Cowan v. Prudential Ins. Co. of America, 703 F. Supp. 177 (D. Conn. 1986)

703 F. Supp. 177 (1986)

Curtis COWAN, Plaintiff,
v.
The PRUDENTIAL INSURANCE COMPANY OF AMERICA, Defendant.

Civ. No. B 81-511 (CB).

United States District Court, D. Connecticut.

December 15, 1986.

Joseph D. Garrison, Garrison, Kahn, Silbert & Arterton, New Haven, Conn., for plaintiff.

*178 Wayne A. Schrader, Gibson, Dunn & Crutcher, Washington, D.C., for defendant.

 
MEMORANDUM OF DECISION

WINTER, Circuit Judge:^[*]

This is an action brought by Curtis Cowan under Title VII of the Civil Rights Act of 1964, 42 U.S.C. §§ 2000e et. seq., for discrimination in employment based on race. The parties agreed to limit the issues at trial to liability, a trial on damages to follow if necessary.

Mr. Cowan, a black man, was an insurance agent in the Stamford, Connecticut district of defendant Prudential Insurance Company of America ("Prudential") from 1975 to 1980. Claiming that he was more qualified than four white agents who were promoted to sales manager in 1978 and 1979, he argues that he was not promoted by reason of his race. Mr. Cowan also claims that Prudential's refusal to promote him, coupled with various other events, made his job situation so intolerable that he reasonably felt compelled to resign, and thus that Prudential's actions constituted a racially motivated constructive discharge.

Notwithstanding the seeming simplicity of the issues, the bench trial consumed five days of testimony, and the stack of exhibits and transcripts are well over a foot in height. The parties subsequently submitted post-trial briefs and presented oral argument. After reviewing the evidence and legal arguments, I conclude that Mr. Cowan has proven by a preponderance of the evidence that because of racial reasons, he was not considered for three promotions for which he was qualified. (One of these promotions is outside the liability period. See note 1 infra.) However, I also conclude that Mr. Cowan was not constructively discharged.

 
I. FAILURE TO PROMOTE

I first address plaintiff's claim that he was passed over for promotion to sales manager in 1978 and 1979 because of his race. Unfortunately, this requires an extended digression into the nature of Prudential's pertinent business and its various methods of evaluating an insurance agent's performance.

 
A. Nature of Business and Evaluation of Performance

As a Prudential insurance agent, Mr. Cowan worked in the Stamford District (roughly Fairfield County in Connecticut), one of approximately 70 districts that comprised Prudential's New England Region ("NEHO"). The Stamford District, which was located in Region "C" of NEHO, was under the authority of Prudential's office in Boston, Massachusetts. The Stamford District was overseen by a district manager, who reported directly to the Boston office. The various agents assigned to the Stamford District were organized into several staffs, each containing six to nine agents. Each staff had a sales manager, whose principal responsibilities were supervising and developing agents. Sales managers generally did not sell insurance or collect premiums. They reported directly to the district manager; however, there was also considerable communication between agents and the district manager.

Each agent was assigned to a staff and given a geographic territory in which he was to sell and service policies. Each agent kept a "premium collection book" ("book") that recorded the status of policies that he serviced. From the record it appears possible for an agent to have sold and serviced policies covering insureds outside the assigned geographic territory.

Prudential agents sold two kinds of life insurance. The first, "debit life policies," carried premiums due monthly, and the policyholders had the choice of either mailing their premiums to Prudential or having them collected. Prudential strongly encouraged the personal collection of premiums on debit policies because it required the agent to go into an insured's home each month, thereby enabling him both to "conserve" *179 existing business and to sell additional insurance. The second type of life insurance, "ordinary life policies," carried quarterly, semi-annual, or annual premiums that the insured sent directly to Prudential.

There were two basic objective measures of an agent's performance: the quantity of insurance sold (production) and the quality of that business (conservation). The former involved a straightforward calculation of either the total face value of policies sold or the amount of premiums owed on those policies in the first year of their term. This, of course, is a threshold requisite of good performance. Without quantity, the quality of an agent's business is of little significance.

Conservation is more complicated to explain. It was measured by a number of statistics relating to "lapse rate." Debit policies with monthly premiums were considered "in arrears" if not paid by the monthly payment due date. On such policies, Prudential allowed a 31-day grace period, honoring claims on policies that were up to 31 days in arrears. If the premium payment was not made within 31 days of its due date, however, the policy was considered to be "lapsed." Thus, an agent's "lapse rate" refers to the percentage of his business that lapsed within a defined period of time after the effective date of the policy and after payment of the initial premium.

If an agent failed to report that a policy had lapsed, the lapse was referred to as a "delayed lapse." A high arrearage rate accompanied by a low lapse rate is thus a strong indication that an agent had not reported policies that had lapsed i.e., the agent had delayed lapses. Prudential discouraged nonreported or delayed lapses because they prevented the timely reinstatement of policies and could cause payments to beneficiaries under policies that were no longer in effect.

Although the methodology described above relates only to debit policies, lapse rates were computed monthly on both debit policies and ordinary policies. For debit policies, Prudential computed a "one year" lapse rate, which represented the percentage of the agent's policies that lapsed within one year of their sale. A "two year" lapse rate was computed on ordinary life policies; a longer period was considered necessary for them because premiums were due less frequently. A weighted average of these two lapse rates yielded a "combined life new business quality" ("CNBQ") lapse rate.

Prudential also computed a figure known as the "debit life net renewal" ("DNR") lapse rate. This figure was an index of an agent's conservation experience on debit policies for which premiums had been paid for at least one year. It was computed by comparing "net renewal lapses" with the size of an agent's book. A net renewal lapse represented chargeable lapses minus "reinstatement and replacement credit" on debit policies for which at least one year of premiums had been paid. An analogous "ordinary life net renewal" ("ONR") lapse rate was also computed for ordinary life policies for which at least two years of premiums had been paid. The ONR lapse rate was expressed as a percentage, while the DNR lapse rate was expressed as a decimal, representing cents per hundred dollars of face value. Because both the CNBQ and DNR lapse rates were based (at least in part) on policies more than one year old and the ONR rate was determined from policies more than two years old, an agent had to have at least two years of experience before the full range of lapse rate figures could be computed.

Prudential also computed an arrearage rate, or percentage of an agent's policies that were in arrears. However, the arrearage rate was less significant than the lapse rate, because policies in arrears were quickly reflected in the lapse rates if premiums were not paid.

Finally, Prudential had an overall measure to rank the performance of its agents, called the "President's Trophy and Citation Program." This took into account an agent's production, commissions, and conservation in calculating "trophy points" designed to recognize and reward, in the words of Prudential's "Manager's Guide *180 for Administering its Sales Manager's Indoctrination Program," "all around performance." The trophy point system thus reflected Prudential's view of the proper relationship or "mix" between statistics relating to production and those relating to conservation. Two Prudential executives, John Collella, a Vice President of District Agencies, and Philip Cistulli, the New England Region's Director of Agencies, confirmed at trial that trophy points were an important signal of an agent's overall success. I find that trophy points were the single most important measure of performance by Prudential agents.

Agents were also often assessed on the basis of an "Agent Career Evaluation" ("ACE") form. ACE forms were completed by an agent, his sales manager, and his district manager, and rated the agent on 33 indices of performance.

 
B. Plaintiff's Prima Facie Case

In reviewing Mr. Cowan's performance, one must bear in mind the dates on which white agents were promoted. John Czel was promoted in July 1978; Steven Shepard was promoted in November 1978; Arthur Lely was promoted in August 1979; and John Cecere was promoted in October 1979.^[1]

 
1) Plaintiff's Performance as an Agent

Having surmounted a drug problem and worked his way up from a maintenance position to salesman for a corrugated box company, Mr. Cowan was hired as an agent by Prudential on February 3, 1975, and assigned to the Stamford District. For most of the pertinent time, Mr. Cowan worked out of a "detached" office in Norwalk, the main district office being in Stamford. His assigned territory was an area on the Westport-Norwalk border.

Despite a rocky start with his first sales manager not relevant to the present action, Mr. Cowan steadily improved his performance from 1975 through the end of 1978, and it remained very strong through much of 1979. In 1975, Mr. Cowan sold debit and ordinary life insurance policies with a total face value of $488,768 and premiums of $9,121. In 1976, his first full year as an agent, Mr. Cowan increased his sales to $1,007,530, with premiums of $13,702. This increase was due in part to his sales to low- and middle-income persons in the inner-city area of Stamford, where he had been raised. During Mr. Cowan's first two years, 80% of his business came from this area. As his career progressed and he became more experienced, Mr. Cowan's business came to be divided equally between the Stamford inner-city area and his assigned Norwalk-Westport area.

Mr. Cowan's sales increased again in 1977, to $1,302,688, with premiums of $19,743. His trophy point ranking among all Stamford agents increased from 13th in 1976 to 11th in 1977. (The number of agents in the district ranged between 33 and 40 from 1976 to 1979).

Although Mr. Cowan's trophy point ranking was high enough to qualify him for the honor of attending Prudential's 1977 regional business conference for the Southern New England Region, he did not receive an automatic invitation to that conference because his combined new business (CNBQ) lapse rate was slightly higher than the maximum allowed for conference qualification. Mr. Cowan's CNBQ lapse rate at the end of 1977 was 29.4%, an increase over 1976 (20.8%) and 1975 (17.2%). Of course, that increase in part reflected the inclusion for the first time in the CNBQ of the two-year ordinary life lapse rate, none of Mr. Cowan's business being two years old until 1977.

Mr. Cowan was ultimately invited to attend the conference, however. Prudential distinguished between inner-city and non-inner-city areas in computing average lapse *181 rates. An agent in a staff or district classified as "inner-city" would thus be invited to a regional business conference, despite having a higher lapse rate than other agents. This rule reflected the increased likelihood that policies sold to low-income households would lapse for reasons beyond the agent's control. Prudential did not officially define Mr. Cowan's sales group or the Stamford District as "inner-city." However, in inviting Mr. Cowan to the conference, Prudential implicitly recognized that Mr. Cowan might justifiably have higher than normal lapse rates.

By the end of 1978, Mr. Cowan's CNBQ lapse rate had decreased dramatically to 10.8%. His debit net renewal rate had decreased from 0.115 in January 1978, to 0.024 in November 1978. Meanwhile, his sales increased to $1,734,590, with premiums of $24,864. Moreover, a high percentage of these sales were debit policies, a fact that enhanced their value in Prudential's eyes. This performance unambiguously qualified Mr. Cowan for the 1978 regional business conference, and came very close to qualifying him for Prudential's "Academy of Honor." In 1978 trophy points, Mr. Cowan ranked sixth among Stamford agents and 82nd in the Southern New England Region. Mr. Cowan's strong performance continued through much of 1979. By the end of July 1979, for example, he had sold $621,012 in policies.

Other aspects of Mr. Cowan's job performance were also very positive until July 1979. He was highly motivated and frequently worked overtime. He got along very well with his sales managers, other agents, and policyholders. His fellow agents not only liked him but also had a great deal of professional respect for him. Mr. Cowan also attended various training courses during his employment at Prudential.^[2] In addition, he engaged in some cross-canvassing, which involved selling policies jointly with another agent. Finally, he recruited several new agent-candidates, although only one was hired by Mr. Amatrudo.

Mr. Cowan's performance as an agent went swiftly downhill in the second half of 1979 as a consequence of his anger at not being promoted for what he believed were racial reasons. After he learned of the promotion of Arthur Lely, Mr. Cowan virtually stopped selling policies and collecting premiums. On January 3, 1980, he submitted his letter of resignation, effective the following day.^[3]

 
*182 2) The Promotions

Mr. Cowan was, on five occasions, passed over for promotion in favor of white agents. Vincent Rella was promoted to sales manager in February 1978 (a promotion now conceded by plaintiff to have been justified on the merits), John Czel in July 1978, Steven Shepard in November 1978, Arthur Lely in August 1979, and finally John Cecere in October 1979. Agents were officially promoted by Mr. Cistulli, the NEHO Director of Agencies. However, the Stamford District Manager, who at all pertinent times was Mr. Amatrudo, had the responsibility of recommending candidates for promotion. Because Mr. Cistulli would almost never reject a district manager's recommendation, it is Mr. Amatrudo's motivation in failing to recommend Mr. Cowan for promotion that is largely at issue.

Although Mr. Amatrudo declined to promote Mr. Cowan in the Stamford District, he made efforts to get him a sales manager's position elsewhere. When Mr. Cowan made his desire for promotion known to Mr. Amatrudo in the fall of 1977, Mr. Amatrudo stated that, while there were no sales manager openings in Stamford, such a position was available in the Norwich, Connecticut district. Although Mr. Amatrudo did not have formal authority to offer Mr. Cowan the job as sales manager in Norwich, I credit Mr. Cowan's testimony that Mr. Amatrudo purported to offer him the Norwich job and urged him to consider it seriously. Mr. Cowan turned down the Norwich position, however, largely because it would have entailed a difficult move across the state, with no guarantee that his wife would be able to find suitable employment.

In early 1979, after the Czel and Shepard promotions but before those of Mr. Lely and Mr. Cecere, Mr. Cistulli suggested to Mr. Amatrudo that he inform Mr. Cowan of a sales manager opening in the Bridgeport District. Mr. Amatrudo arranged a February 1979 interview for Mr. Cowan with the Bridgeport District Manager, Al Spaziano. I credit Mr. Cowan's testimony that Mr. Spaziano offered him the job. Mr. Cowan also declined to accept this promotion. Although Bridgeport was within commuting distance of his home, the district had traditionally suffered from very poor production. For example, in 1978, Bridgeport had ranked 68th out of the 70 districts in NEHO in Prudential's President's Trophy Ranking. Because compensation for sales managers was guaranteed for only two years and was thereafter based on their staffs' production, Mr. Cowan considered the long-run prospects of a sales manager in the Bridgeport District to be bleak. He therefore continued to seek promotion in his home district, which was at that time ranked 12th in NEHO.

After the Cecere promotion, Mr. Amatrudo also helped Mr. Cowan get an offer of a sales manager position in the Boston area. This offer was also declined by Mr. Cowan.

Notwithstanding his efforts to place Mr. Cowan elsewhere,^[4] Mr. Amatrudo promoted *183 a white agent over Mr. Cowan on each of the four pertinent occasions in Stamford (excluding the Rella promotion).

 
3) Mr. Cowan's Qualifications for Promotion

I find as a fact that Mr. Cowan's qualifications for the position of sales manager were generally equal to, and probably better than, those of the agents actually promoted, with the exception of Mr. Cecere, a case discussed in detail infra.

Mr. Amatrudo's offer of the sales manager job in Norwich indicates that he considered Mr. Cowan qualified to be a sales manager as early as the end of 1977. Prudential argues that an agent might be considered ready for promotion to sales manager in a weak district without many strong candidates for promotion, such as Norwich, even if he would not be promoted ahead of various agents in his home district. Even so, the Norwich offer is some evidence of Mr. Cowan's qualification for promotion.

Other evidence demonstrates Mr. Cowan's qualification more strongly. Performance as an agent was by far the most significant factor considered by Prudential in promoting agents to sales manager. Prudential measured that performance through weekly sales bulletins, and encouraged competition for the President's Trophy and Citation Program and for invitations to the regional business conference. These honors were determined solely on the basis of sales, commissions, and lapse rates. In each of the calendar years preceding the Czel, Shepard, and Lely promotions, Mr. Cowan had more trophy points than those promoted. Thus, in 1977, he ranked higher in trophy points than Messrs. Czel and Shepard, both of whom were promoted in 1978. In 1978, he easily outranked Mr. Lely, who was promoted in 1979. Since the trophy point system is the single most important measure of performance, those rankings alone prove Mr. Cowan's qualification for promotion.

Moreover, Prudential also valued improvement. A given sales level or lapse rate was more highly regarded if part of a favorable trend. Mr. Amatrudo testified that a favorable trend is "very important" for an agent. By that measure, Mr. Cowan was performing quite well at the time of the promotions. His already high sales level was steadily increasing. His lapse rates, although relatively high because of his sales to inner-city policyholders, were steadily decreasing. In the first half of 1978, Mr. Cowan's CNBQ lapse rate ranged between 19.9% and 38.8%. These rates did not preclude Mr. Cowan's consideration for sales manager, however. Mr. Cistulli testified that no particular CNBQ lapse rate precluded promotion, and Mr. Cowan's combined lapse rates showed a strong downward trend. From a rate of 26.5% in May 1978, Mr. Cowan's CNBQ lapse rate decreased to 16.0% in July, the month in which Mr. Czel was promoted. Mr. Amatrudo himself testified that a CNBQ lapse rate of 20.0% was acceptable. By November 1978, Mr. Cowan's CNBQ lapse rate was 11.1%. Mr. Cowan's debit net renewal lapse rate also showed good progress, decreasing from a high of 0.115 in January 1978 to 0.024 in November of that year. His DNR lapse rate was only 0.019 in May and June. Based on this performance, I find Mr. Cowan to have been qualified for promotion to sales manager by June 1978, by which time he had completed the ACDP training course and had already sold $917,938 of insurance for that year. Indeed, Mr. Cowan was Prudential's "Agent of the Month" in Stamford for June 1978.

His performance must further be viewed in light of the fact that many of the top-performing agents in the Stamford District did not desire a promotion to sales manager. Because sales managers do not ordinarily sell insurance themselves, they are not paid on a strict commission basis. Rather, a sales manager is paid a guaranteed salary for the first two years based on his previous income as an agent. In later years, however, he is paid according to the production level of his staff of agents. Thus, a good agent may earn more income than a sales manager with a weak staff, no *184 matter how skilled the manager. As a result, many sales managers return to being agents after their two-year guarantee expires, and many top agents do not desire a promotion. Consequently, Mr. Cowan's competition for a sales manager's job did not necessarily come from the agents with the most spectacular performance records.

In addition, Mr. Cowan was specifically named among the "Management potential candidates" on a "Think Sheet" that Mr. Amatrudo submitted to his superior, Mr. Cistulli, on November 14, 1977. On a district personnel form completed by Mr. Amatrudo on December 4, 1978, he listed Mr. Cowan as an agent "[p]romotable to Sales Manager this year." On a Prudential affirmative action implementation plan form, prepared on February 1, 1979, Mr. Amatrudo again listed Mr. Cowan as having "potential for advancement," then circled Mr. Cowan's name to indicate that he was "currently qualified" for advancement. In order to reduce the size of this opinion, my detailed findings as to Mr. Cowan's qualifications relative to those of Messrs. Czel, Shepard, and Lely are attached as Appendix A. On the basis of those findings, I conclude that Mr. Cowan was no less qualified, and probably better qualified, than Messrs. Czel, Shepard and Lely.

I do not find that Mr. Cowan's qualifications were superior to or comparable to Mr. Cecere's at the time of the latter's promotion. After Mr. Lely's promotion to sales manager, effective August 20, 1979, but certainly made known to Mr. Cowan earlier,^[5] Mr. Cowan virtually ceased work. He testified that he decided to "take a stand" in response to the Lely promotion by "not selling any more insurance and letting the office know ... just what was going on as far as I was concerned." His sales declined precipitously and his collection of payments virtually came to a halt. In September, Mr. Cowan had an arrearage rate of 98%, and his year-to-date sales were actually lower than they had been in August, revealing that his policies were lapsing and not being replaced by new sales. Twice a week, agents were required to report new business to their sales manager and to deposit collected premiums. Mr. Cowan's virtual cessation of sales and collections, which probably occurred sometime during July, see note 5 supra, thus had to have been apparent almost immediately to his sales manager, Mr. Czel, and soon thereafter to Mr. Amatrudo. Mr. Cowan should not have been considered for the promotion that went to Mr. Cecere in October, therefore, because Mr. Cowan had effectively resigned before that opening occurred.

 
4) The Prima Facie Case

Mr. Cowan has clearly made out a prima facie case under Title VII as to the Shepard and Lely promotions, the Czel promotion falling outside the liability period. Applying the test articulated in McDonnell Douglas Corp. v. Green, 411 U.S. 792, 802, 93 S. Ct. 1817, 1824, 36 L. Ed. 2d 668 (1973), Mr. Cowan has proven: (1) that he belongs to a racial minority; (2) that he sought and was qualified for promotion to a position for which Prudential was seeking to promote a qualified employee; (3) that, despite his qualifications, he was rejected; and (4) that the promotion was offered to a non-minority employee. See also Verdell v. Wilson, 602 F. Supp. 1427, 1436 (E.D.N.Y. 1985).

 
C. Defendant's Rebuttal

In response to plaintiff's prima facie case, defendant has largely offered evidence that the differences between Mr. Cowan and Messrs. Czel, Shepard, and Lely were so insignificant that the failure to promote cannot be attributed to race. See Texas Department of Community Affairs v. Burdine, 450 U.S. 248, 255, 101 S. Ct. 1089, 1094, 67 L. Ed. 2d 207 (1981).

When the defendant in a Title VII employment discrimination action

 

fails to persuade the district court to dismiss the action for lack of a prima facie case, and responds to the plaintiff's proof by offering evidence of the reason for the plaintiff's rejection, the factfinder must then decide whether the rejection *185 was discriminatory within the meaning of Title VII. At this stage the McDonnell-Burdine presumption `drops from the case,' ... and `the factual inquiry proceeds to a new level of specificity.'

United States Postal Service Board of Governors v. Aikens, 460 U.S. 711, 714-15, 103 S. Ct. 1478, 1481, 75 L. Ed. 2d 403 (1983) (quoting Burdine, 450 U.S. at 255 & n. 10, 101 S. Ct. at 1095 & n. 10) (footnote omitted). This ultimate factual inquiry, on which the plaintiff retains the burden of persuasion, is whether Prudential treated Mr. Cowan "less favorably than others because of [his] race." International Brotherhood of Teamsters v. United States, 431 U.S. 324, 335 n. 15, 97 S. Ct. 1843, 1854 n. 15, 52 L. Ed. 2d 396 (1977).

In Burdine, the Court stated that Title VII plaintiffs may satisfy their burden of persuasion "either directly by persuading the court that a discriminatory reason more likely motivated the employer or indirectly by showing that the employer's proffered explanation is unworthy of credence." 450 U.S. at 256, 101 S. Ct. at 1095. In Aikens, the Court restated this guidance: "In short, the district court must decide which party's explanation of the employer's motivation it believes." 460 U.S. at 716, 103 S. Ct. at 1482. Thus, although it may be true that "a simple finding that the defendant did not truly rely on its proffered reason, without a further finding that the defendant relied instead on race, will not suffice to establish Title VII liability," Clark v. Huntsville City Board of Education, 717 F.2d 525, 529 (11th Cir. 1983), it is up to the trier of fact to decide whether racial motivation more likely than not "motivated the employer." It is clear that if defendant's "proffered explanation is unworthy of credence," there is strong reason to believe that it was offered as a pretext to cover a racial motivation. See Burdine, 450 U.S. at 256, 101 S. Ct. at 1095.

I find that Prudential's "proffered explanation is unworthy of credence," id., and is pretextual. There is no dispute that Mr. Amatrudo made the promotion decisions in question and that his explanations for those decisions constitute the heart of Prudential's rebuttal. I do not credit Mr. Amatrudo's explanations of why he considered Mr. Cowan less qualified than the three agents promoted over him. Mr. Amatrudo's testimony was frequently and materially inconsistent with an affidavit that he had signed in support of Prudential's earlier motion for summary judgment, and often conflicted with his deposition testimony. Moreover, his testimony at trial was often evasive or internally inconsistent.^[6]

Mr. Amatrudo's pretrial affidavit indicated that Mr. Cowan's tardiness was a significant factor in justifying his failure to promote Mr. Cowan. However, Mr. Amatrudo was willing to testify at trial only that he had "a recollection" that Mr. Cowan was late for some district meetings, which were held every six to eight weeks. Furthermore, Mr. Amatrudo, whose office was in Stamford, was in no position to observe whether Mr. Cowan was tardy on the two days each week when he was supposed to report to the Norwalk office. Mr. Cowan concedes that he did report late to the Norwalk office after the Lely promotion, *186 a concession of no relevance in light of the discussion supra, and during an earlier period when his car was not working. Mr. Cowan also acknowledges that he may have deposited premiums late on only a few occasions. This testimony was confirmed by Mr. Czel, Mr. Cowan's sales manager.

That a few instances of tardiness or late filing may have occurred in over four years of work was hardly sufficient to render Mr. Cowan unqualified for a position as sales manager. Mr. Rella was promoted despite a specific notation in Mr. Amatrudo's memorandum recommending him to Mr. Cistulli that "I also have had to speak with Vin on more than one occasion for being tardy in reporting days." No such written record of tardiness in attending meetings exists in Mr. Cowan's case. I conclude that occasional tardiness or late filing of premiums neither disqualified Mr. Cowan nor were the cause of Prudential's failure to promote him.

Mr. Amatrudo's affidavit also stated that, based on company records and his recollection, Mr. Cowan had significantly more "delayed lapses" than agents who were promoted. It thus states that Mr. Cowan had 66 delayed lapses between March 1978 and October 1979, while Mr. Czel had four in November 1977, Mr. Shepard had four in March 1978, and Messrs. Lely and Cecere had none during the pertinent times. I find, however, that the records relied upon by Mr. Amatrudo as well as his "recollection," are neither as significant nor as reliable as Prudential claims. First, roughly a third of Mr. Cowan's delayed lapses were the result of his ceasing to work after he learned of the Lely promotion. Second, the records relied upon in the affidavit disclose only those delayed lapses discovered after a sales manager had decided to check an agent's book. If a particular agent's book was checked more frequently than the books of others, that agent would almost surely have a greater number of delayed lapses attributed to him. Mr. Cowan testified that his book was checked more often, and I credit that testimony. Third, the number of delayed lapses that an agent has may exaggerate the underlying problem. If a family with four policies was frequently not at home when the agent came to collect premiums, the agent would be charged with four lapses although a single successful call would lead to reinstatement.

Fourth, and most important, we know that the records relied upon by Mr. Amatrudo in the affidavit are not accurate. Other records produced at trial show that Messrs. Lely and Cecere each had at least two and as many as four critically delayed lapseslapses unreported for several months beyond the grace periodbetween December 1978 and September 1979.^[7] Because a critically delayed lapse can occur only after a delayed lapse, the statements in Mr. Amatrudo's affidavit indicating that Messrs. Lely and Cecere had no delayed lapses are clearly wrong. This fact lends credence to Mr. Cowan's claim that his book was checked more often than those of other agents. Mr. Amatrudo's explanation for the error was that the records of critically delayed lapses were not in the personnel file when he checked that file in order to prepare his affidavit. This of course indicates that Mr. Amatrudo's affidavit was not a straightforward attempt to explain his thinking at the time in not promoting Mr. Cowan, but was a post-hoc rationalization based on whatever negative evidence was found.^[8]

Finally, Mr. Czel, Mr. Cowan's sales manager during the pertinent period, testified *187 that delayed lapses were commonplace and probably experienced by every agent instead of relatively rare as implied by Mr. Amatrudo. I therefore conclude that the existence of delayed lapses was not a cause for Prudential's failure to promote Mr. Cowan but was a pretext developed for defending this litigation.

Prudential next argues that even if Mr. Cowan was as qualified as those promoted, he still bears the burden of proving that he was better qualified, see e.g., Young v. Lehman, 748 F.2d 194, 198 (4th Cir.1984), cert. denied, 471 U.S. 1061, 105 S. Ct. 2126, 85 L. Ed. 2d 489 (1985); Aikens v. United States Postal Service Board of Governors, 665 F.2d 1057, 1060 (D.C.Cir. 1981), vacated on other grounds, 460 U.S. 711, 103 S. Ct. 1478, 75 L. Ed. 2d 403 (1983). Prudential contends that absent such a showing, it had complete discretion "to choose among equally qualified candidates, provided the decision is not based upon unlawful criteria." Burdine, 450 U.S. at 259, 101 S. Ct. at 1096; see also Loeb v. Textron, Inc., 600 F.2d 1003, 1019 (1st Cir. 1979) (in age discrimination case, employer "is entitled to make its own subjective business judgments, however misguided they may appear" when management-level job is involved); E.E.O.C. v. Trans World Airlines, Inc., 544 F. Supp. 1187, 1220 (S.D.N. Y.1982) ("employer may choose from among several qualified candidates without violating [the equal employment laws]").

Prudential's argument is misplaced in this case. Whether or not Mr. Cowan was unambiguously more qualified than Messrs. Czel, Shepard or Lely is not the issue in light of Mr. Amatrudo's trial testimony. He testified with great emphasis that he did not even consider Mr. Cowan for the sales manager openings filled by those agents, but considered him only for the promotion that went to Mr. Cecere. If Mr. Cowan was not even considered for a promotion, Prudential cannot rely on the claim that he was at best equally qualified to avoid a finding of racial discrimination. Even if each promotion was a "horse race," Mr. Cowan should have been allowed to run.

At oral argument, counsel for Prudential sought to pass off Mr. Amatrudo's testimony as a slip, a lapse of memory, and a remark not intended to be taken literally. This is clearly not the case. I am thoroughly satisfied that Mr. Amatrudo did not consider Mr. Cowan on his merits for the three promotions mentioned. Mr. Amatrudo repeatedly testified that Mr. Cowan was not "a serious candidate for the positions of Rella, Czel, Shepard, or Lely." When asked specifically whether Mr. Cowan's qualifications were "so substantially worse that you did not consider him despite his expressed desires to be promoted" on those occasions, Mr. Amatrudo responded, "That's correct, for one reason orno, he was not considered." In view of the importance of Mr. Amatrudo's testimony on this issue, I sought to clarify his use of the term "considered." I thus asked him, "When you say you didn't consider Mr. Cowan, that means that he was not a candidate, in your view?" Mr. Amatrudo answered, "Yes, your honor." The witness' statements as well as his demeanor thus conveyed the distinct message that Mr. Cowan was not in the running or, switching metaphors, not even in the same league with the agents actually promoted.

That Mr. Amatrudo never gave the possibility of promoting Mr. Cowan any serious thought is reinforced by the strenuous efforts undertaken by Mr. Amatrudo to convince Mr. Cistulli to promote Mr. Lely. Mr. Cistulli, who had rejected Mr. Amatrudo's suggestion in December 1977 that Mr. Lely be considered for sales manager and had specifically suggested to Mr. Amatrudo in early 1979 that Mr. Cowan interview for the Bridgeport position, disagreed with Mr. Amatrudo's recommendation of Mr. Lely in 1979. Mr. Cistulli's unusual resistance to a district manager's recommendation stemmed from the fact that Mr. Lely had performed very poorly as a sales manager several years earlier. See Appendix A. However, Mr. Amatrudo testified that "I continued at it until I got Mr. Cistulli to agree." Thus, with respect to at least one of the candidates promoted ahead of Mr. Cowan, and after Mr. Cistulli had mentioned Mr. Cowan as a management candidate, *188 Mr. Amatrudo found it necessary to argue strenously with Mr. Cistulli over his recommendation of Mr. Lely.

Because the evidence establishes that Mr. Cowan had overall qualifications at least equal to those of the agents promoted, see Appendix A, he should have been considered for the promotions received by those agents. Prudential itself devoted virtually its entire post-trial brief on the promotion issue to establishing the relatively equivalent qualifications of Mr. Cowan and the others and offered no substantial reason whatsoever why he was not even considered. Under these circumstances, the inference that racial motives were involved in the failure to consider Mr. Cowan for promotion is inexorable.

Other evidence confirms that Mr. Cowan's race was a factor motivating Mr. Amatrudo's sales manager recommendations. During 1978, Mr. Amatrudo repeatedly informed his superiors, particularly Mr. Cistulli, that he considered Mr. Cowan to be promotion material. Moreover, Mr. Amatrudo was doing his best to get Mr. Cowan a job in another district. It was unusual for an employee to be promoted from an agent position in one district to a sales manager position in another, however. Less than 10% of Prudential's promotions from agent to sales manager were of this kind. Indeed, in view of Prudential's stated policy of keeping moving expenses to a minimum, district managers were explicitly advised to "`[i]mport' a candidate [for sales manager] when moving expenses are involved only after careful consideration of all candidates." Consequently, Mr. Amatrudo's actions are consistent with a desire to avoid promoting Mr. Cowan in Stamford because of his race and to peddle him elsewhere within Prudential.

Furthermore, Mr. Amatrudo's testimony that he did consider Mr. Cowan for the Cecere promotion demonstrates either how oblivious he was to Mr. Cowan's merits or how pretextual are his efforts to explain the failure to promote Mr. Cowan. The Cecere promotion was formally made on October 5, 1979, indicating that Mr. Amatrudo must have made his recommendation to Mr. Cistulli sometime in September 1979. Mr. Cecere was a relatively new agent, having been hired around the time that Mr. Amatrudo had offered the Norwich job to Mr. Cowan. Mr. Cecere's performance need not be detailed here because of the overriding reasons for Mr. Amatrudo's not seriously considering Mr. Cowan for promotion in September or October. After Mr. Lely's promotion to sales manager, effective August 20, 1979, but surely made known to Mr. Cowan earlier, see note 5 supra, Mr. Cowan virtually ceased work. His sales were thus precipitously declining, and his collection of payments had virtually come to a halt well before the Cecere promotion. By September, Mr. Cowan had an arrearage rate of 98%, and his year-to-date sales were actually lower than they had been in August, revealing that his policies were lapsing and not being replaced by new sales.

Prudential argues that the September 1979 arrearage figures might not have been known to Mr. Amatrudo at the time that the Cecere decision was made. However, it is totally implausible that the precipitous decline of revenues from Mr. Cowan's efforts would not have been known by both his sales manager and district manager by the middle of September 1979. Every Tuesday and Friday, agents were required to report new business to their sales managers and to deposit premiums that they had collected. Mr. Cowan's virtual cessation of sales and collections after Mr. Lely's promotion would have been immediately apparent to his sales manager, Mr. Czel, and would inevitably have come to the attention of Mr. Amatrudo soon thereafter.

Had Mr. Cowan been treated solely on the merits concerning promotion to sales manager, therefore, he would have been considered for the promotion that went to Messrs. Czel, Shepard, and Lely. He would not have been considered for the promotion that went to Mr. Cecere, however, because Mr. Cowan had effectively resigned when that opening occurred. Mr. Amatrudo's testimony that he did consider Mr. Cowan for the promotion that went to Mr. Cecere, but not for the others, demonstrates both that Mr. Amatrudo was indifferent *189 to the merits of promoting Mr. Cowan and that his after-the-fact effort to fashion a non-discriminatory motive is not credible.

The court finds that Mr. Cowan was in fact not considered by Mr. Amatrudo for the Czel, Shepard, and Lely promotions. The record discloses no explanation for this behavior other than racial considerations.^[9] The consequences of this finding will be determined in the damages phase of this litigation.

 
II. CONSTRUCTIVE DISCHARGE

Plaintiff argues that Mr. Amatrudo's failure to consider him for promotion, coupled with a number of other allegedly racially motivated incidents, constituted a "constructive discharge" in January 1980. The law recognizes that an employee's working conditions may be made so intolerable by racial prejudice that he may reasonably feel compelled to resign. In such situations, the employee's resignation is considered involuntary and is deemed a constructive discharge in violation of Title VII.

In this case, Mr. Cowan submitted his letter of resignation on January 3, 1980,^[10] effective January 4, 1980. He argues that he reasonably felt compelled to resign as a result of the following circumstances: (1) the failure to promote him;^[11] (2) various provocative remarks by Mr. Amatrudo; (3) an altercation between himself and Mr. Lely; and (4) the failure of Prudential's high-level management to address his complaints.

Martin v. Citibank, N.A., 762 F.2d 212 (2d Cir.1985), states the applicable legal standard governing a claim of constructive discharge in this circuit:

 

A finding of constructive discharge in violation of § 1981 or Title VII requires that the trier of fact "`be satisfied that the ... working conditions would have been so difficult or unpleasant that a reasonable person in the employee's shoes would have felt compelled to resign.'" Pena v. Brattleboro Retreat, 702 F.2d 322, 325 (2d Cir.1983) (quoting Alicea Rosado v. Garcia Santiago, 562 F.2d 114, 119 (1st Cir.1977)). The employer must have "`deliberately [made] an employee's working conditions so intolerable'" as to force the resignation. Id. (quoting Young v. Southwestern Savings and Loan Ass'n, 509 F.2d 140, 144 (5th Cir.1975)).

762 F.2d at 221.

Prudential argues that this standard requires at least that "the causative conduct complained of must be shown to be intended, be shown to be intended to be adverse, and be shown to be intended or engaged in for racial reasons." Defendant's Post-Trial Brief at 81. Plaintiff argues that there is no need to prove that an employer specifically intended to force the employee out of the job. Rather, he argues, a constructive discharge occurs whenever there is "such a change in [an employee's] working conditions that working conditions are so difficult or unpleasant that a reasonable person in his shoes would resign." Plaintiff's Post-Trial Brief at 88. This dispute need not be resolved, however, because Mr. Cowan has proved neither an intent to *190 force him to resign nor intolerable working conditions.

First, I cannot conclude that anyone at Prudential wanted to force Mr. Cowan to quit. There is no evidence at all that either Mr. Cistulli or Mr. Collella knew that Mr. Amatrudo was not even considering Mr. Cowan for promotion. Nor do I find that Mr. Amatrudo wanted Mr. Cowan simply to leave Prudential. Were that to happen and were Mr. Cowan to sell a competitor's insurance to his established clients, Prudential's Stamford District, already in decline, would lose significant business, a development that Mr. Amatrudo would have trouble explaining to his superiors. Rather, Mr. Amatrudo was content to have Mr. Cowan as an agent^[12] or to get him another job in Prudential so that his book would remain with the company.

Nor were Mr. Cowan's employment conditions so intolerable that a reasonable person would have felt compelled to quit. The basis for my discrimination finding is Mr. Amatrudo's testimony that he did not consider Mr. Cowan for promotion. At the time of the relevant promotions, however, Mr. Amatrudo was informing Mr. Cowan otherwise; although Mr. Cowan may have had grounds to believe he should have been promoted, he did not have the evidence that became available at trial, long after he had resigned. The evidence available to Mr. Cowan at the time demonstrated at best that even if he was slightly better qualified than those promoted, the competition for each promotion was very close. The failure to win these close promotions is understandably disappointing. However, it is not intolerable.

Moreover, Mr. Cowan might well have received the next promotion after Mr. Cecere. There is some evidence that Mr. Amatrudo had virtually run out of suitable agents in mid-1980. The 1980 promotion went to Peter Boucher, who had been an agent only since August 1979, and obviously had not yet worked long enough for his CNBQ and DNR lapse rates to provide accurate information. Because it was Prudential's explicit written policy, confirmed by the testimony of Mr. Cistulli, not to promote agents with "short service" when more experienced agents were available, Mr. Amatrudo might have been forced to acknowledge that Mr. Cowan was clearly superior to Mr. Boucher had Mr. Cowan continued servicing and selling policies for Prudential. At the time of this promotion, moreover, Mr. Amatrudo might well have been under heavy pressure from his superiors to select a proven agent like Mr. Cowan in light of the poor performance of the Stamford District in 1979, when it slipped from 12th in NEHO to 33rd. These conclusions, of course, are speculative because the strength of Mr. Amatrudo's feelings about Mr. Cowan cannot be measured on this record.

Mr. Cowan also relies upon various comments made by Mr. Amatrudo to support the constructive discharge claim. In January 1979, Mr. Amatrudo apparently called Mr. Cowan into his office and told him that neither Mr. Amatrudo nor the older agents in Stamford respected Mr. Cowan, and that Mr. Cowan "was a terrible influence on the younger agents." This incident lends little support to the constructive discharge claim. In Martin v. Citibank, N.A., the Court of Appeals held that an oral warning by an employee's supervisor regarding unfounded complaints from customers and co-workers and the supervisor's "loud mention" in public that the employee had been polygraphed "do not legally suffice to sustain an inference that a reasonable person would have been `compelled' to resign." 762 F.2d at 221. Mr. Amatrudo's comments are much less compelling than those in Martin so far as a constructive discharge is concerned.

Mr. Cowan also complains about a threat made by Mr. Amatrudo on October 24, 1979. On that date, Mr. Amatrudo called Mr. Cowan into his office to discuss Mr. Cowan's low production and suggested that Mr. Cowan perhaps should resign from *191 Prudential if he was not willing to go back to work. Mr. Cowan apparently told Mr. Amatrudo that he would go back to work "when [his] situation was rectified." Mr. Amatrudo told Mr. Cowan that he was going to "take [him] out in the parking lot and kick [his] ass." Although I am willing to credit Mr. Cowan's account of this conversation, I find no reason to believe that this comment was either a serious physical threat or was racially motivated. Mr. Cowan had in fact stopped performing his duties, and Mr. Amatrudo was simply upset that one of his top agents was no longer producing.

Mr. Cowan also points to an incident that occurred between him and Mr. Lely in late November 1979. According to Mr. Cowan's testimony, he was called into Mr. Czel's office, where a rather animated discussion with Mr. Czel took place. Mr. Lely, who was acting district manager on that occasion, entered Mr. Czel's office and, as Mr. Cowan testified, "just started ranting and raving, yelling at the top of his lungs at me." Mr. Cowan decided to leave, and, as he "tried to leave [Mr. Czel's] office, [Mr. Lely] blocked the doorway and he took his hands and pushed me in the chest, completely across his office, into the corner."

The circumstances surrounding this incident clearly indicate that, however unwarranted Mr. Lely's behavior may have been, it was neither motivated by racial considerations nor a cause to resign. Mr. Cowan had filed charges with the Connecticut Human Rights Commission in which he alleged that white agents had received promotions that he deserved. These charges were printed in a newspaper article that mentioned Mr. Lely and Mr. Czel by name. The evidence is overwhelming that Mr. Czel and Mr. Lely were reacting to being described publicly by Mr. Cowan as not deserving promotion, not to Mr. Cowan's race. Mr. Cowan himself testified that Mr. Lely's reaction was prompted by the newspaper story, not by race. It is not even clear that these highly personal reactions can be attributed to Prudential.

Another of Mr. Cowan's complaints stems indirectly from the Lely incident. On the day following the alleged assault by Mr. Lely, Mr. Cowan drove to Prudential's corporate headquarters in Newark, New Jersey. A vice president, Joe Malone, met with Mr. Cowan. Mr. Cowan told Mr. Malone of the Lely incident and described his career leading up to that incident. According to Mr. Cowan's testimony at trial, Mr. Malone was very polite and listened with a "responsive ear." Mr. Malone then told Mr. Cowan that he "wouldn't arbitrarily intervene but once he reviewed all of the facts, he would get back to me." Mr. Malone had not gotten back to Mr. Cowan by the time of his letter of resignation. Mr. Cowan claims that Mr. Malone's failure to respond to his charges reinforced his feelings of racial rejection and isolation.

Mr. Cowan further argues that a meeting with Mr. Cistulli in Boston in December 1979, at which Mr. Cistulli asked Mr. Cowan to sign certain documents, also reinforced those feelings. Mr. Cowan had expected Mr. Cistulli to rectify the wrongs that Mr. Cowan felt had been committed. Similarly, Mr. Cowan points out that Mr. Collella, with whom Mr. Cowan had spoken in October regarding his perceived racial discrimination, never responded to Mr. Cowan's concerns.^[13]

I am unable to attribute this conduct to racial prejudice or to conclude that it made Mr. Cowan's working conditions intolerable. There was no reason for Prudential executives, or even Mr. Amatrudo's direct superior, Mr. Cistulli, to believe that Mr. Cowan had not even been considered for several promotions. There was also no reason for these individuals to believe that Mr. Cowan was so unambiguously the best choice for those promotion, that they *192 should intervene in a decisions for which district managers were as a matter of practice responsible. What they must have known was that Mr. Cowan had stopped work. It was thus entirely reasonable for them to tell Mr. Cowan "to go back and turn it around."

Moreover, although Mr. Cowan may have sincerely felt that his working conditions were intolerable, he was not free to quit working and then complain that Prudential executives were unhappy with his work and unreceptive to his grievances. As the cases almost uniformly recognize, "the policies underlying Title VII will be best served if, whenever possible, unlawful discrimination is attacked within the context of existing employment relationships." Bourque v. Powell Electrical Manufacturing Co., 617 F.2d 61, 66 (5th Cir.1980).

None of the matters described above, whether viewed singly or as a whole, constituted a constructive discharge. I thus hold that Mr. Cowan's resignation was voluntary, whether the date of that resignation was July 1979, when he stopped working; December 15, 1979, when he took a job with another insurance company; or January 1980, when he sent his letter of resignation to Prudential.

 
III. CONCLUSION

Plaintiff has proven by a preponderance of the evidence that Prudential violated Title VII by failing to consider him for promotion because of his race. However, plaintiff has not satisfied his burden of showing that he was "constructively discharged" by Prudential. This opinion should serve as my findings of fact and conclusions of law under Fed.R.Civ.P. 52.

SO ORDERED:

 
APPENDIX A 
Comparison of Curtis Cowan with John Czel, Steven Shepard, and Arthur Lely 
(1) John Czel

By the time of Mr. Czel's promotion in July 1978, Mr. Cowan had established qualifications equivalent to, or better than, those of Mr. Czel. Mr. Czel sold insurance valued at only $651,759 in 1977, compared to Mr. Cowan's total of $1,302,688. In the first six months of 1978, Mr. Czel sold $529,958 worth of insurance compared to Mr. Cowan's sales of $917,933. Mr. Cowan was clearly superior to Mr. Czel in the key overall measure of objective performance, Prudential's Trophy Point System. In 1977, Mr. Cowan had 2,289.8 trophy points; in the same year, Mr. Czel had only 1,185.2. These points ranked Mr. Czel 22nd in the Southern New England region in 1977, compared to Mr. Cowan's rank of 11th. Mr. Czel did not qualify for the regional business conference.

Mr. Czel fares somewhat better in a comparison of lapse rates. Mr. Cowan had significantly higher combined (CNBQ) lapse rates than did Mr. Czel during most of 1977. For six months in 1977, Mr. Czel had CNBQ lapse rates under 5.0%, but those rates began to increase substantially after November 1977. Mr. Czel's lapse rate increased from 5.8% in November 1977 to 37.3% by June 1978. During the same period, Mr. Cowan's rate decreased from 30.9% in November 1977 to 19.9% in June 1978. Again, a favorable trend, in addition to acceptable performance figures, was important.

Mr. Cowan and Mr. Czel had roughly comparable debit net renewal lapse rates and arrearage rates during the months preceeding Mr. Czel's promotion. In 1977, Mr. Cowan's DNR lapse rate ranged between 0.066 and 0.142, while Mr. Czel's rate ranged between 0.069 and 0.110. As of April 1978, Mr. Cowan had decreased his DNR lapse rate to 0.063. It fell again in May to 0.019, and remained at 0.019 in June. Mr. Czel had a somewhat better DNR lapse rate of 0.034 in April 1978, but his rate remained at 0.034 in May, and increased to 0.039 in June. Thus, although Mr. Cowan had DNR lapse rates that were consistently higher than those of Mr. Czel from April 1977 to April 1978, Mr. Cowan's trend in early 1978 and the rate he had reached by May and June 1978 establish that he was at least roughly comparable to Mr. Czel with respect to this index of performance. *193 Similarly, Mr. Czel and Mr. Cowan were virtually neck and neck with respect to arrearage rates in the first half of 1978. Mr. Cowan's monthly arrearage rate decreased from 53% to 28% between January and June 1978, with a low of 18% in May; Mr. Czel's rate was relatively stable, ranging between 30% and 35% over those same months.

Other aspects of Mr. Czel's and Mr. Cowan's qualifications for promotion demonstrate the lack of justification for Mr. Amatrudo's failure to consider Mr. Cowan. Mr. Czel did not get along with his sales manager, Clem Pepe. This fact was reflected in a letter Mr. Amatrudo wrote to Mr. Santello, Region C supervisor. In July 1978, Mr. Pepe rated Mr. Czel substantially lower than did Mr. Amatrudo on Mr. Czel's ACE form. That form was separately completed by the agent, his sales manager, and his district manager. Even Mr. Amatrudo described Mr. Czel as having had a "problem with attitude" in 1977.

Mr. Czel did have more formal training than Mr. Cowan. Mr. Czel had completed both LUTC I and II and had been licensed by the National Association of Securities Dealers. He had also been elected as the local union chairman, a position that supposedly reflected his ability to get along with and be respected by his fellow agents.

For his part, however, Mr. Cowan also had numerous assets beyond the statistical measures of his performance. He got along well with his fellow agents and policyholders. Mr. Cowan was also recognized as a hard worker who was highly motivated to succeed. Moreover, by June 1978, Mr. Cowan had completed ACDP and had just been named "Agent of the Month."

 
(2) Steven Shepard

By the time Steven Shepard was recommended for promotion on October 4, 1978, Mr. Cowan had done nothing but improve his already strong performance. Through September 1978, Mr. Cowan had sold $1,150,991 of insurance, and his CNBQ lapse rate was down to 11.8%. However, his DNR lapse rate had increased somewhat to 0.041. Mr. Shepard's performance was also quite good. Mr. Shepard began his career in late 1975 as an agent in Rochester, New York, servicing an inner-city area. In 1976 he had sales of $1,358,413, with a year-end CNBQ lapse rate of 2.7%. He moved to the Stamford District in January 1977, and by the end of that year had sold insurance valued at $1,330,999, compared to Mr. Cowan's $1,302,688. Mr. Shepard finished the year with a December CNBQ lapse rate of 19.1% and a DNR lapse rate of 0.032. Mr. Shepard and Mr. Cowan finished 1977 very close in President's Trophy Points as well; Mr. Cowan had 2,289.7 points, while Mr. Shepard had 2,156.5.

Mr. Shepard also performed well in 1978. By the end of September 1978, Mr. Shepard had sold $1,191,820 in insurance, slightly more than Mr. Cowan's $1,150,991. Mr. Shepard also had improved his CNBQ lapse rate to 6.2%, and had a DNR lapse rate of 0.021. Mr. Shepard had arrearage rates that were somewhat lower than Mr. Cowan's during the first nine months of 1978. Between January and September 1978, Mr. Shepard's monthly arrearage rate fluctuated between 33% and 20%, while Mr. Cowan's rate was as high as 53% and as low as 18%. However, Mr. Shepard's sales in 1978 were relatively low until July of that year. Through June 1978, Mr. Shepard had sold 42 new policies valued at $696,530. Mr. Cowan, by comparison, had sold 49 policies valued at $917,938. At about this time, Mr. Amatrudo transferred Mr. Shepard to the staff of sales manager Sal Vitanza, who had the highest producing staff in Stamford. Mr. Shepard's performance while working on a book from a former member of Mr. Vitanza's staff may have enabled him to surpass Mr. Cowan in sales by the time of his recommendation for promotion in October. In any event, being on the Vitanza staff certainly enhanced Mr. Shepard's 1978 performance.

Mr. Shepard's intelligence was also considered a positive asset by Mr. Amatrudo. Objective test evidence in part confirms this conclusion. Although Mr. Shepard scored below average in the ACDP tests for "Agency Rate Book" and "Life Insurance Knowledge," he scored above average *194 on the "General Intelligence Test," with a score of 31. Mr. Cowan, by comparison, scored 27 on the latter test, also above average, and, like Mr. Shepard, scored below average on the former tests.

Despite Mr. Shepard's positive assets, his qualifications for promotion were not clearly superior to those of Mr. Cowan. Mr. Cowan's performance was also strong, and he was never thought to lack adequate intelligence. Moreover, Mr. Shepard and Mr. Cowan were in the same LUTC II course when Mr. Shepard was promoted. In addition, during the several months prior to Mr. Shepard's promotion, his sales manager, Mr. Vitanza, evaluated him as needing improvement in many areas of obvious importance to sales managers.

Mr. Shepard's overriding flaw, however, which was fully apparent at the time of his promotion, was his personality. Several employees who worked with Mr. Shepard in 1978 testified that he did not relate well to fellow agents or other employees. Mr. Czel described Mr. Shepard as "brash," and agreed that he "alienated some people." Mr. Giordano, an agent on the same staff as Mr. Shepard, testified that Mr. Shepard acted superior to everyone else, was "uppity," and "never related well to anybody" in the office. Even more damning, Mr. Rella testified that Mr. Shepard was "disliked" by other agents, who viewed him as "very cocky" and lacking in "respect for anybody, for seniority, for anything." Mr. Rella stated that to his knowledge all agents "definitely" did not want Mr. Shepard as "their leader or sales manager." When asked whether he had ever managed Mr. Shepard, Mr. Rella responded, "No, thank God." Mr. Shepard was not successful as a sales manager. Less than a year after his promotion, he returned to being an agent. Two months later, he left Prudential entirely. In November 1979, Mr. Amatrudo wrote, in a report concerning a termination interview, that Mr. Shepard had a "superior attitudewould not take directionnot liked by associateswould not re-hire."

 
(3) Arthur Lely

In the eight months between Mr. Shepard's promotion and Mr. Amatrudo's recommendation of Mr. Lely for promotion on July 25, 1979, Mr. Cowan's strong performance continued. Through June, Mr. Cowan had sold 33.5 policies valued at $338,387. In June, his CNBQ lapse rate was 13.7% and his debit net renewal lapse rate was 0.076. His sales performance, although not as strong as his sales ($917,938) during the first six months of 1978, was nevertheless rather good. By the end of July, he had sold $621,012 of insurance. By comparison, Mr. Lely had sold 40 policies valued at $682,031 through June 1979 ($797,326 through July). In June Mr. Lely's CNBQ lapse rate was 11.4% and his DNR lapse rate was 0.046. In the first six months of 1979, Mr. Lely also had better monthly arrearage rates than Mr. Cowan. Mr. Lely's rates ranged between 18% and 35%, while Mr. Cowan's rates ranged between 21% and 61%, with three months above 50%.

In 1978, on the other hand, Mr. Cowan's performance had been clearly superior to Mr. Lely's. In that year, Mr. Lely sold 72 policies worth $1,364,879 and finished the year with a December CNBQ lapse rate of 15.0% and DNR lapse rate of 0.060. Mr. Cowan sold 111 policies worth $1,734,590 and finished the year with a December CNBQ lapse rate of 12.6% and a DNR lapse rate of 0.038. In the crucial President's Trophy Points category, Mr. Cowan outperformed Mr. Lely with 2,977.8 points in 1978, ranking him 82nd in Region C of NEHO and sixth in Stamford, while Mr. Lely had only 2,489.8 points and ranked only 166th in Region C and ninth in Stamford. Mr. Cowan and Mr. Lely had roughly equivalent arrearage rates in 1978.

Mr. Lely had been a sales manager before 1976, and had many years of experience as an agent. As a result, Mr. Amatrudo claimed to have viewed him as having a good knowledge of the "basics," and as being both a "strong influence on the newer agents" and "of great assistance to the young Sales Managers."

*195 However, there were several serious drawbacks to Mr. Lely as a candidate for promotion. It was Mr. Lely's intention to leave the Stamford area soon to be near his daughter in Florida. Mr. Amatrudo denies having known this fact at the time of Mr. Lely's promotion. However, it was a fact known at the Stamford office at the time, and I conclude that it was certainly known to Mr. Amatrudo, who was a close friend of Mr. Lely's.

Moreover, Mr. Lely's prior performance as a sales manager had been very poor, and he had even been placed on probation as a result of poor sales during his tenure as sales manager. Mr. Amatrudo claims to have made no inquiry into Mr. Lely's earlier experience in a managerial position, testifying at trial that "[i]t didn't interest me." I find that Mr. Amatrudo was generally aware of Mr. Lely's poor performance record as sales manager by the time he recommended him again on July 25, 1979. As early as December 1977, Mr. Amatrudo had raised the issue of Mr. Lely's promotion with Mr. Cistulli and had been rejected. The reason for this rejection must have been Mr. Lely's prior record. Indeed, Mr. Cistulli testified that such a record would warrant special caution in making a re-promotion. When Mr. Amatrudo wrote Mr. Cistulli again on July 25, 1979, he stated that "I understand the short comings [sic] of Art Lely," adding, "Phil, I know and understand [your] feelings on this." Thus, I find that Mr. Amatrudo must have been at least generally aware of Mr. Lely's prior record as sales manager, at the very least through his discussions with Mr. Cistulli concerning the matter.

Mr. Amatrudo's lack of concern about Mr. Lely's past experience also seems at odds with his emphasis on production levels during 1979 because of the Stamford District's declining rank in the President's Trophy Points System. During 1979 the Stamford District was declining from 12th in NEHO in 1978 to 33rd in 1979, a development of which Mr. Amatrudo could not have been wholly ignorant by July 1979. In light of this development, prior performance certainly should have been considered an important factor.

Further, Mr. Lely's supposed strengths in "the basics" and "influence on newer agents" could not have been as significant as Mr. Amatrudo claimed. More than half of the agents on the staff assigned to Mr. Lely either had been sales managers themselves, had been listed by Mr. Amatrudo as "promotable candidates," or were conceded by Mr. Amatrudo to be "veteran agents" who did not require "the basics."

It is also worthy of note that Mr. Lely's performance as sales manager in 1979 and 1980 was consistent with his past performance. When one of Mr. Lely's "newer" agents, Peter Boucher, was being considered for a promotion in early 1980, less than a year after Mr. Lely's promotion, Mr. Amatrudo excused Mr. Boucher's sales levels on the ground that his sales manager, Mr. Lely, was not sufficiently sales oriented. Mr. Amatrudo wrote, "I do feel that if he was with a more aggressive [s]ales oriented Sales Manager that we would have seen a much better production record." Mr. Amatrudo agreed at trial that this remark was critical of Mr. Lely.

In addition, Mr. Lely had apparently never completed the Agent's Career Development Program ("ACDP"), although the program was considered quite important for sales manager candidates. Mr. Amatrudo also deemed it unnecessary to prepare an ACE form on Mr. Lely. Rather, he assumed that one "probably was done when he was promoted" to sales manager for the first time. Mr. Amatrudo also testified that no ACE form was prepared because Mr. Cistulli had not asked for one. Whatever the reason, Mr. Amatrudo did not have the benefit of the systematic agent's evaluation form when making his sales manager selection.

In November 1978, Mr. Amatrudo and Mr. Czel, Mr. Cowan's sales manager, completed an ACE form on which both men gave Mr. Cowan high marks. Notably, Mr. Czel, who had much greater opportunity to observe Mr. Cowan at the Norwalk detached office, rated Mr. Cowan higher than did Mr. Amatrudo in approximately two-thirds of the categories and at least as high *196 as did Mr. Amatrudo in all but three categories. Moreover, in his 1978 year-end inventory of personnel, Mr. Amatrudo described Mr. Cowan as an agent "Promotable to Sales Manager this year." Mr. Lely was not listed in that inventory. On February 1, 1979, Mr. Amatrudo listed Mr. Cowan as a black agent "currently qualified for advancement" on an affirmative action program implementation form.

NOTES

[*] The Hon. Ralph K. Winter, U.S. Circuit Judge for the U.S. Court of Appeals for the Second Circuit, sitting by designation.

[1] Plaintiff originally claimed that he should have been promoted ahead of Vincent Rella in early 1978. At trial and in oral argument, counsel for Mr. Cowan dropped that contention. Plaintiff further concedes that the Czel promotion occurred outside the three-year limitations period in this case. At issue for liability purposes, then, are the Shepard, Lely, and Cecere promotions. However, the circumstances surrounding the Czel promotion remain relevant as probative of intent with regard to the promotions within the limitations period.

[2] In early 1978, Mr. Cowan attended Prudential's in-house Agent Career Development Program ("ACDP"). He was selected for the program by Mr. Cistulli, on the basis of his "progress, commissions, and general attitude." In a letter to Mr. Cowan confirming his selection, Mr. Amatrudo described Mr. Cowan and the other three agents selected as the "cream of the crop." Although he missed the first two sessions for various reasons, Mr. Cowan successfully completed ACDP on March 7, 1978. Mr. Cowan also enrolled in an outside training course, Life Underwriters Training Course ("LUTC"), at the suggestion of Mr. Amatrudo. Although Mr. Cowan's ACE form prepared in November 1978 indicates that he completed both parts of LUTC, Mr. Cowan dropped out of the second part of the course in November after another student in the class, Steven Shepard, was promoted to sales manager instead of Mr. Cowan. Mr. Cowan also attended a number of informal training sessions when he considered them useful, especially after he was told by Mr. Amatrudo that the sessions were important. I find, however, that the majority of these sessions were designed for new, inexperienced agents and thus were of little relevance to Mr. Cowan.

[3] On May 18, 1978, Mr. Cowan was placed on probation by Mr. Cistulli. The incident that gave rise to the probation involved a transaction known as a "surrender of paid up additions." Prudential encourages its agents to induce policyholders to invest dividends from existing policies in new policies. To facilitate this transfer, Prudential offers policies with premium payments exactly matched to the amount of dividends that would otherwise accumulate in existing policies. When one of these new policies was sold, the policyholder was required to sign a form, known as Form 80695, once a year to release the annual dividends for payment of the new policy's premium. Mr. Cowan had made such a sale and had gotten a policyholder, a Mr. Smith, to sign the form to release the dividends. The dividend check was then sent to Mr. Cowan, who, unable to meet in person with Mr. Smith to get his endorsement, himself endorsed the check, signing Mr. Smith's name. After learning what had happened, Mr. Smith agreed that he wanted the money used to purchase a new policy.

Forging the policyholder's name was a violation of Prudential's company policy, and Mr. Cowan's action was ostensibly treated as a very serious matter. The letters written to Mr. Cowan and to the district manager, Mr. Amatrudo, concerning the Smith incident were stern in tone, and the penalty, probation, ordinarily lasted for two years.

Despite these external indications of Prudential's highly negative view of the matter, however, the testimony of Mr. Amatrudo and Mr. Cistulli made it clear that Mr. Cowan's probation was not in fact taken seriously and was not a bar to his promotion. Mr. Cistulli indicated that he would generally remove an agent from probation if the agent's district manager believed the agent had learned his lesson, as would have been the case had Mr. Amatrudo recommended Mr. Cowan for promotion. Mr. Amatrudo testified that he had "completely lost sight of the fact that" Mr. Cowan was on probation during 1978 and 1979. Furthermore, Mr. Cowan's probation was never mentioned when he was offered a sales manager job in the Bridgeport District. See supra. Indeed, Prudential's practice of forwarding dividend checks to the agents instead of the payees in these circumstances deliberately ran the risk that overworked and time-pressured agents would endorse the checks themselves. I conclude that the probation was intended more for "the record" than as a true disciplinary action, and had no impact on Mr. Cowan's consideration for promotion. Prudential does not contend otherwise so far as the promotion is concerned.

[4] Prudential emphasizes that Mr. Amatrudo once recommended that a black agent be promoted to sales manager. However, that was when Mr. Amatrudo was an acting district manager in another district and the agent would not be serving under him as sales manager, a clearly distinguishable situation.

[5] Mr. Amatrudo testified that promotions were known in advance of their effective date.

[6] Many such inconsistencies are noted in the text, and a complete list would unduly burden this opinion. However, one other example is worth noting. In Paragraph 26 of his affidavit, which is referred to in the record as a "Declaration," Mr. Amatrudo stated, "I should note that a high account arrears percentage and a low lapse rate can often indicate the Agent is failing to report policy lapses." This of course is a perfectly sensible observation. Nevertheless, Mr. Amatrudo abandoned it at trial when he was questioned about the fact that Mr. Lely had high arrears percentages and a low lapse rate. He first appeared to waffle on the accuracy of the affidavit's observation. Extended questioning on the issue then led to this testimony:

THE COURT: If a high account arrears percentage is evidence of a number of lapses, doesn't it follow that a low lapse rate may be an indication of delayed lapses?

THE WITNESS: No, sir. I'm sorry, but regardless of what my declaration says, that's not necessarily the way it is. Often the people with the low lapse rates don't have a problem with delayed lapses.

THE COURT: Okay, it doesn't say that. It says when you have both a high arrears percentage and a low lapse rate, that it may often indicate failure to report lapses.

THE WITNESS: I don't think that I declared it completely accurately, your honor.

[7] Prudential either did not keep or has not produced records of critically delayed lapses prior to December 1978.

[8] Another example of a post-hoc rationalization is the statement in Mr. Amatrudo's affidavit that he promoted Mr. Shepard in part because he had completed the LUTC II course and Mr. Cowan had not. At the time Mr. Shepard was promoted, however, both men were in the course; Mr. Cowan thereafter left in anger, believing the course was not helpful. The records later checked by Mr. Amatrudo to prepare his affidavit showed that Mr. Shepard had completed the course and Mr. Cowan had not. Mr. Amatrudo then seized upon that fact as a reason for his decision although it clearly could not have been.

[9] Prudential sought at trial to introduce the Equal Employment Opportunity Commission's finding of no probable cause for Mr. Cowan's administrative complaint. Admissibility was argued under Fed.R.Evid. 803(8) (C). After considering relevant legal authorities, the court indicated its preliminary view that the E.E.O.C. finding would be of little probative value, a view confirmed by the critical importance of the testimony given at trial by Mr. Amatrudo. The court did invite counsel for Prudential to make an offer of proof. However, that invitation was not pursued, and the issue has therefore been waived.

[10] Using January 1980 as the date of resignation perhaps views the constructive discharge claim in an overly favorable light. Mr. Cowan actually stopped work in July 1979, when he got wind of the fact that Mr. Lely was to be promoted. Moreover, he accepted a job with another insurance company on December 15, 1979.

[11] Mr. Cowan also claims that the offers of sales manager positions in poor districts were a reason to resign. These offers, however, are inseparable from the failure to promote him in Stamford, and, if anything, ameliorate the effects of that failure.

[12] For example, Mr. Amatrudo had exerted efforts to get Mr. Cowan the invitation to the regional business conference in 1977. See infra.

[13] Mr. Cowan also complains about certain comments made by Mr. Collella at the October meeting. Those comments related generally to Mr. Collella's feeling that Mr. Cowan should persevere at his job and surmount perceived racial discrimination. Mr. Collella mentioned Jackie Robinson and Martin Luther King as role models. These comments may have been inappropriate, but like other incidents discussed supra, were not motivated by racial animus and could not have justified Mr. Cowan's resignation.