] Additional factors which are part of the overall picture in determining eligibility for promotion are the candidate's attendance, job performance, work habits, and accuracy. A candidate's merit, based on these factors, rather than seniority, is determinative.

66. In instances in which candidates outside the department are considered for promotion, the manager seeking to fill the slot may request that personnel supply him with names of eligible employees from other departments in the company. In order to respond to these requests, defendant's personnel department maintains three promotional inventories, the first of which coincides with its affirmative action plan. This is an inventory of persons in level 5 and above who have the potential to receive three level promotions within five years. The second inventory includes persons from levels 2 through 12 who are promotable at least one level. The third inventory contains names of employees who have requested transfers to other departments.

67. In other instances, the personnel department is not involved in the selection of candidates. For example, in the case of an opening in a highly technical division such as underwriting, the candidate may be selected for promotion from within the department itself by the department manager. In such an instance, defendant's personnel department would have little or no input into the selection process. Thus, if the manager and supervisor are able to fill a vacancy by means of a promotion within the department itself, there is often no necessity for involving personnel.

68. As set forth more fully herein, the principal thrust of plaintiff's case is the lack of promotions to black employees. Central to plaintiff's complaints regarding defendant's promotional policies is the importance of the supervisor's recognition and recommendation of an employee, which plaintiff insists is fraught with the abuses attendant to any system based on subjective evaluations.

In Wade v. Mississippi Cooperative Extension Service, 528 F.2d 508 (5th Cir. 1976), the Court of Appeals criticized three aspects of the defendant's employee evaluation form:

(1) "the questions on the evaluation form were in large part subjective and vulnerable to either conscious or unconscious discrimination by the evaluating supervisors. Cf. United States v. Texas Education Agency, 459 F.2d 600, 606 (5th Cir. 1972);" see also Crawford v. Western Electric, supra, at 1315.

(2) "the evaluation scores themselves were not consistently used as a basis for . . . promotion . . .;" and,

(3) "the defendants wholly failed to make a showing that the test was substantially related to the particular jobs of the individuals being evaluated."

528 F.2d 508, 518. Further, in Rowe v. General Motors Corp., 457 F.2d 348 (5th Cir. 1972), the Court of Appeals found several other conditions in a promotional system reflecting Title VII violations:

(1) "[t]he foreman's recommendation is the indispensable single most important factor in the promotion process, [but he is] given no written instructions pertaining to the qualifications necessary for promotion;"

(2) "those standards which were determined to be controlling are vague and subjective;"

(3) "[h]ourly employees are not notified of promotion opportunities nor are they notified of the qualifications necessary to get jobs;" and,

(4) "there are no safeguards in the procedure designed to avert discriminatory practices."

457 F.2d 348, 358-59.

69. The Court concludes that the evaluation system utilized by defendant avoids the pitfalls of Wade and Rowe in the following respects: First, the categories on the evaluation form are specific and sufficiently capable of objective evaluation to be unlikely pretexts for discrimination. Adams v. Reed, 567 F.2d 1283, 1286 (5th Cir. 1978). For example, in evaluating an employee's production, the supervisor is instructed to consider whether the employee "[c]ompletes as much work as expected from a fully trained employee on this assignment [and] assigns proper priorities to work." See Defendant's Exhibit 20 at page 2. In evaluating an employee's dependability, the supervisor is instructed to consider number of days of absenteeism for the past twelve months, number of incidents of tardiness, the amount of supervision required, and his ability to follow company rules. See Defendant's Exhibit 18 at p. 1. Thus, the evaluator is directed to specific questions in order to appraise an employee's performance in a specific area. Therefore, although the supervisor's evaluation plays a significant role in an employee's promotion, the supervisor is given numerous, specific guidelines to be used in evaluating an employee.

70. Further, there are several safeguards in the evaluation process designed to avert discriminatory practices. First, the evaluations are reviewed by higher level management and by the personnel division. See note 10, supra. The testimony indicated that in at least one instance involving plaintiff, a supervisor's evaluation was revised in the employee's favor after consultation with the manager. Further, the employee himself is afforded an opportunity to review and discuss the supervisor's evaluation and to add his own comments. Id. Thus, an employee is apprised of his own strengths and weaknesses, as well as the standards by which he will be judged for promotion. Finally, the supervisor's evaluation is not the single determinative factor in promotion; the department manager and, in appropriate instances, the personnel department participate in the decision.

71. In addition, defendant has demonstrated through its evidence the relevance of the evaluation in determining an employee's eligibility for promotion. First, it is self-evident that an employee's ability to perform the functions of one job in a department bears heavily on his fitness to advance to a more responsible position, as well as his ability to train and supervise others in his former position. Further, factors such as dependability, relationships with others and supervisory ability are, by definition, pertinent to an employee's ability to perform as a supervisor or assistant supervisor. Also, the evidence clearly demonstrated that current employee evaluations are, in fact, utilized in selecting employees for promotion.

72. Finally, although vacancies are not posted, the testimony amply demonstrated that all employees within a particular division in the promotable level are considered for advancement. If a promotion is to be made from outside the department, personnel maintains sufficient inventories to ensure that all candidates are considered. Thus, even if defendant's promotional system can be faulted pro forma for a failure to post promotional openings, the Court finds from all of the relevant evidence that adequate safeguards are present to preclude any discriminatory impact of this procedure. Therefore, the Court concludes that defendant's system of evaluating employees avoids the hazards proscribed by Wade and Rowe.

73. Nonetheless, subjective evaluation procedures such as those utilized by defendant are not per se violative of Title VII; in order to fall within those practices proscribed by the Act, such procedures must result in discrimination against the protected class. Hester v. Southern Railway Co., 497 F.2d 1374, 1381 (5th Cir. 1974). Thus, courts must look beyond the form of the practice and the employer's motivation to the consequences of the procedures. Griggs v. Duke Power Co., 401 U.S. 424, 432, 91 S. Ct. 849, 854, 28 L. Ed. 2d 158 (1971); Rowe v. General Motors, 457 F.2d 348, 355 (5th Cir. 1972). Therefore, in order to determine whether defendant's policies violated Title VII or Sec. 1981, the Court must examine the effects of these policies on the protected class.

As stated hereinabove, in order to demonstrate the disparate impact of defendant's policies, plaintiff offered statistical evidence in four areas: salary, utilization, promotions and terminations.

7. Salary Analyses

74. Plaintiff's evidence illustrating salary differences between Whites and Blacks consists entirely of a comparison of the 1978 average salaries of Blacks and Whites hired in the same years, 1973 through 1978. Plaintiff's Exhibit 6. This comparison indicates an average weekly salary differential between black and white employees of $31.72 per week for employees hired in 1973; $4.45 per week for 1974 hires; $23.92 per week for 1975 hires; $17.71 per week for 1976 hires; and $20.62 per week for 1977 hires. However, plaintiff's exhibit is a naked comparison of average weekly salaries without regard to level of skill, education and training. No meaningful conclusion can be drawn from any analysis which does not take into account these factors. See Roman v. ESB, Inc., 550 F.2d 1343, 1355 (4th Cir. 1976). More significantly, in failing to consider the level at which each employee was hired, the validity of plaintiff's exhibit is contingent on the assumption that the average salary of all races should be the same, regardless of level of hire. Such a mechanical application of figures can result in no meaningful conclusion, much less a prima facie case. Accordingly, the Court finds that plaintiff has failed to establish a prima facie case of discrimination in the area of salaries.

75. The Court believes that the proper inquiry in an analysis of salaries by race should focus on whether black and white employees with the same tenure at the same job level are paid the same salaries. In an effort to demonstrate that a valid analysis on this basis would militate towards the conclusion that Blacks are not discriminated against in salaries, defendant presented two studies.

Defendant's expert, Dr. Ira Chorush, performed two regression analyses^[11] to determine whether black and white employees, similarly situated with respect to job level and tenure, were paid equally.^[12] Two separate regression analyses were performed for two calendar years: 1974, the first full year in which liability might be found, and 1978, which was the most recent year for which the full year statistics were available. Dr. Chorush utilized the same procedures for each year, proceeding with the 1978 analysis as follows:

76. All 1978 employees up to and including level 60^[13] were coded with respect to annual earnings, job level, tenure and race. Computer cards were keypunched from such data, and white and black employees were separated. White employees were then subdivided into two equal groups by random selection using a random number table.

The first group of white employees was subjected to regression analysis, wherein salary was predicted as a function of job level and tenure.^[14] A multiple correlation coefficient of .925 was obtained, which indicated that job level and tenure furnished a virtually complete explanation of salary among these white employees.^[15] The regression equation so developed, see Defendant's Exhibit 72, was applied to the second group of Whites to determine how well the equation could predict the salaries of an independent group of employees. The multiple correlation coefficient in this case was .918, indicating that the equation was valid for all 1978 white employees. Since the ability to predict salary had thus been established by the use of two randomly selected groups of white employees in 1978, the two groups were combined, and a single regression equation developed.^[16]

77. To address the question whether defendant had consistently and equitably applied the same practices to black employees, the salary prediction equation thus derived was used to predict each black employee's salary.^[17] Having thus determined the salaries which should have been paid to Blacks assuming that the employer applied the same salary policies to Blacks as it did to Whites, Dr. Chorush compared each black employee's actual salary with his predicted salary. The average of these differences is depicted in Defendant's Exhibit 71, which is entitled "Comparison of Observed and Predicted Salaries for Black Employees in 1974-1978." As indicated by this exhibit, the average black salary for 1974 was $72 less than the predicted average salary for this group, and the average salary of black employees for the year 1978 was $154 more than that predicted for these employees.

Dr. Chorush testified that the difference of $72 obtained in 1974 was not statistically significant inasmuch as a difference between actual and predicted salaries of this small a magnitude would occur so frequently by chance that statisticians would be unwilling to say that it is attributable to any systematic difference between actual and predicted salaries.^[18]

For the year 1978, black employees received $154 more than would have been predicted by the regression equation. Dr. Chorush testified that this difference was statistically significant; that is, the probability of a difference of this magnitude occurring by chance is so small that statisticians would say that it represents a meaningful disparity between the salary treatment of black and white employees (in favor of Blacks). Dr. Chorush testified that the odds against a difference of this size occurring by chance are greater than 100 to 1.

According to defendant's expert, the regression analysis of pay practices shows that defendant did not follow any practice which resulted in salary disparities based upon race in 1974. He testified that defendant's pay practices had resulted in a significant disparity in 1978 in favor of black employees. He further stated that although he did not perform any analyses for the intervening years, in his opinion, he would have expected to obtain results similar to those represented within Defendant's Exhibit 71.

Thus, based on the analyses performed by defendant's expert, the Court finds that in 1974 black employees were being compensated in essentially the same manner and to the same extent as white employees with the same job level and tenure and in 1978 were more highly compensated than white employees with the same job level and tenure.

78. Finally, in order to determine whether Blacks and Whites at the same level have obtained comparable salary increases in a given year, Judith Saul, EEO coordinator in defendant's corporate office and an expert in computer programming, prepared Defendant's Exhibit 59 by having a computer examine all personnel files and sort employees according to race, tenure, job level and salary.^[19] This exhibit illustrates the average amount and percentage of salary increases granted to employees, by race and level, for the years 1973 through 1978. These figures indicate overall salary trends, by race, because they are computed without regard to seniority, level of education, previous training or the employee's evaluation. In addition, in summary of each year's activity, the exhibit displays the amount and percentage of the average annual increase for Blacks and Whites.

The statistics for average increases for each job level during each year demonstrate that Blacks' average salary increases exceeded Whites' average salary increases in the same level in forty-five instances, whereas the reverse was true in only twenty instances. Further, the yearly average increases for Blacks exceeded those given to Whites in every year. Therefore, the Court finds that during the period from 1973 through 1978, defendant was awarding greater salary increases to its black employees than to its white employees.

Thus, the Court finds that defendant has established that its salary policies and practices do not impact unfavorably on the class.

8. Promotion Analyses

Plaintiff's proof relative to discrimination against Blacks in promotions was presented by two methods: first, plaintiff catalogued the number and percentage of promotions given to employees, by race, at levels 10 through 84 for the years 1973 through 1978. Plaintiff's Exhibits 4a-4i. Plaintiff then presented the average number of years of seniority possessed by employees promoted to levels 5 through 12 for the years 1973 through 1978.

79. The basis of a disparate impact claim with regard to promotions is the inference which may be drawn from a low percentage of minorities having been selected for promotion from a pool of persons presumptively qualified for promotion. Thus, of primary importance in analyzing promotional statistics is the determination of the relevant labor pool. See Neloms v. Southwestern Electric Power Co., supra, at 1368. It being undisputed that defendant follows a policy or promotion from within, the relevant labor pool for promotions is defendant's own workforce. See Plaintiff's Response to Defendant's Supplemental Post-Trial Brief at p. 2; Defendant's Second Supplemental Post-Trial Brief at page 2.

80. Within defendant's own workforce, there will be, of necessity, definite groups of employees who are eligible for certain promotions, while other employees without the requisite experience and training will be ineligible for the same promotions.^[20] Thus, there was considerable evidence adduced at trial that employees promoted to one level within one division are generally drawn from employees in another particular level within that division.^[21] For example, the uncontroverted testimony at trial established that managers (level 60) are drawn exclusively from the ranks of associate managers (level 20). Therefore, in evaluating defendant's record of promoting employees to level 60 in a particular year, one must consider such promotions in light of those who were available for promotion to level 60, i. e., those present at level 20 when the promotions to level 60 were being effected. Further, because the company would naturally be eager to utilize the experience obtained by an employee while he worked in a particular division, it is logical to assume that when a position in one division becomes open, those considered for promotion to that position will be persons within that same division. Therefore, as is more fully delineated hereinafter, promotion awards in each level and area must be carefully analyzed in the context of the relevant promotional pool inasmuch as the question of whether a prima facie case has been established depends on "the demonstration of a disparity between a minority's presence in a pool qualified for a position and its representation in the group selected for that position. . . ."^[22]

a. Plaintiff's Promotion Utilization Exhibits

81. Plaintiff's Exhibits 4a through 4i, which depict promotions to levels 10 through 84 by year and race, are devoid of reference to the applicable promotional pool. In support of her conclusion that these exhibits demonstrate that defendant's promotional policies have a disparate impact on Blacks, plaintiff's expert, Ms. Rose Battle, compared the percentage of black promotions to each level to the percentage of Blacks in defendant's entire workforce in the year in which these promotions were awarded. By thus ignoring the pool from which promotions to each level are drawn, plaintiff implies that all persons within defendant's workforce are equally qualified for promotions to all levels. Such an assumption is unsupported by the evidence in this case. Neither does the case law support such an analysis or the assumption on which it is based. See Hazelwood, supra, 433 U.S. at 308, n. 13, 97 S.Ct. at 2742, n. 13; Williams, supra, 607 F.2d at 692-93.

82. Nor can a Title VII plaintiff make a prima facie case through such statistics.

"[T]he prima facie showing has meaning only in terms of the relevant labor or applicant pool, as Albemarle and Ironworkers indicate."

Lim v. Citizens S. & L. Ass'n, 430 F. Supp. 802, 807 (N.D.Calif.1976), citing Albemarle Paper Co. v. Moody, 422 U.S. 405, 425, 95 S. Ct. 2362, 2375, 45 L. Ed. 2d 280 (1975) and United States v. Ironworkers Local 86, 443 F.2d 544, 551 (9th Cir. 1971). Particularly when the jobs require a high level of qualifications, the issue of the pool of qualified employees is critical in determining the relevance of statistical comparisons. Bilingual Bicultural Coalition of Mass Media v. FCC, 492 F.2d 656, 658 (D.C.Cir.1974).

The misleading nature of an analysis which is thus compiled without regard to the pool of qualified candidates can be simply illustrated as follows: Plaintiff's Exhibit 4G indicates that in 1973 there were 9 promotions to level 20, eight of which (88.8%) were awarded to Whites and one of which (11.1%) was awarded to a Black. When the 11.1% black promotion rate is compared to the 13.4% black representation in defendant's entire workforce in 1973, it appears to demonstrate that Blacks receive less than their share of promotions to level 20. However, in 1973 Blacks constituted 2% of the employees at level 11, from which level promotions to level 20 are drawn. Defendant's Exhibit 60. Thus, Blacks actually obtained approximately five and one-half times the number of promotions as would have been expected from their representation in the appropriate labor pool.

83. Defendant's Exhibit 43, which was compiled from data contained in plaintiff's promotion (Plaintiff's Exhibit 4) and utilization (Plaintiff's Exhibit 3) exhibits, demonstrates that when promotions are considered in light of the prior year's utilization, black employees received promotions in a percentage substantially higher than their representation in the levels from which these promotions are drawn.^[23]

b. Plaintiff's Exhibits Regarding Service Years Prior to Promotion

84. Plaintiff's Exhibits 7a-7e indicate the average number of years of service of white and black employees in levels 5 through 11 prior to their last promotion, and, when taken at face value, would appear to demonstrate that Blacks had a greater number of service years prior to their last promotion than Whites.

However, the Court finds this exhibit to be of little probative value for several reasons: first, this exhibit does not reflect the level of hire of the employees whose promotion rates are compared. Such an omission could result in misinterpretation of data. As an example, plaintiff's Exhibit 7c indicates that Blacks averaged 7.16 years of service prior to promotion. But if these same white employees had been college hires (hired in at level 8) and had taken an average of 5.8 years to advance one level, and the black employees had been entry level hires (level 2)^[24] and had required an average of 7.16 years to advance 7 levels, the same data would indicate that Blacks actually enjoyed a more rapid rate of advancement than Whites.

Thus, as defendant's expert testified and as these contrasting results demonstrate, no conclusions can be drawn from such an analysis of years prior to promotion, without more, because its results cannot be attributed to any one cause—in this instance, it cannot be determined whether a certain number of years prior to promotion results from the relative level of hire or rate of promotion.^[25]

85. In addition, plaintiff has omitted from this series of exhibits (Plaintiff's Exhibits 7a-7e) employees hired prior to 1965. Doubtless, these pre-Act whites, if present at a given level, competed with others present at that level for the available promotions.^[26]See Finding of Fact No. 57.

86. Accordingly, the Court finds that plaintiff's exhibits concerning promotion utilization are of no probative value because they provide no meaningful analysis in light of the applicable pool of employees eligible for promotion. Further, plaintiff's exhibits concerning years prior to promotion are not probative of discrimination in that they fail to consider the relative level of hire of the promoted employees. In sum, plaintiff has failed to make a prima facie case of the disparate impact of defendant's promotional policies on blacks.^[27]

c. Defendant's Promotion Statistics

Defendant presented statistical analyses of promotions in three areas: analysis of promotional trends (Defendant's Exhibits 57 and 58); a study of the average time taken by white and black employees to reach the various job levels (Defendant's Exhibits 62a and 62b);^[28] and, finally, an analysis of all promotions awarded by defendant during the applicable period of this lawsuit (Defendant's Exhibits 69 and 70).

87. Defendant's Exhibit 57, entitled "Summary of Advancement Percentages, 1973-1978," was compiled by defendant's EEO coordinator, Judith Saul, an expert in computer programming, who had the computer sort employee information from defendant's personnel files. Defendant's Exhibit 57 illustrates for the years 1973 through 1978 the number of white and black employees employed in the Administrative Section of defendant's Southwestern Home Office, and the number and percentage of those employees, by race, who received promotions in that year. Thus, this exhibit illustrates the relative rates of promotion for all black and white employees in defendant's administrative section during the time period relevant to this lawsuit.

Defendant's Exhibit 57 demonstrates that during the years 1973 through 1978, the percentage of black employees receiving promotions was higher than the percentage of white employees receiving promotions. For example, in 1973, the first year of the liability period, 66 of the 163 blacks in defendant's workforce (or 40.49%) received promotions, while 306 of the 982 whites (or 31.16%) received promotions. Likewise, in 1978, the last full year prior to trial, 192 of the 443 black employees (or 43.34%) received promotions while 358 of the 1088 whites (or 32.9%) received promotions. In the intervening years, similar results were achieved. In 1974, 43.52% of defendant's black employees were promoted and 33.07% of defendant's white employees; in 1975, 41.98% of black employees and 32.09% of white employees were promoted. Likewise, in 1976, 52.4% of black employees and 40.31% of white employees received promotions; in 1977, 43.53% of black employees and 35.42% of white employees were promoted.

Stated another way, when promotions at every level are analyzed, a black employee in defendant's workforce had a greater probability than a white employee of receiving a promotion.^[29]

88. Defendant's Exhibit 58 sets forth the same promotional data separated into the major segments of the workforce: levels 2 through 6 (clerical); levels 7 through 12 (advanced clerical, technical and supervision); and levels 20, 60 and 77 (management). Again, the number of white and black employees in the three segments and the number and percentage, by race, of those employees in those segments who received promotions were set forth. In all but two instances, the percentage of black employees by level groups receiving promotions was higher than the percentage of white employees in that same level group that received promotions.^[30] Further, Defendant's Exhibit 58 negates any inference that the favorable statistics concerning black promotions in Defendant's Exhibit 57 resulted from favorable treatment for Blacks in lower levels balanced against their discriminatory treatment at upper levels of defendant's workforce.

89. Second, defendant's expert analyzed promotions within the Administrative Section of defendant's Southwestern Home Office for the years 1973 through 1978. This analysis was performed to ascertain whether discrimination in awarding promotions has occurred by determining whether similarly situated persons who were eligible for promotion received or failed to receive promotions according to their race. In compiling the relevant pool for each promotion, Dr. Chorush included all employees with over six months of service at Prudential^[31] who were employed within the division in which the vacancy occurred in the level from which the promotion would be made.^[32] Neither job performance, time in position, nor overall tenure^[33] (other than the six months' minimum service) was considered.

After thus compiling the applicable pool for each promotion awarded, Dr. Chorush determined whether Blacks received an appropriate share of promotions as follows: First, he calculated the probability of being promoted from one level in a particular division during a particular year by comparing the number of promotions from that level to the number of employees in that level in that department. For example, if there were five promotions given to all employees at level 4 in a particular division, and if there were ten total employees in level 4 in that same division, then the probability of being promoted was evidently .5. This probability figure was then compared to the number of black employees at that level in the department. Thus, if there were four black employees at Level 4 in this department, one would have expected one-half, or two employees, to have been promoted. This assumption yields a result which would have been obtained by a company which promoted at random and was, therefore, non discriminatory by definition. In this example, if three black employees were actually promoted, then Blacks would have received one more promotion than expected; if only one black employee was promoted, then Blacks would have been "shortchanged" by one promotion.

By carrying this analysis through all levels of a division and totalling the results, a division total of the excess or shortfall in black promotions could be obtained. By further adding the results across all divisions, an annual total could be obtained.

Defendant's Exhibits 69 and 70 illustrate the results of this analysis. Defendant's Exhibit 69 compares the total promotions given to black employees with the number which would have been given to Blacks by a company which promoted at random, while Defendant's Exhibit 70 shows this information for level 9 and above.

Defendant's Exhibit 69 shows that black promotions for the years 1973-1978 were not statistically different from what one would have expected based upon their representation within the divisions of the workforce.^[34] As Dr. Chorush testified was likely to occur, there are some natural, random fluctuations during this six-year period. In the first three years (1973, 1974 and 1975) Blacks received slightly more promotions than would have been predicted within a random system; there were also three years (1976, 1977 and 1978) in which black employees received slightly fewer promotions than would have been expected. The differences, in instances in which Blacks were favored as well as those in which they were not, are so small that they would frequently be expected to occur by chance. (See column reflecting "t" factor.) Therefore, they cannot be attributed to any systematic effect.

Out of all of the promotions granted from 1973 to 1978, black employees received a total of 734 promotions, while Dr. Chorush's analysis predicted that, had employees been promoted within divisions purely at random, black employees would have received a total of 734.62 promotions, a difference of .62 promotions, which Dr. Chorush testified was obviously not statistically significant.

90. Further, in order to determine whether the results demonstrated by Exhibit 69 were brought about by an abundance of lower level black promotions as balanced against a deficiency in upper level black promotions, Dr. Chorush studied promotions to level 9 and above. The same methods employed in obtaining Defendant's Exhibit 69 were utilized, and the results of this study are set forth in Defendant's Exhibit 70.^[35]

In a result similar to the analysis of promotions in the workforce as a whole, during three of the years in which upper level promotions were analyzed (1974, 1977 and 1978), black employees received slightly fewer promotions than would have been expected while in the other three years (1973, 1975 and 1976) black employees received slightly more promotions than would have been predicted based upon their representation in the relevant promotional pools.

For the entire workforce at levels 9 and above, black employees would have been expected to obtain 45.99 promotions from 1973 to 1978. These persons actually obtained 48 promotions or 2.01 more promotions than would have been predicted. No "t" statistic was listed in Defendant's Exhibit 70 because the numbers were so small that no meaningful test of the significance of differences could be performed.

The Court therefore finds that when the appropriate promotional pools are considered, the statistics clearly demonstrate that Blacks received as many promotions as they would have received in a non-discriminatory system. Therefore, the Court concludes that defendant's promotional policies have not resulted in a discriminatory impact on defendant's black employees.

9. Utilization Statistics

91. Utilization statistics are a compilation of the percentage representation of white, black and other minority employees at the various levels of defendant's workforce in each year; they provide a "snapshot" picture of the workforce at a given moment in time.

Plaintiff offered its exhibits illustrating utilization by race in an effort to demonstrate that Blacks are underrepresented in upper levels of defendant's workforce when compared both with their percentage in defendant's total workforce and with the workforce as a whole.

Plaintiff's Exhibit 3 reveals that Blacks occupied the following percentages of positions in each level during the years 1973 through 1975.

           1973    1974   1975
Level 1    33.3%   21.4%   --[*]
Level 2    31.2    35.2   28.6%
Level 3    16.0    26.9   24.5
Level 4    20.3    21.8   23.9
Level 5    19.9    19.3   22.2
Level 6    24.4    21.8   23.8
Level 7    12.3    21.2   16.7
Level 8    14.9    21.1   24.6
Level 9     4.9    10.7   11.6
Level 10    0.0     8.0   12.5
Level 11    2.5     2.4    4.2
Level 12   10.0     7.7    0.0

Plaintiff's Exhibit 8 indicates that Blacks occupied the following percentage of certain job categories in the years 1973 through 1975.

                   1973   1974   1975
Officials and
Managers            0.0%   0.7%   1.4%
Professionals       0.0    7.1    7.6
Technicians         7.2    8.7   13.2
Office—Clerical    19.5   21.7   23.7
[*][**]Service Workers    30.9   34.8   52.0

92. The Fifth Circuit has held that a Title VII plaintiff may establish a prima facie case of employment discrimination by demonstrating a substantial disparity between the percentage of Blacks in a specific job classification and their percentage representation in the workforce. Robinson v. Union Carbide Corp., 538 F.2d 652, 660 (5th Cir. 1976); Sagers v. Yellow Freight System, Inc., 529 F.2d 721, 729 (5th Cir. 1976). Thus, in any category in which black representation falls substantially below 18.2%, which is their percentage representation in the workforce, plaintiff has established a prima facie case.

93. The burden then shifts to the defendant to establish that the discrepancy results from causes other than racial discrimination, or to otherwise explain the disparity. Id. Defendants may carry this burden by the use of "more refined, accurate and valid statistics." Movement for Opportunity and Equality, Inc. v. General Motors Corp., 622 F.2d 1235, 22 EPD ¶ 30, 863 (7th Cir. 1980), citing Furnco Construction Corp. v. Waters, 438 U.S. 567, 576-78, 98 S. Ct. 2943, 2949-50, 57 L. Ed. 2d 957 (1978); Teamsters v. United States, 431 U.S. 324, 339, 360, 97 S. Ct. 1843, 1856, 1867, 52 L. Ed. 2d 396 (1977). Accordingly, defendant introduced statistical evidence reflecting its employment actions taken during the years relevant to this lawsuit.

94. As emphasized by the Supreme Court in Hazelwood School District v. United States, supra, utilization statistics, standing alone, are not necessarily probative of discrimination because they reflect the results of many years of personnel decisions in the area of hiring, promotion and termination. Id. at 309, 97 S.Ct. at 2742; see also Movement for Opportunity and Equality, supra. For example, utilization statistics relative to defendant's workforce in 1974 will include many employees who were hired and promoted by defendant during many years prior to 1974.

Therefore, in searching for evidence that is actually probative of discrimination, most Courts look to statistics reflecting defendant's actions during the proper time frame rather than static work force statistics. See, e. g., Movement for Opportunity, supra; EEOC v. Local 14, Operating Engineers, 553 F.2d 251, 255 (2nd Cir. 1977); Swint v. Pullman-Standard, 539 F.2d 77, 105 (5th Cir. 1976); United States v. Strickland Transportation Co., 17 FEP 1081, 1083 (N.D.Tex.1978); Neloms v. Southwestern Electric Power Co., supra; Croker v. Boeing Co., 437 F. Supp. 1138, 1184-1186 (E.D.Pa. 1977).

Accordingly, the Court should look beyond the raw data of utilization to examine employment decisions which have taken place during the time period relevant to this lawsuit, i. e., for purposes of Title VII, the period beginning October 13, 1974, and for purposes of § 1981, on or after October 31, 1973. See Findings of Fact 54 and 55. For purposes of analyzing employee utilization, the change in utilization which has occurred during the relevant time period will be probative. See, e. g., Green v. McDonnell Douglas Corp., 528 F.2d 1102, 1105 (8th Cir. 1976).

95. Defendant's Exhibit 42, which was constructed from Plaintiff's Exhibit 3, analyzes utilization during the relevant time frame by presenting the change in utilization from 1973 to 1975. Defendant's Exhibit 42 indicates that during the period from 1973 to 1975, Blacks received 46.9% of the increases in positions in levels 2 through 12.

In addition, Defendant's Exhibits 61a and 61b demonstrate that in the years 1974 through 1978, defendant has hired Blacks into positions from level 1 to manager at a percentage rate that exceeds their 18.2% representation in the work force, with Blacks participating in a relatively steady increase in the percentage of hires from 1973 to 1978. In fact, the average percentage of Blacks hired during 1973 through 1978 is 28.6%, some 10% above the percentage representation of Blacks in defendant's workforce and in the workforce as a whole. Thus, the change in utilization of Blacks as evidenced by their increasing percentage representation within all levels of defendant's workforce is inconsistent with any policy of discrimination during the time period relevant to this law suit.

The following example illustrates the results of a company's non-discriminatory hiring policies on a previously all-white workforce. Assume that a company has discriminated against black persons by refusing to hire them but, at some point, entirely reverses its policy of discrimination and begins to hire black persons in proportion to their representation in the surrounding community. Assume further that at the point when the employer becomes non-discriminatory, its workforce consists of 1000 employees who are, of course, all white. Assume a hiring rate of 50 employees per year (5%) from a community workforce which is 20% black. Then the employer's progress over the next five year period would be as follows:

              Workforce        Hires
            White  Black    White  Black
1st Year    1000     0       40      10
2nd Year    1040    10       40      10
3rd Year    1080    20       40      10
4th Year    1120    30       40      10
5th Year    1160    40       40      10

Thus, at the end of five years, the hypothetical employer would have a workforce of 1200 employees of whom 40 would be black. This would represent a utilization of 3.3% in a community in which the workforce was assumed to be 20% black. Nevertheless, this hypothetical employer would not have been guilty of discrimination during the past five years. Hazelwood v. United States, supra, 433 U.S. at 309, 97 S.Ct. at 2742. See also Schlei and Grossman, supra, at p. 1182.

Further, assuming the company promoted from within and administered its promotional program in a non-discriminatory manner during the same five-year period, the black employees, being comparatively recent hires, would be expected to occupy relatively lower positions within the employer's workforce than their white co-workers who had had a greater amount of time to move up through the various lines of progression.

96. The operation of this percolation effect within defendant's workforce is illustrated by Defendant's Exhibits 60 and 76. Defendant's Exhibit 60 shows the representation, by number and percentage, of black employees at levels 2 through 60 during 1973 through 1978, as well as the percent of percentage increase of black representation by level.^[36] For example, in 1973 Blacks comprised 12% of the level 7 work force in 1973 and 27% of the level 7 work force in 1978, thus yielding a percentage increase of 225%. Similar increases were experienced by Blacks at all other levels, ranging from a 129% percentage increase at level 6 to 800% at the associate manager level.

Granted, black representation decreases as the job level is ascended. See Defendant's Exhibit 60. However, as the years progress, black employees have percolated up through the job level system so that in succeeding years the job level at which black representation falls below their percentage representation in the community work force becomes higher. See Defendant's Exhibit 76.

In any event, the touchstone of a utilization analysis is not whether the employer's workforce at all levels mirrors the racial makeup of the community. See Hazelwood v. United States, supra, at 308, n. 13, 97 S.Ct. at 2742, n. 13, citing Teamsters v. United States, supra. Rather, the crux of the matter is the progress made by the employer during the relevant time period.

97. Defendant's expert performed several analyses and prepared a mathematical model in an attempt to evaluate the progress made by defendant during the liability period as measured against the performance of the model, which represents a company with a non-discriminatory system. The model was designed with a workforce makeup identical to defendant's in 1973, with a turnover rate by level identical to defendant's.

The only difference between the operation of the mathematical model and the actual operation of defendant's personnel practices would be that defendant maintains that its personnel decisions are made on the basis of merit, while the mathematical model makes these decisions at random, on the basis of relative representation within the workforce. Assuming "merit" to be equally distributed between the two races and given enough personnel decisions, roughly equivalent results would be expected.

In order to construct the model, Dr. Chorush first calculated, by level, the average hiring rate and average rate of terminations, each as a proportion of the number of employees at that level in previous years. See Defendant's Exhibit 66. In addition, he took into account the rate of growth of the number of jobs at each level between 1973 and 1978.

With regard to the labor force from which new hires would be drawn, Dr. Chorush made the following assumptions: Hires into levels 8 and above would be made from a labor force that is 6.5% black,^[37]see Defendant's Exhibits 73a, 73b, 75 and 77, and hires into levels 4 through 7 would be from a labor force that is 18% black, see Defendant's Exhibits 74 and 75. Based on Defendant's hiring experience during the years in question, Dr. Chorush assumed that 40% of the level 2 hires and 35% of the level 3 hires would be black. Defendant's Exhibit 75.

The model thus constructed operated as follows: First, for each level, hires according to assumption were made. That is, at level 60, for example, if the hire rate were 5% and there were 40 incumbents at level 60, the model would hire 5% of 40, or two people. By assumption, 6.5% of these hires would be black.

Next, terminations from level 60 would be made. If the termination rate were 2%, then 2% of the white employees and 2% of the black employees would be removed.

Finally, sufficient employees would be promoted to level 60 (from level 20, in this case) to obtain the necessary number of personnel at level 60. These promotions would be given to Whites and Blacks in proportion to their representation in the level below. See Defendant's Exhibits 67a and 67b.

Dr. Chorush wrote a computer program to carry out these calculations, which would otherwise be rather lengthy. The model, starting with 1973 actual data, computed expected 1974 utilization, which was then compared with actual 1974 utilization. Expected 1975 utilization was then computed from expected 1974 utilization, and so on. As a by-product of these calculations, the program also computed the percentage of promotions at each level in each year that ought to be received by black employees, which figures were obtained through the natural operation of the workforce.

The results of the comparison of defendant's performance in the area of utilization with the model's predictions are illustrated by Defendant's Exhibits 65a-e.^[38] Of the fifty comparisons made (ten levels during each of five years), the company's actual performance exceeded that predicted by the model in thirty-six instances; in the remaining fourteen instances, the company's utilization of black employees was less than that predicted by the model.

In some instances where company performance did not measure up to that predicted by the model, however, examination of the degree of percentage difference in light of the number of persons involved reveals that this disparity is not alarming. For example, black representation in the manager (level 60) category for 1974 (0.00%) was less than that predicted by the model (0.07%), i. e., the company had no black managers while the model predicted .07% black managers. However, since the total number of persons within this category was 36, the total number of persons predicted by the model to be occupying the manager category was .025 (36 × .0007). Thus, if the company had had as many as one black manager, this would have been 40 times as many as predicted by the model.^[39] In fact, throughout the four year period of 1974 through 1977 when the actual utilization for the manager category was less than that predicted by the model, the company would have been greatly in excess of the model percentages if it had had as many as one black manager. In other cases, the numerical disparity between the number of Blacks predicted by the model and the number of black employees actually present in that level was very slight. For example, the model predicted that 3.33 Blacks would be present at level 11 in 1974, while only two were present, a difference of only 1.33 black employees.^[40] In view of this slight degree of underrepresentation in at least four of the instances in which defendant's utilization was less than the model predictions, as well as the numerous instances in which company performance greatly exceeded that predicted by the model, the Court finds that defendant has affirmatively established by the use of accurate statistics that its personnel policies do not have a discriminatory impact on Blacks.

98. Defendant's Exhibits 64a-e compare the percentage of promotions actually awarded by defendant to black employees in level 10 through 60 with the model predictions of percentage of promotions which would have been expected for black employees in a non discriminatory company. This date is given for each year, and there is shown a weighted average for all years.^[41] For example, Defendant's Exhibit 64a shows that for each of the years 1974-1978, black employees were promoted into level 10 at a higher rate than was predicted by the mathematical model. Although in some years the actual promotion rate does not equal the predicted rate, the weighted average promotion rate for black employees for the years 1974-1978 is more than twice that predicted by the model.

99. Finally, Dr. Chorush prepared Defendant's Exhibit 78, which shows predicted model utilization given a hiring assumption of 18% black persons in all levels^[42] in order to compare defendant's performance to a situation in which the highest possible number of blacks was available for hiring into all levels. Even under this overly generous, 18% across-the-board hiring assumption, defendant's actual utilization and promotion of black persons compares favorably with that predicted by the model.

100. The Court accepts the assumptions on which defendant's expert's mathematical model are based as reasonable and, if in any way erroneous, such error tended to favor the plaintiff. Accordingly, the Court finds that defendant's statistics concerning predicted black promotions establish that defendant's personnel policies do not unfavorably impact upon Blacks or prevent their progress into upper level positions. In fact, defendant's model illustrates the validity of the "percolation" theory as it impacts on a workforce in which there is a low turnover in upper level positions, see Defendant's Exhibit 66, to which it takes many years for any employee to progress, see Defendant's Exhibit 62a and 62b. As the Court stated in Lee v. City of Richmond, 456 F. Supp. 756, 769-770 (E.D.Va.1978), the absence of Blacks in upper level positions and their predominance in lower level positions do not, standing alone, establish racial discrimination when the upper levels have a low turnover and few black applicants.

In sum, although plaintiff has established a prima facie case by means of static utilization statistics, see Finding of Fact No. 92, the Court finds that defendant has conclusively rebutted it by comparing defendant's actual personnel-related performance during the years in question with that predicted by a validly constructed model of a non-discriminatory company whose workforce makeup, vacancies and promotional opportunities are identical to that of defendant. Thus, by the use of more refined, accurate and valid statistics, Movement for Opportunity, supra, defendant has demonstrated that the disparities in black representation at certain levels resulted not from racial discrimination or from the disparate impact of facially neutral policies on Blacks, but rather from the natural operation of the workforce. Therefore, the Court concludes that plaintiff has failed to demonstrate that, in the area of utilization, defendant has discriminated against Blacks.

10. Termination Statistics

As evidence of the allegedly discriminatory treatment received by Blacks, plaintiff offered evidence that Blacks are also discriminated against in the area of terminations.

In support of his general allegation of discriminatory treatment in terminations, plaintiff introduced only the evidence surrounding his own discharge. As set forth more fully hereinabove, see Findings of Fact Nos. 15 through 21, the Court has found that plaintiff's discharge was not the result of racial discrimination.

However, plaintiff also attempted to prove that defendant's policies, practices and procedures in the area of terminations have a disparate impact on black employees. As a consequence, the Court has reviewed the testimony concerning defendant's termination policies, as well as statistical evidence offered by plaintiff and defendant, in order to determine whether these policies have impacted adversely on the plaintiff class.

101. Defendant's termination procedures were fully ventilated by the testimony of several of defendant's witnesses. Terminations are monitored by the personnel division. Typically, a manager implements a termination decision. He consults the EEO coordinator and/or the director of personnel to review the basis for the decision and to evaluate it in terms of the merits of the individual case. He further ascertains whether company procedures have been followed and whether like treatment is being afforded members of both races in the manager's division. (The employee being terminated has often been reviewed previously if he or she has been on probationary (INR) status.) The personnel division has the authority to reverse line management in its decision to terminate an employee, and testimony indicated that this course of action has occurred in the past.

If, after exhausting all efforts to explore alternative approaches, a termination is necessary, the situation is again reviewed by the EEO coordinator and then independently reviewed by the director of personnel. In all cases of involuntary termination, an exit interview is conducted by the EEO coordinator or director of personnel. According to the testimony of the personnel director, even at such a late stage, a termination decision may still be reversed if new information comes to light. The Court finds these policies to be consistent with the procedures employed in the case of plaintiff's termination.

The Court finds that the termination procedures utilized by defendant are facially neutral. In addition, these procedures contain adequate safeguards such as warnings to the employee, see Bolton v. Murray Envelope Corp., supra, and an investigation and the opportunity for the employee to present his position, see Turner v. Texas Instruments, supra. Thus, the Court determines that defendant's termination policies neither manifest an intent to discriminate against Blacks nor, in and of themselves, appear to impact discriminatorily on Blacks.

102. By contrast, plaintiff also introduced Plaintiff's Exhibits 5 and 26^[43] in an effort to demonstrate the adverse impact of defendant's policies on its black employees. A comparison of the percentage representation of Blacks in defendant's workforce (derived from Plaintiff's Exhibit 28) to the percentage of terminations received by Blacks (Plaintiff's Exhibit 5) indicates that in four of the six years in question, Blacks received a percentage of total terminations higher than their percentage representation in defendant's workforce.^[44]

Plaintiff's Exhibit 5 demonstrates that there were very few terminations prior to 1976 (3 from October 3, 1973, to December 31, 1973, 6 in 1974, and 13 in 1975) and a marked increase thereafter (40 in 1976, 69 in 1977 and 27 in 1978). Frances Early testified that when she assumed the position of director of personnel in 1977, she found that managers were extremely reluctant to terminate employees, particularly minorities. Ms. Early testified that in her opinion, this was an unreasonable policy and unfair to employees of both races who were performing their jobs well. Therefore, she informed managers that involuntary terminations would be implemented where warranted and necessary. Thus, while the number of terminations of Whites, Blacks and others increased to a large degree in 1976, the total number of terminations nearly doubled in 1977 and decreased greatly in 1978.

103. Defendant's Exhibit 54 lists all involuntary terminations between 1973 and 1978 (except for those terminated for failure to report) and summarizes the review memorandum prepared by personnel for each termination. Even if plaintiff had established a prima facie case of discriminatory discharge in the case of each black employee listed, defendant has demonstrated a legitimate, non-discriminatory reason for terminating each employee through the information supplied in Defendant's Exhibit 54. Finally, plaintiff has failed to prove that Whites committing the same disciplinary infractions were not terminated or that the reasons established by defendant were otherwise pretextual. Accordingly, plaintiff has failed to prove that the members of the plaintiff class were the victims of discriminatory discharge.

104. The Court has also examined Plaintiff's Exhibit 5 and Defendant's Exhibit 54 with a view to whether they evidence a pattern of disparate impact of defendant's personnel policies on its black employees. First, the sample of involuntary terminations is not large enough that one can infer any probative trend in the numbers. See, e. g., Adams v. Reed, 567 F.2d 1283, 1287 (5th Cir. 1978); Ochoa v. Monsanto Co., 473 F.2d 318, 320 (5th Cir. 1973) (per curiam). In addition, a careful review of Defendant's Exhibit 54 reveals no pattern of discharges relatable to race; on the contrary, it is clear that both black and white employees were terminated for good cause, with reasons falling within the categories of poor performance, poor attendance and unacceptable conduct. Further, the higher percentage of black terminations is not altogether unexpected, since Blacks experience a proportionately higher representation in the lower, entry level positions, where most involuntary terminations occur. Such a result is logically attributable to the relative inexperience of employees in the lower levels, see Croker v. Boeing Co., 437 F. Supp. 1138 at p. 1188 (E.D.Pa.1977), along with the obvious reluctance of employees in upper level positions to engage in activities which would jeopardize their more responsible and lucrative positions. Finally, plaintiff has failed to rebut defendant's position that the higher percentage of black terminations is attributable to the number of Blacks newly hired. Id. at p. 1188. Accordingly, the Court finds that plaintiff has also failed to establish that defendant's termination policies have a disparate impact on the plaintiff class.

105. In arriving at the above Findings of Facts, the Court has weighed carefully the credibility of the witnesses who have testified and further finds that in instances wherein significant factual conflicts exist, the defendant's witnesses are more believable.

III. Conclusions of Law

1. The Court has jurisdiction over the parties and subject matter of this lawsuit pursuant to Title VII of the Civil Rights Act, 42 U.S.C. § 2000e et seq., and 42 U.S.C. § 1981.

2. Plaintiff has failed to establish a prima facie case that he was denied a promotion on the basis of his race. Finding of Fact No. 10. Even if plaintiff had established a prima facie case, defendant rebutted it by proving a legitimate, non-discriminatory reason for plaintiff's non-selection, Findings of Fact Nos. 11 and 12, and plaintiff has failed to rebut it by establishing that the reasons established by defendant were pretextual. Finding of Fact No. 12. Accordingly, plaintiff has failed to prove that he was denied a promotion on the basis of his race.

3. Plaintiff has failed to establish a prima facie case that he was terminated because of his race. Finding of Fact No. 17. Even if plaintiff had done so, defendant has adequately rebutted plaintiff's case by proving that plaintiff's termination was based on a legitimate, non-discriminatory reason, Finding of Fact No. 18, and plaintiff has failed to demonstrate that this reason was pretextual. Id. Accordingly, plaintiff has failed to demonstrate that he was discharged because of his race. See also Findings of Fact Nos. 19, 20 and 21.

4. None of plaintiff's witnesses, Esther DuPont, Emma Hagger, Murreba Harrison, Luretta Moses, Mary Byrd and Vernel Armstrong, offered any evidence that is probative of discriminatory treatment of Blacks within the defendant company. Findings of Fact Nos. 26, 30, 33, 37, 40, 43 and 45.

5. Plaintiff has failed to establish a prima facie case of discriminatory impact of defendant's policies on the class in the area of salaries. Finding of Fact No. 74. Defendant has affirmatively established that its policies and practices with regard to salaries do not result in a discriminatory effect on the class. Finding of Fact No. 78.

6. Plaintiff has failed to make a prima facie case of disparate impact of defendant's promotional policies on black employees. Finding of Fact No. 86. Further, defendant has conclusively established that defendant's promotional policies do not have a disparate impact on its black employees. Finding of Fact No. 90.

7. Although plaintiff has established a prima facie case of disparate impact in the area of utilization by means of raw data reflecting the racial makeup of various levels in defendant's workforce, Finding of Fact No. 92, defendant has conclusively rebutted it by more relevant statistics reflecting defendant's actual personnel decisions during the time frame relevant to this lawsuit. Finding of Fact No. 100. Further, defendant has affirmatively demonstrated that the disparities in racial makeup of certain levels result from forces other than racial discrimination or the disparate impact of facially neutral policies on Blacks. Id.

8. Plaintiff has failed to establish a prima facie case of racial discrimination in the area of terminations under either a theory of discriminatory treatment or disparate impact. Findings of Fact Nos. 101 and 104. Even if plaintiff had made a prima facie case, defendant has rebutted it, id., and plaintiff has failed to establish that the reasons articulated for the individual discharges or for the statistical results were pretextual. Id.

In the event that the foregoing Findings of Fact also constitute Conclusions of Law, they are adopted as such. In the event that the foregoing Conclusions of Law also constitute Findings of Fact, they are adopted as such.

Defendant is hereby instructed to file with the Court within ten (10) days an appropriate final judgment which incorporates by reference the foregoing Findings of Fact and Conclusions of Law.

NOTES

[1] See discussion of defendant's level system, section II.B.5, infra.

[2] Plaintiff complains that he, as the only Black in the duplicating section, was passed over for promotion because only Whites were involved in the selection process. The vacancy in the assistant supervisor position to which plaintiff aspired was created when Ms. Wanda Simms, a white female who had been the assistant supervisor, replaced Mr. E. J. Wilkins, the section supervisor, who transferred to another division. See Finding of Fact No. 5.

Ms. Simms's predecessor was Mr. Richard Kindle, Jr., a black male who was promoted to assistant section supervisor on July 7, 1967, upon the recommendation of Mr. Wilkins, the same supervisor who recommended Ms. Aschenbeck over plaintiff. Defendant's Exhibit 45. Mr. Kindle currently occupies a Level 60 position as the manager of defendant's data processing services division at defendant's Southwestern Home Office.

[3] A prima facie showing pursuant to McDonnell Douglas is not the equivalent of a factual finding of discrimination. "Rather, it is simply proof of actions taken by the employer from which we infer discriminatory animus because experience has proved that in the absence of any other explanation it is more likely than not that those actions were bottomed on impermissible considerations." Furnco Construction Corp. v. Waters, 438 U.S. 567, 579-80, 98 S. Ct. 2943, 2951, 57 L. Ed. 2d 957 (1978). When the plaintiff makes out a prima facie case under McDonnell Douglas, the burden shifts to the employer to prove that the employment decision made was based on a legitimate consideration and not an illegitimate one such as race. Id. at 577, 98 S.Ct. at 2949; see also Turner v. Texas Instruments, Inc., 555 F.2d 1251, 1255 (5th Cir. 1977). The employer must establish the existence of nondiscriminatory reasons for its action by a preponderance of the evidence. Id.; see also Burdine v. Texas Dept. of Community Affairs, 608 F.2d 563, 567 (5th Cir. 1979) cert. granted, ___ U.S. ___, 100 S. Ct. 3009, 65 L. Ed. 2d 1112 (1980) (legally sufficient proof is required) (emphasis in original); accord, Crawford v. Western Electric Co., Inc., 614 F.2d 1300, 1320 (5th Cir. 1980) (citing Turner); but see, Tortorici v. Harris, 610 F.2d 278, 279 at n. 1 (5th Cir. 1980) (per curiam) ("We do not decide what level of proof constitutes a sufficient rebuttal after Furnco . . ..").

[4] "Pretextual" in this context does not require that plaintiff prove that he was not chosen for promotion solely on the basis of his race without regard to skills and abilities; rather plaintiff is required to show that his race was a "but for" cause. McDonald v. Santa Fe Trail Transportation Co., 427 U.S. 273, 282 at n. 10, 96 S. Ct. 2574, 2579 at n. 10, 49 L. Ed. 2d 493 (1976); see also Albemarle Paper Co. v. Moody, 422 U.S. 405, 425, 95 S. Ct. 2362, 2375, 45 L. Ed. 2d 280 (1975).

[5] See note 3, infra.

[6] See note 4, infra.

[7] Two of the six Whites not tested before transfer from the commissary to clerical positions in other divisions of the company were employed in the clerical division of the commissary. This evidence is consistent with testimony offered by defendant's witnesses that transferees were tested only if they had not yet demonstrated their ability to perform in the area to which they sought transfer.

[8] However, evidence of conduct on which action would be time-barred is relevant, United Air Lines v. Evans, 431 U.S. at 558, 97 S.Ct. at 1889, 52 L.Ed.2d at 578, and may be utilized to examine current practices. Cf. Local Lodge No. 1424 v. N. L. R. B., 362 U.S. 411, 80 S. Ct. 822, 4 L. Ed. 2d 832 (1960); Hazelwood, supra, 433 U.S. at 309, n. 15, 97 S.Ct. at 2742, n. 15, 53 L.Ed.2d at 778.

[9] Level 1 has been abolished.

[10] Employee evaluations are reviewed by higher level management and by members of the personnel division. In addition, the results of these evaluations are discussed with the employee, who is afforded the opportunity to provide his views on his evaluation.

Factors which are evaluated in the personnel appraisal reports and performance reviews are: production; volume, quality and knowledge of work; dependability, including attendance; attitude, including willingness, cooperativeness and initiative; development; relationships with others; and supervisory ability. See Defendant's Exhibits 18, 19 and 20.

[11] The technique of regression analysis has been described as follows:

"A multiple regression analysis is a computer-assisted method of assigning numerical weights, or "influence factors", to certain independent variables (e. g., age, race, education) in relation to a single dependent variable (e. g., salary). The analysis enables an observer to determine whether a particular independent variable has influenced a dependent variable and to estimate the extent of that influence. Furthermore, the analysis enables an observer to cumulate the various influence factors and to determine the degree to which the independent variables as a unit have influenced or "explained" the dependent variable."

Agarwal v. McKee & Co., 16 EPD ¶ 8301 at p. 5573 (N.D.Calif. 1977).

In the instant case, the independent variables of job level and tenure were utilized by Dr. Chorush to predict the dependent variable of the employee's salary. See note 12, infra.

[12] Defendant's analysis thus differs from the regression analysis rejected by the Fifth Circuit in James v. Stockham Valves & Fittings Co., 559 F.2d 310 (5th Cir. 1977), cert. denied, 434 U.S. 1034, 98 S. Ct. 767, 54 L. Ed. 2d 781 (1978), inasmuch as it analyzes salary in relation to job level and tenure alone. Other factors that would influence salary-promotion rate and utilization—were examined through studies independent of this regression analysis.

[13] Employees at levels 77 and above, which include physicians, attorneys, and a few very highly placed executives, and which represent less than 2% of the work force in the aggregate, were not included in the regression analysis. These employees are outside the typical hiring, promotion and salary practices of the defendant. Moreover, since there was only one black employee in these high level positions in 1978 (and none in 1974), the inclusion of these employees as a group in the analysis would have resulted in no material change in conclusions about whether the typical black employee is appropriately paid.

[14] Thus the possibly race-related factors such as education, skill level, previous training and unchecked supervisor's evaluations that caused the Fifth Circuit to discount the regression analysis of James v. Stockham Valves, supra, are not contained in defendant's analysis.

[15] Multiple correlation coefficients may range from 0 to 1, whereby 0 indicates no prediction and 1 represents complete predictability. Thus, a multiple correlation coefficient this high is indicative of a highly structured compensation plan based on job level and tenure.

[16] This technique of separating the test group into two randomly selected groups in order to confirm the validity of the equation is known as cross-validation.

[17] Thus, defendant's expert avoided another of the pitfalls of the regression analysis enumerated by the Fifth Circuit in James v. Stockham Valves, supra. Developing an equation on the white population and then applying it to the black population precludes the possibility of developing an equation that incorporates a race-related variable to the disadvantage of Blacks. Defendant's expert's formula thereby determined what salary value defendant attached to a particular job level and tenure for white employees (a group in which racial bias cannot exist) and thereupon determined to what extent black employees in the same job level and with the same tenure obtained the same salary.

[18] In evaluating these differences in actual and predicted black salaries in order to ascertain whether they were statistically significant, defendant's expert utilized a standard "t" test, which measures the probability of differences in average values. The evaluation of significance depends upon the magnitude of the differences in comparison to the variance among the values from which the averages are derived. A measure of this variance is called the standard deviation (or in Defendant's Exhibit 71, "standard error").

Differences which are less than two standard deviations apart would occur so frequently by chance (more than one time in 20) that social scientists are unwilling to assume that they are the result of any systematic effect. This is called the .05 level of significance. When disparities are greater than two standard deviations, the probability of this occurring by chance is less than 1 in 20 (.05) and it is assumed that there is some effect causing the difference. See Castaneda v. Partida, 430 U.S. 482, 496 at n. 17, 97 S. Ct. 1272, 1281 at n. 17, 51 L. Ed. 2d 498 (1977) and Hazelwood School District v. United States, 433 U.S. 299, 308-09 at n. 14, 97 S. Ct. 2736, 2741-42 at n. 14, 53 L. Ed. 2d 768 at n. 14 (1977), in which the Supreme Court specifically approved and adopted the .05 level of significance.

[19] Defendant's Exhibit 59 indicates the following average annual salary increases by level for each year:

                   Percent     Percent         Average
                     of          of            Annual
                   Average     Average         Dollar
                   Annual      Annual         Increase-
                   Salary      Salary         Difference
                  Increase    Increase          Black
                    for         for             over
Year      Level    Blacks      Whites           White
1973        2      No active employees.
n            3       16.00       19.85           -198
            4       13.07       15.93           -176
            5       13.23       14.79            -94
            6       11.38       10.95             39
            7       11.67       11.28              2
            8       12.30        9.61            157
            9       12.33       10.08            111
           10      No active black employees.
           11       12.95        9.52            214
           12       15.13        8.64            621
         Assoc.
          Mgr.      23.42        8.39          1,898
          Mgr.     No active black employees.
Average
Increase
in 1973             14.14       11.90         257.40
1974        2      No active employees.
            3       16.95       16.76            -17            4       16.06       16.84            -50
            5       13.27       15.68           -177
            6       13.20       13.17             11
            7       13.39       12.86             62
            8       13.49       10.52            270
            9       12.80        9.45            270
           10       13.05       11.42             79
           11        9.14        9.36           -130
           12        8.40        9.91           -250
         Assoc.
          Mgr.      21.46       10.46          2,100
          Mgr.     No active black employees.
Average
Increase
in 1974             13.74       12.40         197.09
1975        2      No active employees.
            3       15.57       16.20            -58
            4       15.37       14.52             52
            5       12.98       11.82             77
            6       10.67       10.90            -10
            7        9.75       10.70            -82
            8       10.42        8.93            101
            9       11.01        9.16            142
           10       10.55        8.82            156
           11       15.26        8.35            802
           12      No active black employees.
         Assoc.
          Mgr.       2.79        1.32            241
          Mgr.     No active black employees.
Average
Increase
in 1975             11.43       10.07         142.10
1976        2       13.04        8.98            260
            3       13.76       11.72            160
            4       17.10       15.47            110
            5       14.25       13.79             32
            6       10.82       12.20            -95
            7       11.08       10.77             20
            8        9.35        9.86            -38
            9       10.51        8.56            174
           10       11.55        9.17            197
           11       11.58        8.28            377
           12       13.82        9.26            497
         Assoc.
          Mgr.      13.52       9.48             463
          Mgr.   No active black employees.
Average
Increase
in 1976             12.53       10.62         179.75
1977        2     No active employees.
            3       21.38       20.62            114
            4       20.39       19.80              5
            5       19.96       19.19             44
            6       16.31       17.42            -76
            7       15.80       15.85              6
            8       14.45       14.47             -8
            9       14.55       13.55             78
           10       14.81       15.32           -166
           11       14.27       12.62             90
           12       11.76       11.70            -86
         Assoc.
          Mgr.      13.58       11.50            176
          Mgr.      19.51       11.15          1,660
Average
Increase
in 1977             16.33       15.26         153.08
1978        2      No active employees.
            3       14.19       14.59            -66
            4       13.84       13.72             23
            5       13.77       13.65              1
            6       11.20       11.04             16
            7        9.80       10.73           -101
            8        9.29        9.41              0
            9        9.88        9.78            -16
           10        9.71       10.63           -165
           11       10.94        8.72            247
           12       10.42        8.94             98
         Assoc.
          Mgr.      11.01        8.82            279
          Mgr.      14.26        9.45            914
Average
Increase
in 1978             11.52       10.79         102.50

[20] See also Hazelwood School District v. United States, 433 U.S. 299, 97 S. Ct. 2736, 53 L. Ed. 2d 768 (1977):

"When special qualifications are required to fill particular jobs, comparisons to the general population (rather than to the smaller group of individuals who possess the necessary qualifications) may have little probative value."

433 U.S. at 308, n. 13, 97 S.Ct. at 2742 n. 13. See also Williams v. Tallahassee Motors, 607 F.2d 689, 692-93 (5th Cir. 1979); Hester v. Southern Railway Co., 497 F.2d 1374, 1379 at n. 6 (5th Cir. 1974).

[21] See also Defendant's Level System, Section II.B.5, infra, and Defendant's Promotional System, Section II.B.6, infra.

[22] Comment, Beyond the Prima Facie Case in Employment Discrimination Law: Statistical Proof and Rebuttal, 89 Harv.L.Rev. 387 (1975).

[23] Defendant's Exhibit 43 is a comparison of the 1976 promotions of Blacks to levels 10, 11, 12 and 20, with the 1975 utilization of Blacks in levels 9, 10 and 11, from which promotions to levels 10, 11, 12 and 20 would necessarily be drawn. This exhibit indicates that while in 1975 Blacks occupied 4.2% of the level 11 employees, they received 10% of the promotions to level 20 in 1976. In 1975, Blacks occupied 12.5% of the level 10 positions but received 20% of the promotions to level 12 in 1976. In 1975, Blacks occupied 11.6% of the level 9 jobs but received 13.9% of the promotions to level 11 and 22.7% of the promotions to level 10 in 1976.

[24] The availability of black college graduates is lower than their percentage representation in the population as a whole. Defendant's Exhibits 73a and 73b; Plaintiff's Exhibit 21, p. 8-9. Similarly, Blacks constitute a lower percentage of the population possessing clerical skills than they do of the population as a whole. Defendant's Exhibit 74. Defendant's expert testified that, based on such data, Blacks could be expected to be hired into job levels which would be, on the average, lower than levels into which Whites were hired.

[25] Defendant's Exhibits 62a and 62b depict the average time taken by employees to reach the various job levels, by race, for the workforce as it appeared on 12/31/77, and 12/31/78, respectively. For example, in 1977, the average black associate manager (level 20), had taken 75 months to attain that level while the average white associate manager had taken 178 months to attain level 20. These two exhibits demonstrate that for the workforce as a whole, not taking into account level of performance, training, education or seniority, black employees generally have taken a much shorter time to reach a given level than their white counterparts. However, the Court finds the same weakness in these two exhibits as in Plaintiff's Exhibits 7a-7e, i. e., there is no consideration given to the relative level of hire of the employees. Thus, the Court finds Defendant's Exhibits 62a and 62b to be of little probative value in comparing rates of promotion of black and white employees. However, inasmuch as plaintiff has failed to make a prima facie case of disparate impact with regard to promotions, see Finding of Fact No. 86, this deficiency in defendant's rebuttal exhibits is irrelevant.

[26] All of defendant's promotional exhibits include pre-Act hires.

On the other hand, plaintiff included pre-Act hires in his exhibits concerning utilization (Plaintiff's Exhibit 3; see Finding of Fact No. 91) and promotion utilization (Plaintiff's Exhibit 4; see Findings of Fact Nos. 81 and 82).

[27] The burden is on plaintiff to produce a meaningful statistical comparison in order to establish a prima facie case. Miller v. Weber, 577 F.2d 75, 77 (8th Cir. 1978), citing James v. Wallace, 533 F.2d 963, 967 (5th Cir. 1976).

[28] See note 25, infra.

[29] In rebuttal of Defendant's Exhibit 57, plaintiff prepared his Exhibit 28, which analyzes the same figures in a slightly different fashion by setting forth the percentage of the total promotions that were received by Whites and Blacks in defendant's workforce in a given year.

For example, Plaintiff's Exhibit 28 illustrates that in 1973, of the 372 promotions awarded to employees in defendant's administrative section, Whites received 306 (or 82.2%) and Blacks received 66 (or 17.7%). Similarly, Plaintiff's Exhibit 28 indicates that in 1978, Whites received 358 (or 65.1%) of the 550 promotions awarded while Blacks received 192 (or 34.9%). Yet Plaintiff's Exhibit 28 indicates that in 1973, 1978, and the intervening years, the percentage of promotions awarded to Blacks was higher than the percentage of Blacks employed in defendant's administrative section. Thus, even when the overall promotion statistics are analyzed in another manner, they provide no evidence of discrimination.

[30] Defendant's Exhibit 58 reveals the following advancement percentages by group of levels each year:

                         Percentage      Percentage
                          of white        of black
                         employees       employees
                         receiving       receiving
Year   Levels            promotions      promotions
1973    2-6                38.56           39.84
        7-12               28.06           40.00
        20, 60 & 77        12.31        Incalculable*
1974    2-6                38.85           37.96
        7-12               31.35           59.26
        20, 60 & 77        15.45            0.00
1975    2-6                40.09           44.76
        7-12               28.12           36.36

        20, 60 & 77        14.96           33.33
1976    2-6                46.95           51.98
        7-12               38.25           53.33
        20, 60 & 77        24.46           50.00
1977    2-6                41.43           43.49
        7-12               34.72           41.90
        20, 60 & 77        19.21           80.00
1978    2-6                36.32           44.41
        7-12               35.50           41.86
        20, 60 & 77        15.63           30.00

* At the beginning of 1973, there were no Blacks present within these levels, but one black employee was apparently hired into and promoted within this level group during 1973.

[31] In the approximately 3,000 instances of promotion studied by Dr. Chorush, there was only one instance in which an employee with less than six months' service received the promotion.

[32] Thus, each promotion was analyzed in terms of the available labor pool when and where the opening occurred. See Swint v. Pullman Standard, 539 F.2d 77, 105 (5th Cir. 1976).

[33] Dr. Chorush testified that any use of overall tenure would have lessened expected black promotions, because the average white employee has more than twice the tenure of the average black employee. See Defendant's Exhibit 63. In addition, black employees have less seniority at any given level because they have been promoted into their jobs more recently, on the average, than white employees. See Defendant's Exhibit 68.

[34] Defendant's Exhibit 69 reveals the following result of defendant's expert's studies:

                   Expected
         Blacks     Blacks
Year    Promoted   Promoted   Difference    t
1973       66        59.06       6.94     1.325
1974       85        76.30       8.70     1.702
1975       89        82.63       6.37     1.188
1976      144       145.73     - 1.73     -.241
1977      158       168.44     -10.44    -1.269
1978      192       202.46     -10.46    -1.311
Total     734       734.62       -.62       -

[35] Dr. Chorush's study of upper level promotions yielded the following results:

                   Expected
         Blacks     Blacks
Year    Promoted   Promoted   Difference
1973        3         1.37       1.67
1974        2         2.41       -.41
1975        6         3.83       2.17
1976       13        10.95       2.05
1977       10        11.48      -1.48
1978       14        15.95      -1.95
Total      48        45.99       2.01

[*] Level 1 was abolished in 1974.) See also Defendant's Exhibit 60.

[*] The Court did not consider Plaintiff's Exhibit 8F, concerning skilled craftsmen, or 8G, concerning semiskilled operators, since these employees would apparently be classified as wage-band employees, who have been excluded from the class.

[**] Presumably, this category includes commissary workers, who have also been excluded from the class.

[36] Defendant's Exhibit 60 reflects the following representation of black employees, by level, from 1973 to 1978.

                                                                                | % Increase
Level   |    1973   |    1974   |    1975   |    1976   |    1977   |    1978   |  1973-1979
   2    |  24% (16) |  23% (15) |  26% (22) |  35% (24) |  48% (34) |  49% (22) |  204% (133)
   3    |  16% (19) |  25% (26) |  25% (37) |  42% (56) |  48% (66) |  43% (46) |  269% (250)
   4    |  20% (30) |  21% (31) |  23% (34) |  32% (58) |  43% (75) |  45% (75) |  225% (303)
   5    |  19% (40) |  19% (36) |  23% (47) |  30% (63) |  33% (71) |  41% (67) |  216% (324)
   6    |  24% (30) |  23% (32) |  24% (37) |  25% (45) |  26% (58) |  31% (58) |  129% (260)
   7    |  12% (8)  |  21% (14) |  18% (13) |  24% (21) |  28% (27) |  27% (27) |  225% (109)
   8    |  15% (18) |  22% (24) |  25% (31) |  23% (33) |  23% (30) |  28% (29) |  187% (165)
   9    |   5% (5)  |  11% (12) |  12% (14) |  16% (20) |  17% (26) |  20% (30) |  400% (109)
  10    |   0% (0)  |   9% (2)  |  13% (4)  |  17% (6)  |  24% (12) |  23% (13) |   *
  11    |   2% (2)  |   2% (2)  |   4% (4)  |   7% (8)  |   9% (11) |  10% (12) |  500% (39)
  12    |  11% (1)  |   8% (1)  |   0% (0)  |  15% (2)  |   7% (1)  |  16% (3)  |  145% (8)
Assoc.  |           |           |           |           |           |           |
 Mgr.   |   1% (1)  |   2% (2)  |   3% (3)  |   4% (4)  |   7% (7)  |   8% (9)  |  800% (26)
 Mgr.   |   0% (0)  |   0% (0)  |   0% (0)  |   0% (0)  |   4% (2)  |   8% (4)  |   *
* Percentage increases at levels 10 and 60 could not be calculated since black representation was 0% in 1973.

[37] At trial, plaintiff introduced evidence that Blacks constitute 53.5% of the college graduates in the immediate area where defendant's Southwestern Home Office was located until 1978. See also Plaintiff's Post Trial Brief at p. 8. However, the area thus carved out by plaintiff is a small, predominately black area of the inner city, and the undisputed testimony at trial demonstrated that defendant's college hires were recruited from the entire Houston Standard Metropolitan Statistical Area and, to a lesser degree, from the Southwestern United States. Accordingly, the Court regards the figure of 53.5% black college graduates as irrelevant to any meaningful analysis.

In fact, Plaintiff's Exhibit 21 indicated a 5.7% availability for blacks with four years of college, and plaintiff's expert testified that she had no reason to disagree with this figure. See also note 24, supra. Of course, if defendant's expert had utilized the 5.7% figure from Plaintiff's Exhibit 21 rather than the 6.5% figure, the model would have predicted lower black availabilities for the hypothetical, non-discriminatory company, thus rendering defendant's action performance more favorable by comparison.

[38] Defendant's Exhibit 65 indicated the following comparisons between percentage utilization by race and model prediction:

                  1974                  1975                  1976                  1977                  1978
         |  Actual | Predicted |  Actual | Predicted |  Actual | Predicted |  Actual | Predicted |  Actual | Predicted
Level 60 |   0.00% |   0.07%   |   0.00% |   0.23%   |   0.00% |   0.44%   |   0.00% |   0.70%   |   3.70% |   1.00%
Level 20 |   1.30  |   0.71    |   2.33  |   1.29    |   3.49  |   1.90    |   4.12  |   2.55    |   6.80  |   3.21
Level 12 |  11.11  |   0.78    |   8.33  |   1.40    |   0.00  |   2.04    |  15.38  |   2.71    |   7.14  |   3.41
Level 11 |   2.47  |   4.11    |   2.35  |   5.06    |   4.17  |   5.99    |   7.14  |   6.91    |   9.48  |   7.90
Level 10 |   0.00  |   3.77    |   8.70  |   4.65    |  12.50  |   5.57    |  16.67  |   6.51    |  24.49  |   7.56
Level 9  |   5.10  |   7.28    |  10.81  |   8.37    |  11.67  |   9.46    |  16.13  |  10.89    |  17.39  |  12.38
Level 8  |  15.13  |  10.62    |  21.82  |  12.04    |  25.20  |  14.45    |  22.92  |  16.51    |  22.56  |  17.81

Level 7  |  11.76  |  14.24    |  20.59  |  17.97    |  18.06  |  19.74    |  23.86  |  20.28    |  27.96  |  20.96
Level 6  |  23.81  |  19.47    |  22.54  |  20.44    |  24.18  |  20.56    |  24.59  |  21.26    |  26.36  |  22.72
Level 5  |  19.32  |  21.70    |  18.85  |  21.24    |  22.93  |  22.59    |  30.00  |  25.07    |  32.87  |  27.25

[39] Other instances in which actual utilization fell short of model predictions but model predictions yielded less than one black employee include level 10 in 1974 (3.77% predicted × 21 = .79 Blacks); level 60 in 1975 (.23% × 34 = .0782); level 60 in 1976 (.44% × 44 = .1936); level 12 in 1976 (2.04% × 10 = .204); and level 60 in 1977 (.70% × 47 = .329).

See Roman v. ESB, 550 F.2d 1343, at 1355 (4th Cir. 1976), in which the Court of Appeals noted:

"There were no black officials and managers. 27 employees were in this category. Two percent were available in the labor force, so the number of black managerial employees should have been .54. One black employee in this category would have more than compensated for the statistical imbalance."

[40] Other instances of slight numerical disparities were found at level 9 in 1974, where the model predicted utilization of 7.13 Blacks while actual utilization was only 5; level 7 in 1974 (prediction of 9.68; utilization of 8); level 5 in 1974 (prediction of 44.19; utilization of 40); level 11 in 1975 (prediction of 4.3; utilization of 2); level 5 in 1975 (prediction of 40.57, utilization of 36); level 11 in 1976 (prediction of 5.75; utilization of 4); and level 7 in 1976 (prediction of 14.2; utilization of 13).

[41] Defendant's Exhibit 64 reveals the following comparisons of actual and predicted percentages of black promotions.

                  1974                  1975                  1976                  1977                  1978                 Total
         || Actual | Predicted || Actual | Predicted || Actual | Predicted || Actual | Predicted || Actual | Predicted || Actual | Predicted
         ------------------------------------------------------------------------------------------------------------------------------------
Level 10 || 14.28% |   3.53%   || 13.33% |   7.23%   || 22.72% |   8.23%   ||    25% |   9.24%   || 12.50% |   10.57%  || 18.18  |   8.03
Level 11 ||  0.00  |   3.53    ||  8.70  |   7.23    || 13.95  |   8.23    || 14.28  |   9.24    || 15.00  |   10.57   || 11.27  |   8.03
Level 12 ||  0.00  |   3.85    ||  0.00  |   3.77    || 20.00  |   4.65    ||  0.00  |   5.57    ||  0.00  |    6.51   ||  5.88  |   5.06
Level 20 ||  0.00  |   4.35    || 16.67  |   4.18    || 10.00  |   5.10    || 15.00  |   6.00    || 15.38  |    6.90   || 11.27  |   5.40
Level 60 ||  0.00  |   0.20    ||  0.00  |   0.89    ||  0.00  |   1.45    || 25.00  |   2.05    ||  0.00  |    2.60   ||  6.67  |   1.54

[42] Of course, this assumption is contrary to the Supreme Court's pronouncement in Hazelwood, supra, that comparisons to the workforce as as whole are of little probative value when special skills are required to fill particular jobs. 433 U.S. at 308, n. 13, 97 S.Ct. at 2742, n. 13.

[43] Plaintiff's Exhibit 26, which is a list of all employees who retired from the Southwestern Home Office after October 31, 1973, and are now receiving retirement, indicated that no black employees have retired from Prudential. The Court does not view the lack of black retirees as probative of discrimination since no Blacks were employed by defendant prior to 1965 and, therefore, no black employee would have over 14 years of tenure.

[44] Plaintiff's Exhibits 5 and 28 indicate the following:

                          Percentage of
         Terminations and
        Percentage of     Resignations
          Blacks in        Requested
         Defendant's      Received by
Year      Workforce         Blacks
1973       14.2%               0%
1974       16.2%            33.3%
1975       18.0%            12.5%
1976       20.0%            31.7%
1977       24.9%            56.4%
1978       28.9%            60.7%