Concurrence · Public-domain judicial opinion

Dewitt v. Proctor Hospital

Dewitt v. Proctor Hospital, 517 F.3d 944 (7th Cir. 2008).

POSNER, Circuit Judge, · 2008-02-27 · No. 07-1957

Opinion text

This transcription is from Harvard’s Caselaw Access Project. Consult the original source for printed page breaks and corrections.

POSNER, Circuit Judge, concurring. I agree with the decision and with most of Judge Evans’s characteristically lucid majority opinion. I write separately to raise two questions that seem to me to be worth flagging although they do not have to be answered to decide the case — the alternative ways of establishing a prima facie case of discrimination and their suitability to the discrimination charged in this case — and a third question — the difference between discrimination on grounds of expense and discrimination on grounds forbidden by federal law — regarding which the discussion in the majority opinion seems to me incomplete and potentially misleading. The standard understanding is that there are two ways to make out a prima facie case of discrimination — which is to say, a showing in advance of trial sufficient to defeat the defendant’s motion for summary judgment. They are the “direct method” and the “indirect method,” the latter being that of McDonnell Douglas Corp. v. Green, 411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973); see, e.g., Timmons v. General Motors Corp., 469 F.3d 1122, 1126-27 (7th Cir.2006). The direct method is the one that litigants would use if there were no McDonnell Douglas test. The plaintiff would marshal all his direct and circumstantial evidence of discrimination (“direct” here meaning, as distinct from its use in the phrase “direct method,” evidence that is not circumstantial' — which, in a discrimination case, usually means admissions) and the defendant would do the same with regard to evidence countering an inference of discrimination. The question would then be whether a reasonable jury could infer discrimination if the record at trial were the same as the record on summary judgment (except that some of the evidence given in documentary form in the summary judgment proceedings would have to be given in oral form at a trial). If so, the defendant’s motion for summary judgment would be denied. That is the basis for the reversal of the dismissal of the claim of disability discrimination in this case. The evidence used to establish a prima facie case under the indirect method would often be part of the circumstantial evidence presented under the direct method. In a case of racial discrimination, for example, the fact that a black employee had been replaced by a white one and the defendant was mum on the reason for the replacement would be some evidence of racial motivation. Another element of the indirect method might be missing — maybe the white had better qualifications and a slightly different kind of job and reported up a different chain of command. But that hole could be plugged by some additional evidence of discrimination, perhaps statistics on the racial composition of the defendant’s work force in relation to the pool of potential workers. “Any demonstration strong enough to support a judgment in the plaintiffs favor if the employer remains silent will do, even if the proof does not fit into a set of pigeonholes.” Carson v. Bethlehem Steel Corp., 82 F.3d 157, 159 (7th Cir.1996) (per curiam). So it was a mistake for the parties in this case to think that the way to litigate it was to address the two methods of establishing a prima facie case as if each were in its own sealed compartment. One consequence was that the defendant’s lawyer made a damaging tactical error by failing to produce any evidence (even a simple sworn denial) of a nondiscriminatory reason for terminating the plaintiff (though he claims to have such evidence), on the ground that the plaintiff had failed to establish a prima facie case by either method taken separately. Maybe so, but the plaintiff presented other evidence of discrimination besides the defendant’s silence on why it had fired the plaintiff, and that silence, though insufficient in itself, was enough (at least given another tactical blunder by the defendant, which I discuss later) to make out a prima facie case when added to the balance. “[W]hen a plaintiff in a discrimination case has direct evidence of discrimination as well as the indirect evidence required to make out a prima facie case under McDonnell Douglas he does not have to show that either approach, taken in isolation from the other, makes out a prima facie case — he can combine them.” Simple v. Walgreen Co., 511 F.3d 668, 670-71 (7th Cir.2007); cf. Logan v. Kautex Textron North America, 259 F.3d 635, 638-40 (7th Cir.2001); Burns v. AAF-McQuay, Inc., 96 F.3d 728, 732 (4th Cir.1996). So far I have assumed that McDonnell Douglas provides an appropriate method of establishing a prima facie case under the rarely litigated “association” provision of the Americans with Disabilities Act, the provision that forbids discrimination against “a qualified individual because of the known disability of an individual with whom the qualified individual is known to have a relationship or association.” 42 U.S.C. § 12112(b)(4); see Larimer v. International Business Machines Corp., 370 F.3d 698 (7th Cir.2004). As we explained in Larimer, and repeated in a slightly different context in Timmons v. General Motors Corp., supra, 469 F.3d at 1127-28, an attempt to clone the McDonnell Douglas test for use in an association case would enable a plaintiff to establish a pri-ma facie case merely by showing that the employer, knowing that the plaintiff (a qualified worker in the sense of meeting his employer’s expectations) had an association with a disabled person, fired the plaintiff or took other adverse employment action against him and replaced him with (or did not take similar adverse employment action against) someone who did not have a known association with a disabled person. 370 F.3d at 701-02. Yet the inference from these scanty facts that the association had induced the employer’s action would be too weak to justify forcing the defendant to produce evidence that he had taken the action for a different, an innocent reason. Some people do feel distaste for associating with a disabled person. But the employer (and employees whose prejudices the employer might share or condone) are not being asked in an “association” case to associate with a disabled person. He is not another employee, but a stranger to the workplace with whom one of the employees happens to have a relationship. Prejudice against an employee who merely has a relationship with a disabled person doubtless exists; as we pointed out in Larimer v. International Business Machines Corp., supra, 370 F.3d at 699-700; see also Tyndall v. National Education Centers, Inc., 31 F.3d 209, 214 (4th Cir.1994), an employer might worry that the plaintiff would be distracted from his work by the disability of the person with whom he had a relationship, or might think the disease creating the disability catching (perhaps the person is the plaintiffs husband and has AIDS). But that is a sufficiently rare occurrence to require the plaintiff to go beyond a bobtailed McDonnell Douglas test and show, as the court in Den Hartog v. Wasatch Academy, 129 F.3d 1076, 1085 (10th Cir.1997), put it, that “the adverse employment action occurred under circumstances raising a reasonable inference that the disability of the relative or associate was a determining factor in the employer’s decision.” See also Timmons v. General Motors Corp., supra, 469 F.3d at 1126-27; Larimer v. International Business Machines Corp., supra, 370 F.3d at 701; Hilburn v. Murata Electronics North America, Inc., 181 F.3d 1220, 1230-31 (11th Cir.1999); Ennis v. National Ass’n of Business & Educational Radio, Inc., 53 F.3d 55, 57-59 (4th Cir.1995). Otherwise the number of spurious suits would soar. An employee who, perhaps fearing the axe, wanted to prepare a discrimination case would have only to talk up at work his relationship to a disabled person; such relationships are not uncommon. An employer’s most likely concern about an employee who has a disabled relative, especially a spouse or child, is that the relative’s medical expenses may be covered by the employer’s employee health plan. There is a positive correlation between being disabled and having abnormally high medical expenses, just as there is a positive correlation between the age of an employee and his salary because most employees receive regular raises as long as they perform satisfactorily. Suppose a company encounters rough waters and decides to retrench by firing its most expensive employees. They are likely to be older on average than the employees who are retained, but as we said many years ago, and the Supreme Court confirmed in Hazen Paper Co. v. Biggins, 507 U.S. 604, 611-12, 113 S.Ct. 1701, 123 L.Ed.2d 338 (1993), “nothing in the Age Discrimination in Employment Act forbids an employer to vary employee benefits according to the cost to the employer; and if, because older workers cost more, the result of the employer’s economizing efforts is disadvantageous to older workers, that is simply how the cookie crumbles.” Karlen v. City Colleges of Chicago, 837 F.2d 314, 319 (7th Cir.1988); see also EEOC v. Francis W. Parker School, 41 F.3d 1073, 1076-78 (7th Cir.1994); Bramble v. American Postal Workers Union, 135 F.3d 21, 25-26 (1st Cir.1998); Adams v. Florida Power Corp., 255 F.3d 1322, 1325-26 (11th Cir.2001); cf. Troupe v. May Department Stores Co., 20 F.3d 734, 738 (7th Cir.1994). The majority opinion in this case does not cite any of these precedents or acknowledge the difference between distaste and expense as grounds for a discrimination suit — though the fault for the omission, as we’ll see, is the defendant’s. Now it is true, as we know from the discussion in the Larimer case of the “expense” form of association discrimination, 370 F.3d at 700-01, that an employer who discriminates against an employee because of the latter’s association with a disabled person is liable even if the motivation is purely monetary. But if the disability plays no role in the employer’s decision — if he would discriminate against any employee whose spouse or dependent ran up a big medical bill — then there is no disability discrimination. It’s as if the defendant had simply placed a cap on the medical expenses, for whatever cause incurred, that it would reimburse an employee for. This appears to be such a case. So far as the record reveals, the defendant fired the plaintiff not because her husband was disabled but because his medical expenses— which might not have been any lower had they been due to a condition that did not meet the statutory definition of a disability — were costing the defendant an amount of money that it was unwilling to spend. All the evidence recited in the majority opinion concerns costs (“cutting costs,” “high cost of Anthony’s medical treatment,” “financial albatross,” etc.) that a person who had a nondisabling medical condition could equally incur. If cost was indeed, as appears to be the case, the defendant’s only motive for the action complained of, the defendant was not guilty of disability discrimination. Christian v. St. Anthony Medical Center, Inc., 117 F.3d 1051,1052-53 (7th Cir.1997). But it has never made this argument, and so reversal is proper. Since, however, a defendant does not have to file a motion for summary judgment at all, 11 Moore’s Federal Practice, § 56.32[1], at pp. 56-260 to 56-261 (3d ed.1997); Fed.R.Civ.P. 56(b), and if he does file one doesn’t have to include all his arguments in it, Smith v. Richert, 35 F.3d 300, 305 (7th Cir.1994), the defendant will be able to argue the cost point on remand unless the district judge finds that it has been forfeited by being withheld for so long. The majority opinion in this court need not be an obstacle to the defendant’s making the argument. In not remarking the distinction between disability discrimination and expense discrimination, the opinion merely accepts the parties’ framing of the issues.