concurring in part and dissenting in part.
I must respectfully disagree with the majority’s view that the district judge abused his discretion in declining to award retroactive competitive seniority to the EEOC. Moreover, I do not think the district judge erred in his definition of the relevant available labor pool for purposes of determining liability and computing backpay. In all other respects, I concur in the majority’s opinion.
Denial of Retroactive Seniority
I do not agree that the district judge abused his discretion in denying retroactive competitive seniority. “[T]he statutory scheme of Title VII ‘implicitly recognizes that there may be cases calling for one remedy but not another, and * * * these choices are, of course, left in the first instance to the district courts.” Franks v. Bowman Transportation Co., 424 U.S. 747, 779, 96 S.Ct. 1251, 1271, 47 L.Ed.2d 444 (1976).
As the majority notes, Rath’s opposition to full retroactive seniority is based among other concerns on the prospect that long-term employees will be bumped to less desirable jobs. We recognized in Briseno v. Central Technical Community College Area, 739 F.2d 344, 348 (8th Cir.1984), cited by the majority, that the, choice of remedies discussed in Franks permits a district court to limit relief so that innocent incumbent employees will not be displaced. In Moore v. City of San Jose, 615 F.2d 1265, 1272 (9th Cir.1980), also cited by the majority, the Ninth Circuit noted that retroactive seniority in the circumstances of that case would cause no existing employees to lose their jobs, and “the number of returning employees was small enough that their impact on the seniority-based benefits of incumbent employees would be minimal.”
*339Thus, while I am not opposed to an award of noncompetitive “benefit” seniority1 to appropriately identified victims of Rath’s discriminatory policies, I would affirm the district court’s denial of competitive seniority in this ease. In light of the number of persons now rightfully entitled to a place in Rath’s work force as the result of the company’s years of discriminatory hiring, as well as Rath’s current financial status and the certainty that incumbents will be bumped to lower positions, I consider the district court’s adjustment of the remedy to account for incumbent employees appropriate. The retention of a full work force by a company which has experienced “horrendous losses” since 1977 and which is presently in bankruptcy seems to me exceedingly unlikely. See Romasanta v. United Air Lines, Inc., 717 F.2d 1140, 1147-56 (7th Cir.1983), cert. denied, 466 U.S. 944, 104 S.Ct. 1928, 80 L.Ed.2d 474 (1984) (declining to award full, retroactive competitive seniority in light of the adverse impact on a substantial number of incumbent employees and certain economic conditions adversely affecting the defendant company’s potential for growth). Labor Force Statistics
In Green v. Missouri Pacific Railroad Co., 523 F.2d 1290, 1293-94 (8th Cir.1975), we recognized that generally three kinds of statistical comparisons may be used to establish whether a challenged employment practice has a disproportionate impact on a protected group in violation of Title VII. Two of these procedures involve resort to general population figures. The first examines whether “[women] as a class (or at least [women] in a specified geographical area) are excluded by the employment practice in question at a substantially higher rate than [men].” Id. at 1293. Another involves comparing the composition of the employer’s work force with the composition of the population at large. Id. at 1294. The procedure which does not rely on general population data and which the district court used “focuses on a comparison of the percentage of [male and female] job applicants actually excluded by the employment practice * * *.” Id.
The district court considered the EEOC’s evidence in support of its reliance on general population statistics defective:
While the Court believes that general population or similar work force data may be appropriate under certain circumstances, the Court is of the opinion that the failure to break the work force statistics for Louisia [sic] County into the job categories of “durable goods manufacturing” or “non-manufacturing” industries, eliminates any substantive probative value the statistics on the general population may have had. Because of the nature of the jobs offered by the Rath plant, the Court is of the opinion that many persons engaged in jobs in non-manufacturing industries would not be attracted to that type of work, although it may also be questionable whether this type of work would be attractive to many in the durable goods manufacturing. Such statistics would have had more validity than the general population figures.
I agree with the district court’s analysis. I cannot accept the unsubstantiated premise advanced by the EEOC that the entire female component of the Louisa County, Iowa work force would have been interested in or qualified to perform the hog slaughtering and processing jobs at Rath. See New York City Transit Authority v. Beazer, 440 U.S. 568, 586 n. 29, 99 S.Ct. 1355, 1366 n. 29, 59 L.Ed.2d 587 (1979):
*340Although “a statistical showing of disproportionate impact [need not] always be based on an analysis'of the characteristics of actual applicants,” Dothard v. Rawlinson, 433 U.S. 321, 330, [97 S.Ct. 2720, 2727, 53 L.Ed.2d 786], “evidence showing that the figures for the general population might not accurately reflect the pool of qualified job applicants” undermines the significance of such [general population] figures. Teamsters v. United States, supra, at 340 n. 20 [97 S.Ct. at 1856-57 n. 20].
I therefore consider the district court’s use of applicant flow data appropriate.
. “Benefit”-type seniority refers to the use of a worker's earned seniority credits in computing his level of economic "fringe benefits.” Examples of such benefits are pensions, paid vacation time, and unemployment insurance. “Competitive”-type seniority refers to the use of those same earned credits in determining his right, relative to other workers, to job-related "rights” that cannot be supplied equally to any two employees. Examples can range from the worker’s right to keep his job while someone else is laid off, to his right to a place in the punch-out line ahead of another employee at the end of a workday.
Franks v. Bowman Transp. Co., 424 U.S. 747, 782 n. 1, 96 S.Ct. 1251, 1273 n. 1, 47 L.Ed.2d 444 (1976) (Powell, J. concurring in part and dissenting in part).