Volume XIV, Issue V
September/October 2000

Correcting a Supreme Mistake

by Geoffrey William Hymans

"I lie, therefore I discriminate."

In spite of the lack of any logical connection between these two actions, the United States Supreme Court has concluded that proof of the one can serve as proof of the other. It is rare that a State Supreme Court has the opportunity to "correct" a mistake made by the US Supreme Court, but in the case BCTI v. Hill, 97 Wn.App. 657, review accepted (1999), the Washington State Supreme Court has just such an opportunity.

Allowing proof of a lie to serve as proof of discrimination will lead to a great increase in the number of discrimination lawsuits filed. This boon to the plaintiff's bar came in a recent US Supreme Court decision that has been overlooked in favor of decisions that immediately resonate in the media, but the case has potentially far reaching implications in the employment law field. Further, the case seems to have been given short shrift by the Court itself, perhaps because it was allocating the bulk of its efforts to cases it knew would be more closely scrutinized in the popular press. Yet this case demonstrates, as Justice Scalia aptly noted in dissent in another ill-reasoned decision this term, that opinions of the current Court which command a majority of the nine votes often fail - in rudimentary terms - to even "make sense."

The case at issue is Reeves v. Sanderson, No. 99-536, June 12, 2000. The case purported to resolve a rare 6-5 conflict in the Federal Circuits on the issue of, in Justice O'Connor's words, "whether a defendant is entitled to judgment as a matter of law when the plaintiff's case consists exclusively of a prima facie case of discrimination and sufficient evidence for the trier of fact to disbelieve the defendant's legitimate, nondiscriminatory explanation for its action." In simple terms: does proof of a lie, together with being old, female, disabled, or a minority, and having been demoted or fired equal proof of invidious discrimination? Amazingly, the Court said "Yes."

The issue arises out of the Court's framework for proving discrimination cases: the ubiquitous McDonnell-Douglas test. This "burden-shifting scheme" initially forces the plaintiff to make out a "prima facie" case. The plaintiff must show that they were a member of the protected class, that they were doing satisfactory work, that they suffered an adverse employment action (e.g. demotion, reprimand, firing), and that they were replaced by someone outside the protected class. The purpose of this preliminary showing is to force the defendant to put forward a "legitimate, nondiscriminatory reason" for the adverse employment action. Once the "presumption" created by this prima facie case serves its purpose, it becomes "irrelevant" and "drops from the case."

After the employer produces a neutral reason for its actions, the burden again shifts to the plaintiff, but the burden is now one of persuasion rather than production. The plaintiff must prove that "the proffered reason was not the true reason for the employment decision and that race was." Put another way, "the ultimate burden of persuading the trier of fact that the defendant intentionally discriminated against the plaintiff remains at all times with the plaintiff."

The question that arose after the Court's last major foray into its McDonnell-Douglas test, the St. Mary's case in 1993, turned on the language of the test itself. In seeking to get past the "pretext" prong of the test, is it sufficient to show that the employer lied about its proffered reason for firing the plaintiff (i.e. that its reason was a "pretext"), or the plaintiff must not only show that the employer lied, but introduce some evidence that the employer lied because it was covering up for discrimination?

Thus, in the typical shorthand of legal specialists, the circuit Courts and the states sorted themselves into "pretext-only" or "pretext-plus" camps. "Pretext-only" states allowed discrimination cases to go to a jury if a plaintiff produced evidence that the employer's proffered reason was "unworthy of credence," even if they produced no evidence of discrimination, while the "pretext-plus" Courts insisted on at least some additional evidence.

The St. Mary's case led to the rare 6-5 split in the circuits. It also led to a split among the Divisions of the Washington State Court of Appeals (see further discussion below). The US Supreme Court faced the converse of the Reeves question in the St. Mary's case: whether proof that the employer's decision was a lie entitled the plaintiff to judgment as a matter of law. The Court held that it did not, but uttered seemingly conflicting statements on the issue of what minimum proof the plaintiff must offer. The Court stated that disbelief of the employer's proffered reasons will permit an inference of intentional discrimination, citing the Court of Appeals below that "no additional proof of discrimination is required." Yet the Court contradicted this astounding statement later in the opinion, when the Court explained that McDonnell-Douglas "does not say ... that all the plaintiff need do is disprove the employer's asserted reason," but rather that the employee demonstrate "that whatever the stated reasons for his rejection, the decision was in reality racially premised." St. Mary's review of the prior caselaw (focusing on McDonnell-Douglas and Burdine) led the Court to conclude that "nothing in law would permit us to substitute for the required finding that the employer's action was the product of unlawful discrimination, the much different (and much lesser) finding that the employer's explanation of its action was not believable."

Thus, the confusion evidenced in the circuits and reflected in Washington's appellate divisions. Clever appellate courts have either faced these competing statements head on and chosen the side that suits them, or have resolved the dilemma by reexamining definitions. But the Supreme Court chose not to clear up the confusion created by its St. Mary's opinion until Reeves, a case which was factually unsuited for the task. Indeed, the Court, in its haste to resolve this split, may have hamstrung itself by taking the first available case to come down the pike, rather than waiting for a more suitable opportunity.

Reeves was an unpublished age discrimination decision out of the 5^th Circuit. In addition to the facts summarized in Reeves which cast doubt on the employer's proffered reason for firing the plaintiff, there were comments made by the employer to the effect "(1) that Reeves was so old that he 'must have come over on the Mayflower' and (2) that he was 'too damn old to do the job.'" Needless to say, in a slightly more sympathetic Court these comments may have provided evidence that would even have met the "pretext-plus" standards. Perhaps it was the Supreme Court's visceral response to the 5^th Circuit's position that even in light of this evidence the case should not see a jury that led to the result in Reeves.

That result is illogical, irrational, and just flat-out wrong. Amazingly, the Court spends little time sorting through the St. Mary's maze, or reviewing the meaning of the standards developed in McDonnell-Douglas or Burdine. Indeed, in flat contradiction to the statements in St. Mary's that the "presumption" (which is really no more than a burden of production) created by the prima facie case "drops out of the picture" once the employer produces a neutral reason, the Court stated "the trier of fact may still consider the evidence establishing the plaintiff's prima facie case and inferences properly drawn therefrom ... on the issue of whether the defendant's explanation is pretextual." The Court then drops the bomb:

"Proof that the defendant's explanation is unworthy of credence is simply one form of circumstantial evidence that is probative of intentional discrimination, and it may be quite persuasive.... In appropriate circumstances, the trier of fact can reasonably infer from the falsity of the explanation that the employer is dissembling to cover up a discriminatory purpose. Such an inference is consistent with the general principle of evidence law that the factfinder is entitled to consider a party's dishonesty about a material fact as 'affirmative evidence of guilt.'" (Citations omitted.)

What? It is essential to parse these statements carefully, for they represent either a misapplication of a general evidence doctrine, an incredibly cynical view of the state of race relations in this country, or perhaps both.

First, what inferences that can be drawn from the "prima facie case" can shed light on whether the employer's explanation is a pretext? The purpose of the doctrine, as enunciated in McDonnell-Douglas and later confirmed in St. Mary's, is to force the employer to explain his or her action; the burden shifted is one of production, not persuasion. The prima facie case, which can be reduced to the fact that something bad happened to someone who is a member of a legally protected class, sheds no light on the veracity of the employer's explanation. "Inferences" of discrimination can only be drawn if one accepts the premise that in the absence of any explanation, the most likely force behind an "adverse employment action" taken against someone in a protected class is invidious discrimination. This incredibly cynical deduction will be revisited later in analyzing the substantive outcome -- a finding of discrimination, as opposed to proof of pretext -- but it is no less disturbing in the warm-up. This is not the world I want to live in, nor the world I believe most Americans inhabit today.

Turning to the substantive case, i.e. what is proof of invidious discrimination, the Court's next statement is baffling. The Court cites a perfectly valid general evidentiary principle, but then distorts and misapplies this principle.

First, the employer's explanation for the firing may not even be a "material fact." Black's Law Dictionary, 6^th Edition, defines a material fact as one in which "proof of that fact would have the effect of establishing or refuting one of the essential elements of a cause of action or defense asserted by the parties..." The only "elements" in a discrimination case are that an adverse employment action occurred (usually not disputed), and the motivation of the person who took that adverse action. So without being able to read minds and determine conclusively what motivated an individual — and remember, this is in the absence of comments or other indications that might meet a "pretext-plus" standard and shed further light on the defendant's motivation — we are forced to conclude that the only way the lie about the reason for the adverse employment action is material is if we presume that in the absence of any other explanation, the most likely reason is invidious discrimination.

Lying about the reason for an adverse employment decision may reinforce other evidence of invidious discrimination, but the Court treats the lie itself as sufficient evidence on the ultimate issue — whether the employer discriminated — to sustain liability. This is, as I said before, simply wrong.

People lie everyday in our society. Any rational person would concede that lying is far more prevalent than discrimination, and that people lie to cover up all sorts of unpleasant motives which are not discriminatory: nepotism, personal animosity, jealousy, megalomania, or just being a jerk. Indeed, most Americans today would agree that far more people are simply jerks than are bigots. A lie about the reason for terminating someone, even a member of a protected class, is just as likely — and in today's society probably far more likely — to be a "pretext" for a nondiscriminatory "moral" wrong as for discrimination. Proof of a lie about the reason for firing someone simply does not produce sufficient evidence of discrimination, even if that person is of a "protected class," to sustain liability.

Consider the following hypothetical: a female office worker is fired and replaced with a male office worker. The reason given was that she was accessing inappropriate internet sites from her computer terminal. During discovery in her gender-discrimination case, it is learned that a different password was used to access the computer network during the times when inappropriate material was viewed on the terminal than the password assigned to the female employee. In addition, during discovery the plaintiff learns that her replacement is a nephew of her former boss. Under the Reeves decision, a jury verdict of gender discrimination with this evidence will be upheld.

The Courts are not an employment ombudsman, seeking to correct every injustice that takes place in the office. The Courts are empowered by specific statutes to provide compensation for a specific wrong: invidious discrimination. But it is the Supreme Court's dated view of American society that underlies its Reeves decision, and indeed underlies the entire McDonnell-Douglas framework, that may have led to the Court's erroneous decision.

That view is simply summarized: In the absence of any other explanation for an adverse employment decision concerning a member of a legally protected class, the Court will presume invidious discrimination. This is at the heart of the McDonnell-Douglas prima facie case, but perhaps this is acceptable -- even thirty years after the McDonnell-Douglas decision (which came toward the end of the civil rights movement's heyday) -- because the burden placed was only one of production. As shown in St. Mary's, this view of society merely forced an explanation from the employer: it did not result in liability for the employer, even if that explanation were untrue, because the prima facie case did not add anything to the burden of persuasion that invidious discrimination had taken place.

If the fact that something bad happened to a member of a protected class did raise any sort of presumption of discrimination that survived the employer's proffer of a lie, St. Mary's would have to have resulted in a verdict for the plaintiff. In the absence of any other explanation (the employer's explanation having been disproven) a presumption of discrimination should have been sufficient "evidence" to support liability. If no other evidence was adduced, this should logically have resulted in a directed verdict for the plaintiff. But the Court in St. Mary's said that this wasn't the case, and that the employee still had to prove discrimination. Of course, such proof would only be necessary if any presumption didn't itself serve as such proof.

But in Reeves, the Court states the opposite. Proof that the employer's reason is a lie leaves us with an adverse employment action visited on a member of a protected class for which we have no explanation. This is the same situation as in St. Mary's, and if there isn't a presumption of discrimination that is revived by proof of the lie, then allowing liability in this situation "permit[s] us to substitute for the required finding that the employer's action was the product of unlawful discrimination, the much different (and much lesser) finding that the employer's explanation of its action was not believable." This is exactly what St. Mary's stated was not permitted by anything "in law."

What is truly amazing is that Justice Scalia, the author of St. Mary's, and the other four judges who signed on to the opinion effectively refuted this statement by signing on to Reeves. Thus, these Justices acquiesced -- nay, were complicit in -- the judge-made expansion of liability for lying under the rubric of discrimination.

The Washington State Supreme Court has an historic opportunity to "correct" this error. How, you might ask? Because many, if not most, discrimination claims in this state are filed under state statute (Chapter 49.60 RCW) rather than under federal law, such as the ADEA. While Washington adopted the proof scheme in discrimination cases from that developed in federal courts, it goes without saying that RCW 49.60.180 (outlawing discrimination in employment) is a state statute, and the highest authority interpreting state statutes is the Washington State Supreme Court. "The [Washington] Supreme Court has the ultimate authority to determine the meaning and purpose of a [state] statute." Hubbard v. Department of Labor & Industries of State of Washington, 140 Wn.2d 35 (2000). As noted in one of the foundational Washington age discrimination cases:

"While these federal cases are a source of guidance, we bear in mind that they are not binding and that we are free to adopt those theories and rationale which best further the purposes and mandates of our state statute." Grimwood v. University of Puget Sound, 110 Wn.2d 355, 136-362 (1988).

Indeed, the Washington State Supreme Court has expressly addressed the precedential effect of the US Supreme Court's interpretation of a federal statute that is similarly worded to a Washington statute:

"It is well settled that the Supreme Court's construction of a similarly worded federal statute, although often persuasive, 'is not controlling in our interpretation of a state statute.' " Aviation West Corp. v. Washington State Dept. of Labor and Industries, 138 Wn.2d 413, 424 (Wash. 1999), citing Hoffer v. State, 113 Wash.2d 148, 151 (1989), quoting State v. Gore, 101 Wash.2d 481, 487 (1984)) (additional citations omitted).

The United States Supreme Court's error in reasoning in the Reeves decision thus has no binding effect on the Washington State Supreme Court, which can return the proof standard for discrimination cases to the rational requirement of actually proving discrimination.

The BCTI case serves as the perfect vehicle for this return to rationality. The plaintiff in that case introduced no evidence of age discrimination. After suing for disability, race, and age discrimination, and having had the disability discrimination dismissed at the close of her case, the plaintiff was awarded a verdict on the basis of age discrimination. The plaintiff introduced no evidence of age discrimination: no direct evidence, and not even a lick of circumstantial evidence. There were no ageist comments, not even the most circumstantial of statistical studies purporting to show that older workers were fired in greater numbers than younger workers. Nothing.

There was also no evidence of race discrimination, but there was an employer who, from the evidence introduced, appeared to have acted wrongly; from the evidence presented it could appear that the plaintiff's supervisors fired her rather than lose her through a transfer to another region of the company. Evidence was also introduced indicating that the company covered up this immoral motivation by claiming that the plaintiff was fired for "dishonesty." Thus, this case appeared to be the perfect instance of a jury doing "rough justice" – acting as an ombudsman to punish a company that may have done something morally questionable, even if the acts were not illegal.

But, as noted in the discussion of the Reeves case above, the antidiscrimination laws are designed to get at a specific wrong: discrimination. RCW 49.60.010 finds discrimination to be "a matter of state concern." It does not find "immoral acts by employers" to be a matter of state concern, and it goes without saying that the doctrine of employment-at-will, long the law in Washington, is hardly consonant with holding employers legally accountable for immoral actions.

In BCTI the Washington State Court of Appeals, Division II, surveyed the pretext/pretext-plus precedents from around the county, and closely examined the St. Mary's case (certainly more closely than the United States Supreme Court appeared to…). The court noted that Division I, in Sellsted v. Washington Mutual Savings Bank, 69 Wn.App. 852 (1993), had followed the federal 3rd Circuit Court of Appeals and opined that Washington was a "pretext-only" state. Division II noted, however, that Sellsted had been decided before St. Mary's.

Division II then returned to the source of the proof scheme in discrimination cases. The court found that the solution to the Gordian knot of pretext lies in reexamining the decision that started it all, McDonnell-Douglas. As Division II noted in BCTI, the definition of "pretext" used in McDonnell-Douglas was substantially broader than the dictionary definition. All of the examples cited by the Court as "evidence relevant to any showing of pretext" are actually direct and indirect evidence of actual discrimination, such as would meet even what is considered a "pretext-plus" standard today. Thus, the definition of pretext, as used in McDonnell-Douglas, includes both the elements of falsity and discrimination. The two apparently conflicting statements in St. Mary's can thus be harmonized, and the Reeves decision, and the footnote in Burdine on which the Reeves decision ultimately rests (see infra footnotes 17 & 18), can be shown to be an unwise expansion of the McDonnell-Douglas test that will ultimately result in liability for many employers who lie, but do not discriminate.

Perhaps the Reeves result was dictated by the 5^th Circuit's failure to give sufficient weight to statements by the employer which, in addition to proof that the employer's proffered reason was a lie, would have satisfied even a pretext-plus standard. Perhaps it reflects underlying views of society and assumptions about most folks' motivations that seem anachronistic today, yet may have been necessary in an earlier era through which all the Justices lived and practiced. Whatever the cause, the US Supreme Court should reverse Reeves at it earliest opportunity and return to the standards set forth in McDonnell-Douglas. Further, since Reeves and its proof requirements are simply the interpretation of a federal statute, states with their own anti-discrimination statutes should reject the reasoning of Reeves, and adopt the pretext-plus standard in interpreting their own statutes. The Washington State Supreme Court should take the lead in this regard, using as its vehicle the case perfectly suited to the purpose, BCTI v. Hill.

The Reeves and BCTI cases represent a fascinating microcosm of our evolving society. In the absence of any other explanation, can we as a matter of law assume that discrimination most likely motivated an adverse employment decision, and can we hold an employer liable on the basis of that assumption? One's opinion of whether discrimination is still endemic in our society will likely determine one's opinion on the legal question of the appropriate proof standard.

Yet the concept of precedent is dangerous here. Most individuals, even those who believe that rampant discrimination still exists, probably believe – or at least hope – that one day our society will have conquered this scourge. But once we have won the battle for the hearts and minds, the law, based on legal precedents such as Reeves, will still permit recovery for discrimination based on proof of a lie. Even if the pendulum of injustice still swings in favor of those who claim to be the victims of discrimination, at some point that pendulum will shift, and we will begin holding ever more employers who are innocent of discrimination legally liable. This is the peril of presumptions instead of proof. Yet, as Justice Scalia stated in another decision this term, "in judge-made law at least, logic will out."

Back to table of contents >>