MANSFIELD, Circuit Judge:
Citibank, N.A. ("Citibank") appeals a decision of the Southern District of New York, Robert L. Carter, District Judge, refusing to set aside a jury award of $75,000 in damages to Eva Martin, a black woman who formerly worked as a teller at one of its branches. The jury found discrimination under 42 U.S.C. § 1981 (1982) and intentional infliction of emotional distress under New York law, based on Citibank's selection of six minority employees, including Martin, for polygraphing during an investigation of missing funds at the bank. Only one white employee was polygraphed. Citibank also appeals a decision by Judge Carter that the bank's conduct constituted discrimination under Title VII of the Civil Rights Act of 1964, 42 U.S.C. §§ 2000e to e-17 (1982). The bank maintains that there was insufficient evidence as a matter of law to demonstrate either discrimination in violation of § 1981 or Title VII or to constitute intentional infliction of emotional distress under state law. We agree and reverse.
Martin cross-appeals from Judge Carter's dismissal of her claims under § 1981 and Title VII of constructive discharge from the branch to which she was transferred after the polygraph incident and denial of her claim for injunctive relief against future Title VII violations. Finding no error in either of these rulings, we affirm them.
At the end of September 1980, Eva Martin began work as a teller at Citibank's City Hall branch, 250 Broadway, New York City. Martin's responsibilities as a teller included processing deposits at Citibank's automated teller machines and express deposit boxes. Her work record was admirable, her relations with co-workers were very good, and at one point she assisted the bank in foiling a robbery.
In March, April, and May of 1981, the bank experienced several unexplained disappearances of items including cash, checks, a money order and a passbook. About half of the missing items had been deposited in the automated teller machines or express deposit boxes. The others disappeared from the work area near the head teller's station.
Citibank's Internal Investigations Unit began an inquiry, directed by Geoffrey Obici, an assistant manager in the bank's investigation and potential loss department and an investigator in its audit department. Branch officials forwarded to Obici four memoranda describing the various disappearances and identifying seven individuals known to have been involved in transactions affecting some of the items that later disappeared. Neither the memoranda nor bank records contained any information as to the race of these seven individuals, five of whom were black, one hispanic and one white. Obici also spoke several times with Assistant Branch Manager Larry Cook, who gave him the names of the operations employees at the bank. Martin's name was not mentioned in any of the memoranda analyzing the missing items.
During the relevant period the City Hall branch had 26 employees, including Martin, of whom 15 were on its operations staff,
Obici began to set up interviews with employees on the operations staff in early June 1981. By the time he concluded his investigation, seven of the bank's operations staff employees had been polygraphed, including Martin. Of those seven, five were black, one hispanic and one white. The first person polygraphed, a hispanic, tested inconclusively, but after further discussion was sent back to work. The second person tested was cleared, but the third failed the polygraph and was discharged. Martin, tested immediately afterwards, was cleared and sent back to work when her test revealed nothing suspicious. Three others were subsequently tested, including one white woman. All three cleared.
Despite being exonerated, Martin was upset at having been suspected and questioned Cook repeatedly as to why only blacks had been tested, particularly since he had told her initially that everyone at the branch would be polygraphed. Still discontent after her return from sick leave (due to a burn) she arranged, at her own request, to be transferred to the bank's 23rd Street Branch. At the 23rd Street Branch, Martin complained of harassment by her superiors and co-workers, including: the loud mention by her supervisor in public of the polygraphing; twice being required (contrary to bank policy) to process deposits while serving customers; being given the wrong combination to the night deposit safe on one occasion; once having her deposits blocked from being credited by someone using her supervisor's computer card; and having her lunch hour repeatedly changed. As a result of these events, she resigned about four weeks later. Bank records reflect that a week before she left she had received a warning regarding complaints from customers and co-workers about her attitude.
After exhausting her administrative remedies Martin sued Citibank, alleging claims of disparate treatment and constructive discharge under Title VII and § 1981 and alleging intentional infliction of emotional distress under New York law. She based both her disparate treatment claims and her intentional infliction of emotional distress claim on the theory that she was polygraphed because of her race.
At the three-day trial held during April 1984 Martin testified about her experiences at both branches. She also stated that Clarence Sexton, a black supervisory teller at the City Hall Branch and one of those polygraphed, had in response to her question as to why he was polygraphed despite his eight years experience there, replied,
Martin's sole other witness was a statistical expert, Dr. Jeffrey Tanaka, who testified that there was only a .5% chance that a randomly selected group of seven individuals chosen from the 26 employees at the bank would include six or more minorities. In addition, he said that there was a 3.2% chance of randomly selecting from the 15 people on the operations staff a group of seven that included at least six minority persons. This analysis counted blacks and hispanics as minorities. Tanaka stated that statisticians consider a phenomenon statistically significant (that is, unlikely to have happened accidentally) if the probability of it happening by chance is under 5%. On this basis, he found race a significant factor in the bank's selection for testing.
Citibank presented testimony from Obici that he did not know the race of the employees at the branch, that bank records do not contain race information, and that he did not learn of Eva Martin's race until she arrived for the interview. He stated that he was the one who selected individuals for polygraphing and that he made his choices from the operations staff, based chiefly on each person's potential access to the missing funds. He stated that he placed three people with whom he had had prior dealings when he had personally worked in the City Hall Branch (two white, one black) on the bottom of his mental list because he considered them "least likely" to have been involved in the thefts and that he did not recall having been given the names of two operations staff members (one white and one Indian). Similarly, Cook testified that he had given Obici the names of the operations employees and that he himself had no say in determining who would be tested or in discontinuing the investigation. Cook specifically denied having told any investigator the race of the branch employees.
A statistical expert for Citibank testified that the analytic method used by Martin's expert was inappropriate since the selection of individuals for testing was not a random process. The Citibank statistician also reanalyzed the data examined by plaintiff's expert, using the same statistical method. He pointed out that if the hispanic teller were not tallied as a minority person, the selection from the operating staff of five blacks and two non-blacks for testing would have happened by chance 10% of the time, which is not statistically significant. Citibank also presented testimony from bank employees denying Martin's allegations of harassment and relating instances of rudeness and uncooperativeness by Martin after her transfer to the bank's 23rd Street Branch.
At the close of the evidence Judge Carter granted a defense motion for a directed verdict on the constructive discharge claim but denied that motion as to the other claims. He reserved decision on the Title VII discrimination claim (which was tried to him, not to the jury). The jury returned a verdict for plaintiff under § 1981 and New York law and awarded her $50,000 in compensatory damages plus $25,000 in punitive damages. Judge Carter then decided against Martin on the Title VII claim, stating that she had not demonstrated that defendant's non-discriminatory explanation for its actions was pretextual.
Citibank then moved under Fed.R.Civ.P. 50(c) for judgment notwithstanding the verdict (JNOV) on the ground that there was insufficient evidence to support the jury's finding of discrimination. Martin in turn moved to change the judgment on the Title VII claim on the ground that the judge was bound to accept the jury's finding of intentional discrimination. The judge denied Citibank's motion for JNOV and reversed his ruling on the Title VII discrimination claim as Martin requested. He did not, however, award additional damages, and he denied Martin's request for injunctive relief. Both sides appeal.
Title VII and § 1981
In a disparate treatment case the standard for liability under Title VII, which the parties have agreed is the same as that
In reviewing a case after it has been tried, the focus must be on the central issue in the case — whether there has been intentional discrimination against the plaintiff — rather than on each separate step of the McDonnell analysis, see U.S. Postal Service Bd. of Governors v. Aikens, 460 U.S. 711, 713-15, 103 S.Ct. 1478, 1481-82, 75 L.Ed.2d 403 (1983), and the court must refrain from weighing the evidence or the credibility of witnesses. See Mattivi v. South African Marine Corp., 618 F.2d 163, 167-68 (2d Cir.1980). Rather, viewing the evidence in the light most favorable to the plaintiff and drawing all reasonable inferences in her favor, we may reverse only if we find that no reasonable fact-finder could have reached a verdict for the plaintiff. See C-Suzanne Beauty Salon, Ltd. v. General Insurance Co., 574 F.2d 106, 112 n. 10 (2d Cir.1978); Simblest v. Maynard, 427 F.2d 1, 4 (2d Cir.1970). When a finding of discrimination under Title VII is (as here) explicitly based on the verdict on the § 1981 claim, reversal of judgment for the plaintiff on the latter would require reversal as to the former.
In the present case Martin presented no direct evidence from which discrimination might be inferred. The hearsay testimony about Sexton's comment has no bearing on the existence of discriminatory selection: Sexton had nothing to do with the selection process and his remark was made after Martin had complained about polygraph selection on the basis of race and had asked Sexton why he had been tested. Cf. Haskell v. Kaman Corp., 743 F.2d 113, 121 (2d Cir.1984) (finding error in the admission of comments from other employees about the suspected age motivation behind their terminations). Sexton's comment is of even less value in view of the fact that there were clear non-discriminatory reasons for the bank's decision to have him polygraphed: he was named in Cook's pre-polygraph memoranda as having handled three transactions involving items that later disappeared.
The evidence at trial plainly established that there was good cause for an investigation at the bank and that Martin was one of the employees with access to the areas in which items had disappeared. There was no basis for the jury to find that Obici, the person who made the selections for polygraphing, knew the race of those he selected for polygraphing. He testified that he did not know the race of employees, and there was no evidence that he received such information from Cook — in fact both denied that it was furnished to him. Any inference to the contrary would have been speculation.
Admittedly, as Judge Learned Hand has observed, "the denial of one, who has a motive to deny, may be uttered with such hesitation, discomfort, arrogance or defiance, as to give assurance that he is fabricating, and that, if he is, there is no alternative but to assume the truth of what he denies." Dyer v. MacDougall, 201 F.2d 265, 269 (2d Cir.1952). However, the Dyer opinion went on to hold that while "in strict theory a party having the affirmative might succeed in convincing a jury of the truth of his allegations in spite of the fact that all the witnesses denied them, we think it plain that a verdict would nevertheless have to be directed against him." Id.; accord Davis v. National Mortgagee Co., 349 F.2d 175, 178 (2d Cir.1965); 9 C. Wright & A. Miller, Federal Practice & Procedure § 2527, at 563 (1971) ("If all of
The only other evidence of discrimination presented by Martin is the statistical analysis, based on the selection of six minority persons and one white for polygraphing. We have previously held that such statistical proof alone cannot ordinarily establish a prima facie case of disparate treatment under Title VII or § 1981. Hudson v. International Business Machines Corp., 620 F.2d 351, 355 (2d Cir.) (finding no error in the ruling that statistics about a company's record of promoting employees of different races, standing alone, could not establish a prima facie case of disparate treatment), cert. denied, 449 U.S. 1066, 101 S.Ct. 794, 66 L.Ed.2d 611 (1980); accord Grigsby v. North Mississippi Medical Center, Inc., 586 F.2d 457, 459-61 (5th Cir.1978); Murphy v. Middletown Enlarged City School District, 525 F.Supp. 678, 692 (S.D.N.Y.1981) (applying the Hudson rule). Although plaintiff tries to draw a distinction between "generalized" and "particularized" statistical proof and argues that the latter can be sufficient to prove discrimination, we find no logical basis for attaching any legal significance to these characterizations and no supporting case law for the distinction. As the statistical proof is not sufficient to carry the case to the jury in itself, neither is the statistical evidence in conjunction with the jury's possible disbelief of Cook and Obici. Any other result would cripple the operation of the Hudson rule.
The value of the statistical proof alone is slight here, since the analysis conducted by Tanaka, Martin's expert, is weak. That analysis depended on the erroneous assumptions that racial bias was the only factor that might have explained the composition of the group polygraphed and that aside from any race-related prejudice, the selection of employees for polygraphing was made at random. In fact, however, the evidence at trial made clear that selection quite reasonably was not random: Obici testified that he arranged polygraphs for those he believed to have been involved in transactions affecting the missing items, that he had planned to delay polygraphing the individuals believed by him on the basis of his prior association with them to be "least likely" to be involved, and that he did not recall being given the names of two of the 15 employees on the operations staff. Obici ordered polygraphs for five of the seven employees named in Cook's memoranda as known to be connected with transactions involving the items that were missing from the later part of the relevant period. The only persons named in the memos but not polygraphed were (1) a black woman, who was implicated to a lesser extent than the others and was personally known to Obici, and (2) a black man, who no longer worked at the bank.
In the context of an investigation into all of the members of the operations staff, the decision to polygraph Martin is in no way unusual. Nor does it in any way, standing alone, support an inference that she was selected because of her race. The mere fact that the investigation concluded after a scant half of the operations staff had been polygraphed bears no racial implications. Obici testified that after he had seen the results for the seven individuals tested, he and his supervisor decided to bring the investigation to a halt. He stated that the bank tries not to over-use the polygraph because the testing adversely affects morale, and that, having had one person fail the polygraph and five people clear it, he and his supervisor felt that continued polygraphing would not be productive. There was no contrary testimony by any witness.
Martin suggests that we can infer racial animus from the fact that the first five people polygraphed were black or hispanic and from the further fact that the only white person tested was a person whose name appeared four times in the memoranda and who was selected after Martin had complained about the polygraphing of blacks. Given the stipulation at the trial that "Interviews were scheduled according to the availability of Mr. Obici and the employee to be interviewed," we cannot agree that the ordering of those selected supports an inference of discrimination.
Intentional Infliction of Emotional Distress
Since Martin's state law claim of intentional infliction of emotional distress (IIED) was directly based on her allegation that she was polygraphed because of her race, the evidence is also insufficient, for the reasons detailed above, to establish her New York law claim. Even if this were not the case, under New York law the acts alleged by Martin, even if fully established, would not provide legally adequate grounds for a verdict of IIED.
New York, which uses the Restatement (2d) of Torts definition of intentional infliction of emotional distress, requires that the conduct be "so outrageous in character, and so extreme in degree, as to go beyond all possible bounds of decency, and to be regarded as atrocious, and utterly intolerable in a civilized society." Fischer v. Maloney, 43 N.Y.2d 553, 558, 402 N.Y.S.2d 991, 992-93, 373 N.E.2d 1215, 1217 (1978) (quoting Restatement (2d) Torts § 46(1)). The conduct must also be intentionally directed at the plaintiff and lack any reasonable justification.
New York courts have been very strict in applying these principles. One New York court recently dismissed an IIED claim despite allegations by the plaintiff that he was: transferred and demoted for reporting fraud at his company; told that he could not be fired because of his age but that he would never advance; discharged and ordered to leave immediately; forcibly and publicly escorted from the building by guards when he returned the next day to pick up his belongings; and ordered out of the building two weeks later when he came back, as instructed, to pick up his possessions, which were then dumped in the street. Murphy v. American Home Products Corp., 112 Misc.2d 507, 447 N.Y.S.2d 218, 219-20 (N.Y.Sup.), aff'd, 88 A.D.2d 870, 451 N.Y.S.2d 770 (1st Dept.1982). In affirming this dismissal, the New York Court of Appeals stated that the allegations fell "far short" of the standards for an IIED claim. Murphy v. American Home Products Corp., 58 N.Y.2d 293, 461 N.Y.S.2d 232, 236, 448 N.E.2d 86, 90 (1983). When such conduct does not establish an IIED claim the proof in the present case is a fortiori inadequate. See also Belanoff v. Grayson, 98 A.D.2d 353, 471 N.Y.S.2d 91, 94 (1st Dep't 1984) (finding insufficient proof of IIED based on allegedly unjustified negative performance ratings and termination claimed to result from sex discrimination).
Thus, despite the unacceptability of racial discrimination in civilized society, we have seen no indication that New York courts would consider the conduct here "outrageous" per se or would otherwise conclude that it sufficed to prove Citibank's liability for intentionally inflicting emotional distress.
Since Martin did not present sufficient evidence to prevail on any of her claims,
Martin's Cross-Appeal: The Constructive Discharge Claim
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 Saving and Loan Ass'n, 509 F.2d 140, 144 (5th Cir.1975)).
In this case the evidence was insufficient as a matter of law to establish constructive discharge. In considering the sufficiency of evidence, we accept as true plaintiff's testimony that: her supervisor loudly mentioned her having been polygraphed; complaints concerning her attitude to co-workers were unfounded; her supervisor had once given her the wrong combination to the night deposit box and that someone using his card had once interfered with her deposits; and that she had been required to process deposit records while serving customers. Plaintiff admitted that she had not been formally disciplined but acknowledged that she had received an informal, oral warning concerning complaints about her attitude from several customers and from co-workers.
As Judge Carter correctly concluded in directing a verdict in Citibank's favor on Martin's claim of constructive discharge, these incidents do not legally suffice to sustain an inference that a reasonable person would have been "compelled" to resign. Compare Bourque v. Powell Electrical Mfg. Co., 617 F.2d 61 (5th Cir.1980) (resignation due to lower pay resulting from sex discrimination does not constitute constructive discharge); Muller v. United States Steel Corp., 509 F.2d 923, 929 (10th Cir.), cert. denied, 423 U.S. 825, 96 S.Ct. 39, 46 L.Ed.2d 41 (1975) (unfavorable job assignment and discriminatory failure to promote does not constitute constructive discharge), with Goss v. Exxon Office Systems Co., 747 F.2d 885, 888-89 (3d Cir.1984) (holding it sufficient for constructive discharge that, allegedly as the result of sex discrimination, plaintiff was verbally abused and threatened and forcibly transferred to a position with substantially lower salary and inferior working conditions); Meyer v. Brown & Root Construction Co., 661 F.2d 369, 371-72 (5th Cir.1981) (involuntary transfer of pregnant employee to heavy manual labor that posed substantial risks to her health constituted constructive discharge).
We reverse Judge Carter's decision not to grant a JNOV in favor of Citibank on the claims of § 1981 discrimination and intentional infliction of emotional distress, and reverse his decision amending his ruling on the Title VII discrimination claim. We affirm the directed verdict for defendant on the constructive discharge claim. The case is remanded to the district court with instructions to enter a judgment in accordance with this opinion.
Conversely, if the Indian employee were counted with the blacks and hispanics as a minority person, then using the same statistical method employed by plaintiff's expert, there would be more than an 8% chance that six or more minority persons would be included in a random sample of seven employees. See P. Pfeiffer & D. Schum, Introduction to Applied Probability 21, 34 (1973) ("Pfeiffer").
Martin suggests that the statistics concerning the probability of selecting the polygraphed individuals from the entire branch staff were more probative of discrimination than the statistics concerning selection from the operations staff. The analysis performed by her expert, however, erroneously assumed that aside from any racial effect he was able to identify, each employee had an equal chance of being selected for testing. Since Obici testified that he selected employees to be polygraphed from the operations staff and since every individual tested was a member of the operations staff, it is incorrect to assume that members of the entire branch staff, including operations staff employees, had (apart from their race) an equal chance of being selected for polygraphing. As such, the value of statistics concerning the chances of selecting the polygraphed individuals from the staff of the entire branch is much reduced, if not minimal. See Pfeiffer, fn. 4, supra.