dissenting.
The history of sex discrimination in America is replete with arguments based on the supposed fragility of women, and their inability to endure conditions which arise in various employments. The basic tenor of legislation against sex discrimination has been to give a woman, if she chooses, the opportunity to engage in an occupation previously substantially limited to men, however onerous the conditions of that occupation. In this case, the court holds that Title VII protects a woman from working at a job where conditions may be more hazardous to her health than to the health of men. I believe this ruling confuses the dictates of Title VII with the functions of the Occupational Safety and Health Administration, and I therefore dissent.
The court properly states the distinction between disparate treatment and disparate impact analysis, but it begins to go astray by analyzing physical conditions of employment through a disparate impact analysis. The landmark case, Griggs v. Duke Power Co., 401 U.S. 424, 91 S.Ct. 849, 28 L.Ed.2d 158 (1971), properly emphasizes that the theme of disparate impact analysis concerning job qualification is that we must measure the person for the job, not just the person in the abstract. In other words, the most important thing is whether the person can do the job.
The force of this policy has also been the genesis of the “business necessity” test in rebuttal, holding that if a company completely shuts out persons from consideration, the practice must be one “necessary to the safe and efficient operation of the business,” Head v. Timken Roller Bearing Co., 486 F.2d 870, 879 (6th Cir.1973) (quoting Robinson v. Lorillard, 444 F.2d 791, 798 (4th Cir.1971)), or “such as to substantially promote the proficient operation of the business,” Chrisner v. Complete Auto Transit, 645 F.2d 1251, 1262 (6th Cir.1981). This is a stringent test.
This type of disparate impact analysis is almost completely unsuited to examination of conditions. By hypothesis, the conditions being examined are in fact facially neutral. However, it can rarely, if ever, be said that conditions, especially any that would be complained of, are “necessary.” Almost any condition can be improved or ameliorated at a certain cost. I do not believe Title VII empowers us to analyze just how much a condition may reasonably be improved without eliminating the business. Instead, OSHA is specifically charged with determining the safety standards and levels of health risk to be observed. It does not appear that OSHA has any standards that differentiate safe working conditions for women from those for men.
I do not agree with the court that working conditions that apply to all workers can be viewed as a method by which an employ*390er would “limit, segregate, or classify” employees, in the words of § 703(a)(2). An employer could conceivably use working conditions as a way to “limit ... employees ... in a way which would deprive” them of opportunities on account of sex. Thus, in E.E.O.C. v. Ball Corp., 661 F.2d 531 (6th Cir.1981), different working conditions applied to different groups which were heavily differentiated by sex. In such a case, disparate impact analysis may be appropriate because there in fact has been a classification of employees. Here, there is no “limit” on employees, other than in the sense that every reaction to work conditions or requirements may be a limit to some employee. In Ball, employees did have a specific rule that limited them. Here, that is not true. There is also not the slightest evidence in this case that the employer’s toilet practices were based on any intention or desire to discourage women employees.
At page 387 of its opinion, the court argues that the entire rationale of disparate impact would be undone if “apparent equality of facilities could shield an employer from Title VII liability,” because it would overlook differences in effect from a “challenged practice.” But this argument is simply another way of pointing out that if disparate impact analysis is extended to conditions, then every job arrangement is subject to scrutiny under that standard.
A more sensible view is that disparate impact analysis does apply to classifications and differentiation in doing the work of a job. But a working condition applicable to all — by hypothesis people who already are employed and are paid equally if they do the job — is not a limitation or classification of employees. It is simply not a “practice” as that term has been used in the context of § 703(a)(2).
The various components of the complaint here also indicate the unsuitability of “business necessity” analysis. Thus, plaintiff complains of at least the following:
1. the unsanitary and soiled conditions of the facilities generally;
2. the absence of paper seat covers;
3. the absence of unsoiled toilet paper;
4. the absence of sanitary napkin dispensers; and
5. the absence of running water.
It is obvious that these items fall on a continuum of cost, trouble, and prevalence in the industry. The general condition of the toilets was contrary to the contract, and represented a failure of the contractor, and of TVA’s supervision. The absence of paper covers was the same, though it appears to have been more episodic. The unsuitability of the toilet paper that was furnished may have been the result of the contractor, or of actions by fellow employees. The absence of running water was clearly a choice dictated by cost and conveniences, and an alternative was made available in the form of the waterless hand cleaner, which plaintiff did not use. Sanitary napkin dispensers would clearly be a convenience for women only, which TVA chose not to provide. This wide range of items, and of justifications, also leads me to believe that the court was hasty in holding that a violation occurred, rather than remanding to the district court for consideration of the business necessity on the new basis announced by the court.
At page 388 of its opinion, the court says the decision to use portable toilets is not the issue. However, under the court’s analysis it would seem that an employer who failed to provide toilets at all would be in violation of Title VII, whether or not in violation of OSHA regulations. The consequences developed in this case would surely be the same, or worse. Again, this would be a result likely to have startled the members of Congress who enacted this legislation.
Nor are these contingencies purely hypothetical. There certainly are a reasonable number of outdoor-type jobs, such as firefighting, itinerant construction, or maintenance, where this court’s ruling could have real impact. It would also pose the familiar dilemma for an employer of choosing between incurring higher costs by employing women or risking a discrimination complaint by attempting unobtrusively to exclude them. Our decision today certainly *391makes the latter course more likely than it would be otherwise.
Virtually every strenuous, unusual, or dangerous job contains conditions that may have a differential impact on a sex, age group, or race. In most cases, this disparity could be eliminated by a greater or lesser expenditure of money and effort to provide amenities, redesign jobs, and so on. This type of “reasonable accommodation” analysis is appropriate in cases of discrimination against the handicapped1 under § 504 of the Rehabilitation Act of 1973, 29 U.S.C. § 794 (1982 Ed.), but Title VII does not mandate such a standard for race or sex.
On a relative and statistical basis, women may well be impacted more by dirty toilets than are men. But so too are women working at jobs designed to require heavy lifting, so too are whites working in conditions of extreme heat and sunlight; so too are those of lowland national origin in working at great altitudes, and so too are older people working at jobs requiring great flexibility, strength and speed. If all of these are to be analyzed under the rubric that working conditions are the type of qualifications to which disparate impact analysis applies, and further that only “business necessity” can overcome them, we have imported wholesale into Title VII exactly the type of cost versus risk considerations that are the staple expertise of OSHA.
As has been frequently remarked, the legislative history of Title VII is remarkable for its sparseness. However, I see nothing there, or in the whole movement toward sexual equality in the workplace embodied in Title VII, to enact a requirement that working conditions for all must be upgraded to some unstated standard before women can have full access to the workplace. Quite the contrary, the keynote of that movement has been the removal of barriers that are special to women so that they may then compete on the basis of their ability to do the actual jpb; given its conditions.
I would AFFIRM.
. It is interesting to note that in the recent case of Alexander v. Choate, 469 U.S. 287, 105 S.Ct. 712, 83 L.Ed.2d 661 (1985), the Supreme Court refused to hold that all cases of disparate impact made out even a prima facie case of discrimination, even under the more favorable standards of § 504. Id., 105 S.Ct. at 720.