Mrintunjoy Sengupta, an East Indian whose skin pigment is black, was employed as a senior engineer by the Morrison-Knudsen Co. (M-K). In early 1982, M-K experienced economic difficulties that necessitated significant reductions in its work force. Sengupta was discharged. Thereafter he brought suit, claiming that his discharge violated Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e et seq.; 42 U.S.C. § 1981 ; the Idaho Human Rights Act; and Idaho contract law. The district court granted summary judgment for M-K on all of Sengupta’s claims. We affirm.

FACTS AND PROCEDURAL BACKGROUND

Sengupta worked for M-K as a junior engineer from 1975 to 1977. He voluntarily left M-K in 1977 to obtain a master’s degree in industrial management engineering. In 1981, Sengupta was rehired as a senior engineer and assigned to the Underground Team of the Chevron Shale Oil Project. Sengupta’s immediate supervisor was Ian Dorling, the Underground Team leader. The project engineer and project manager were Brent Hurst and Dick Grass, respectively. Dorling, who was most familiar with Sengupta’s work performance, was responsible for preparing Sengupta’s job evaluation at the conclusion of the Chevron project. Hurst and Grass, who also had contact with Sengupta, participated in reviewing Sengupta’s work.

Sengupta worked on the Chevron project from February through August, 1981. Dorling and the other two supervisors were displeased with Sengupta’s work during this time, and Dorling prepared a highly critical evaluation of Sengupta. The evaluation form used by Dorling identified eleven areas of job performance for evaluation and directed the evaluator to rate the employee with a score of one to six in each area. This form was used for all Mining Group employees, of which Sengupta was one, and, at the time of his evaluation, it was used exclusively for determining salary increases. Some of the areas in which Sengupta received low marks were creativity, leadership and verbal communication.

In early 1982, the shale oil industry collapsed, forcing M-K to make layoffs. The Mining Group is one of several groups within M-K operating independently of each other; it is divided into districts that, in turn, are subdivided into departments. The Mining Group Personnel Department, which reports directly to Group level management, devised the layoff procedures that were implemented in January, 1982. Sengupta’s department, Department 222, had no involvement in the development of this procedure. Department 222’s participation in the layoff process was limited to preparing the performance evaluations and selecting the number of employees to be terminated.

Pursuant to the layoff plan, all Mining Group employees were ranked according to their most recent performance evaluation. Sengupta’s ranking was based on Dorling’s evaluation. The employees with the lowest scores were discharged, according to the needs of each department. The head of Department 222 decided five people should be laid off.

In 1982, the Mining Group began with 281 employees, of which 248 were non-minority workers and 33 were minority workers. Of the former, 68 were discharged, while of the latter the number was 9. Or, expressed in percentages, 27 percent of the nonminority Mining Group employees and 27 percent of the minority Mining Group employees were laid off. However, within Department 222 the pattern was different. From a total of 28 employees, 4 of the 5 employees selected for layoff were black, including Sengupta.

Sengupta made a Title VII disparate impact as well as a discriminatory treatment claim. In 1984, the district court entered summary judgment in favor of M-K on Sengupta’s disparate impact claim but reserved its ruling on his discriminatory treatment claim until Sengupta had an opportunity to conduct further discovery and thoroughly brief his position. In 1985, the district court reaffirmed its ruling on Sengupta’s disparate impact claim and entered summary judgment for M-K on the plaintiff's treatment and state law claims. Sengupta brought this appeal.

II.

STANDARD OF REVIEW

We review the district court’s grant of summary judgment de novo to determine whether there are any genuine issues of material fact and whether the district court correctly applied the relevant substantive law. Ashton v. Cory, 780 F.2d 816, 818 (9th Cir.1986).

III.

DISCRIMINATORY TREATMENT

In order to succeed on his section 1981 claim, Sengupta must prove intentional discrimination. General Bldg. Contractors Ass’n v. Pennsylvania, 458 U.S. 375, 389 , 102 S.Ct. 3141, 3149 , 73 L.Ed.2d 835 (1982). But of the two Title VII theories of liability, only his treatment claim requires proof of discriminatory intent. International Bhd. of Teamsters v. United States, 431 U.S. 324 , 335 n. 15, 97 S.Ct. 1843 , 1854 n. 15, 52 L.Ed.2d 396 (1977). A disparate impact claim does not depend upon proof of a discriminatory motive. Id. at 336 n. 15, 97 S.Ct. at 1854 n. 15.

The structure of the pleading and proof of a discriminatory treatment claim is familiar ground. The plaintiff “has the initial burden of establishing a prima facie case that his employer discriminated against him on account of his race, color, religion, sex, or national origin.” Cooper v. Federal Reserve Bank, 467 U.S. 867, 874 , 104 S.Ct. 2794, 2799 , 81 L.Ed.2d 718 (1984). To establish a prima facie case of racial discrimination, the plaintiff has the burden of proving by a preponderance of the evidence that he was discharged “under circumstances which give rise to an inference of unlawful discrimination.” Texas Dep’t of Community Affairs v. Burdine, 450 U.S. 248, 253 , 101 S.Ct. 1089, 1094 , 67 L.Ed.2d 207 (1981).

In McDonnell Douglas Corp. v. Green, 411 U.S. 792, 802 , 93 S.Ct. 1817, 1824 , 36 L.Ed.2d 668 (1973), the Supreme Court established a framework within which a Title VII plaintiff can use circumstantial evidence to establish a prima facie case of racial discrimination in a hiring context:

The McDonnell Douglas standards are adaptable to different factual situations. Id. at 802 n. 13, 93 S.Ct. at 1824 n. 13. In the discriminatory discharge context, the plaintiff must show (i) that he was within the protected class; (ii) “that he was doing his job well enough to rule out the possibility that he was fired for inadequate job performance”; and (iii) “that his employer sought a replacement with qualifications similar to his own, thus demonstrating a continued need for the same services and skills.” Loeb v. Textron, Inc., 600 F.2d 1003, 1013 (1st Cir.1979); see Potter v. Goodwill Indus., 518 F.2d 864, 865 (6th Cir. 1975).

We hold that Sengupta failed to establish a prima facie case of discriminatory treatment under the McDonnell Douglas approach. M-K did not attempt to fill Sengupta’s position following the layoff. McDonnell Douglas does not proscribe discharges because of an absolute or relative performance failure or a reduction in the labor force. See Gay v. Waiters’ and Dairy Lunchmen’s Union, Local No. 30, 694 F.2d 531, 547 (9th Cir.1982). When neither of these reasons justifies the discharge, “courts infer, in light of their experience, that the employment decision was ‘more likely than not’ based upon race.” Id. (quoting Fumco Constr. Corp. v. Waters, 438 U.S. 567, 577 , 98 S.Ct. 2943, 2950 , 57 L.Ed.2d 957 (1978)).

Sengupta’s failure to prove his claim rests on the fact that M-K no longer needed his services because of its economic difficulties. Without the demonstration of a need for the same services and skills that Sengupta possessed, the conceptual underpinnings of McDonnell Douglas crumble.

It is true, of course, that we have said that a prima facie case of disparate treatment may be established “with a combination of direct, circumstantial and statistical evidence of discrimination.” Gay, 694 F.2d at 553 . However, there is no direct evidence of racial discrimination in this case. Sengupta argues that the use of subjective criteria in the evaluation process and the fact that his supervisor, Dorling, gave another black person, Kendrick Lee, a negative evaluation are evidence of a discriminatory intent. This is not so. Dorling’s evaluation of Lee was below average, but it was more complimentary than that of Sengupta. In addition, Lee found his evaluation to be satisfactory.

Moreover, the use of subjective employment criteria is not unlawful per se. Gay, 694 F.2d at 554 . Although subjective practices may provide “ready mechanisms for discrimination,” Smith v. Union Oil Co., 17 Fair Empl.Prac.Cas. (BNA) 960 , 991 (N.D.Cal.1977), their relevance to proof of a discriminatory intent is weak. Honorable employers frequently use subjective criteria also. Indeed, in many situations they are indispensable to the process of selection in which employers must engage. In any event, after nearly two years of discovery Sengupta has been unable to amass any evidence indicating that Dorling, or any other M-K supervisor, used the subjective evaluation to mask unlawful and pretextual discrimination.

Sengupta primarily relies upon statistical evidence to thrust upon M-K the burden of going forward with evidence to show the non-discriminatory reasons for his discharge. He asserts that discriminatory intent can be inferred from the fact that four of the five persons laid off from Department 222 were black. Although statistics have a place in a disparate treatment case, see Diaz v. American Tel. & Tel., 752 F.2d 1356, 1363 (9th Cir.1985), their utility “depends on all of the surrounding facts and circumstances,” International Bhd. of Teamsters, 431 U.S. at 340 , 97 S.Ct. at 1857 . Sengupta’s ability to prove discriminatory intent based on statistical evidence depends upon selecting the proper labor pool. The issue is whether that pool is the entire Mining Group or Department 222.

First, Department 222 is too small. It had only 28 employees at the time of Sengupta’s discharge. In Monta v. Southern California Permanente Medical Group, 541 F.2d 217 (9th Cir.1976), cert. denied, 429 U.S. 1050 , 97 S.Ct. 761 , 50 L.Ed.2d 765 (1977), this court stated that “ ‘statistical evidence derived from an extremely small universe’ ... ‘has little predictive value and must be disregarded.’ ” Id. at 220 (quoting Harper v. Trans World Airlines, Inc., 525 F.2d 409, 412 (8th Cir.1975)); see also International Bhd. of Teamsters, 431 U.S. at 340 n. 20, 97 S.Ct. at 1856 n. 20 (considerations such as small sample size may detract from the value of statistical evidence). The problem with small labor pools is that slight changes in the data can drastically alter appearances. Consequently, in 1979, the EEOC concluded that statistics from small applicant pools such as a pool of 30 persons are not dispositive. 44 Fed.Reg. 11,999 (1979).

By necessity, courts sometimes must rely on statistics derived from small sample groups. Not to do so would deny employees in small companies some of the protections that Title VII provides. M-K is not, however, a small company; nor is the Mining Group a small group. There exists no necessity to employ a small group.

Even if the data from the small Department , 222 were statistically relevant, reliance on it would be inappropriate. The impact of a practice on the protected class should generally be measured against the actual pool of employees affected by that practice. See Moore v. Hughes Helicopters, Inc., 708 F.2d 475, 482 (9th Cir.1983). In discriminatory hiring cases this principle is occasionally suspended to permit the examination of the pool of potential applicants rather than actual applicants. An example is when the challenged employment practice is an entrance requirement that discourages individuals from applying for the position in question. See Dothard v. Rawlinson, 433 U.S. 321, 330 , 97 S.Ct. 2720, 2727 , 53 L.Ed.2d 786 (1977) (height and weight specifications may have discouraged females from applying for jobs as prison guards). Here we have the data for the entire Mining Group. We need no external subgroup.

We return, therefore, to the basic data. Within the Mining Group, the relevant group, 27 percent of the minority employees and 27 percent of the non-minority employees were laid off. Accordingly, there is no evidence of a discriminatory pattern. We find that Sengupta’s statistical and circumstantial evidence, properly viewed together, fails, as a matter of law, to establish a prima facie case of disparate treatment.

IV.

DISPARATE IMPACT

Sengupta, as already pointed out, asserts a cause of action under the disparate impact theory of Title VII. To establish a prima facie discriminatory impact case, the plaintiff does not need to show discriminatory intent. International Bhd. of Teamsters, 431 U.S. at 336 n. 15, 97 S.Ct. at 1854 n. 15. What he must show is that a facially neutral employment practice has a “significantly discriminatory” impact upon a protected class. Connecticut v. Teal, 457 U.S. 440, 446 , 102 S.Ct. 2525, 2530 , 73 L.Ed.2d 130 (1982). If the plaintiff makes this initial showing, the burden of proof then shifts to the defendant to show that the employment practice has “a manifest relationship to the employment in question,” Griggs v. Duke Power Co., 401 U.S. 424, 432 , 91 S.Ct. 849, 854 , 28 L.Ed.2d 158 (1971), or that the practice is justified by “business necessity,” Gay, 694 F.2d at 537 (citing Contreras v. City of Los Angeles, 656 F.2d 1267, 1275-80 (9th Cir.1981), cert. denied, 455 U.S. 1021 , 102 S.Ct. 1719 , 72 L.Ed.2d 140 (1982)). The employer, of course, can “rebut the employee’s prima facie case by showing the inaccuracy of the employee’s statistics.” Moore v. Hughes Helicopters, Inc., 708 F.2d at 481 .

Sengupta argues that the M-K layoff procedure, as implemented within Depart ment 222, was a facially neutral process that had an adverse impact on black employees. The flaw in Sengupta’s analysis is clear. When the relevant statistical pool, the entire Mining Group, is considered, there is no disparate impact. The challenged discharge procedures resulted in an equal percentage of minority and non-minority employees being laid off.

Sengupta argues that the Supreme Court’s decision in Connecticut v. Teal, 457 U.S. 440 , 102 S.Ct. 2525 , 73 L.Ed.2d 130 (1982), precludes use of the layoff statistics for the Mining Group. According to Sengupta the defendant’s “bottom line” statistics for the Mining Group do not insulate M-K from a successful disparate impact claim because Teal requires that employment opportunities be equal for minorities and non-minorities.

In Teal , the Supreme Court analyzed a two-tier promotion system that achieved a statistically nondiscriminatory result. The first step in the process, a written examination, had a disparate impact on black candidates. However, a nondiscriminatory result was achieved in the second step when a disproportionate number of blacks who had passed the written examination were promoted. The Court held that the nondiscriminatory result did not defeat the plaintiff’s prima facie case of disparate impact with respect to the written exam. Id. at 454 , 102 S.Ct. at 2534. The Court observed that Title VII was designed to protect employment opportunities for individuals, id. at 451, 102 S.Ct. at 2532, and racially neutral statistics would provide little comfort for those black applicants who had been screened out by the initial examination.

Teal is inapplicable to this case. Sengupta was not discharged pursuant to a two-tier screening process in which the disparate impact under the first tier was compensated for under the second tier. The Personnel Department’s layoff procedure consisted of only one step and it was applied to all Mining Group employees. There is no evidence, other than the flawed statistical evidence that Sengupta attempts to . use, that Department 222 was treated more harshly than any other department. When all is said and done, the employer, M-K, treated the Mining Group as a unit and Sengupta has given us no reason to treat this as a mask concealing discriminatory evil. To hold otherwise would permit the employee who alleges disparate impact to select the group, large or small, that most convincingly supports his claims.

STATE LAW CLAIMS

Sengupta asserts that when M-K discharged him in 1982, it breached an implied covenant of good faith and fairness. Under Idaho law the general rule is that an employer may discharge an employee for any reason without incurring liability unless the employee is hired pursuant to an employment contract that specifies the duration of employment or expressly limits the reasons for which an employee may be discharged. Watson v. Idaho Falls Con-sol. Hosps., Inc., Ill Idaho 44, 720 P.2d 632, 635 (1986); MacNeil v. Minidoka Memorial Hosp., 108 Idaho 588, 589 , 701 P.2d 208, 209 (1985). Idaho recognizes one exception to the above rule: an employer may be liable for wrongful discharge if the motivation for the discharge contravenes public policy. Watson, 720 P.2d at 635 . It is undisputed that Sengupta had no contract and was an employee at will. Additionally, Sengupta has failed to demonstrate that his discharge contravenes public policy in any way. Consequently, Sengupta’s contract claim must fail.

Sengupta’s final claim is that his discharge violated the Idaho Human Rights Act (IHRA), Idaho Code § 67-5901 et seq. The Idaho Supreme Court has held that the same standards of proof applicable in Title VII cases govern actions under IHRA. Bowles v. Keating, 100 Idaho 808, 812 , 606 P.2d 458, 462 (1979). Thus, if Sengupta’s claim must fail under Title VII, it must also fail under IHRA.

The district court’s grant of summary judgment is affirmed.

AFFIRMED.

. One of the four black employees selected for layoff was eventually retained because of his special skills in rock mechanics. Another black employee from Africa was laid off because his six month work visa was about to expire. . A prima facie case may also be established through the production of direct evidence of discrimination. Douglas v. Anderson, 656 F.2d 528 , 531 n. 2 (9th Cir.1981). In this case, there is no direct evidence of racial discrimination. . Sengupta has alleged that the district court’s grant of summary judgment in favor of M-K denied him his Sixth Amendment right to trial by jury. Initially, since this is not a criminal case, the Sixth Amendment is not implicated. Moreover, the Seventh Amendment has not been violated by the district court’s action. The Constitution only requires that bona fide fact questions be submitted to a jury. DeBry v. Transamerica Corp., 601 F.2d 480, 490 (10th Cir.1979). In its judgment for M-K the district court concluded that there were no genuine issues of material fact at issue. Thus, the plaintiffs only possible claim is that the district court erred in this determination.