dissenting.
I respectfully dissent. The majority opinion artificially manipulates the Equal Protection Clause requirement of “similarly situated” and misapplies Fourteenth Amendment standards at the expense of female inmates incarcerated at NCW.
The majority opinion errs in concluding that male inmates incarcerated at NSP are not similarly situated to female inmates incarcerated at NCW for purposes of comparing institutional programs and services.1 On a general level, the female inmates at NCW and male inmates at NSP are similarly situated for purposes of comparing the programs and services offered to them because both groups of inmates are confined by the State of Nebraska as a result of a criminal conviction. More importantly, they are similarly situated because, the State of Nebraska and the Department of Corrections view the purpose of incarceration to be the same for all inmates regardless of gender: to enable inmates to seek reform within the institution and rehabilitate themselves to function in society as law-abiding citizens. Irrespective gender, the prison system strives to prevent revolving-door recidivism. A close and *735critical nexus exists between inmate rehabilitation and the provision of recreational opportunities and educational and vocational training programs and services to inmates within the prison system. It is unquestionable, with respect to the goal of rehabilitation, that female inmates at NCW and male inmates at NSP are similarly situated for purposes of comparing programs and services.
Female inmates at NCW are also similarly situated to male inmates at NSP for purposes of programs and services because they are equally as capable of participating in, and benefitting from, the recreational, educational and vocational opportunities offered to their male counterparts. See More v. Farrier, 984 F.2d 269, 271 (8th Cir.1992) (concluding wheelchair-bound inmates were similarly situated to ambulatory inmates for purposes of challenging denial of in-eell cable television because they were as capable of participating in, and benefitting from, the challenged program). The female inmates incarcerated at NCW are for the most part poor, undereducated, and lack the vocational training necessary to become self-supporting. The female inmates’ gender places them at the bottom of the list of the unemployed and unemploya-bles in this country. In other words, gender itself will disadvantage the female inmates as they attempt to enter or re-enter the workforce. Clearly, their need for educational and vocational training to obtain employment and become self-sufficient upon release is equal to that of males at NSP, who are also in general poor, undereducated, and lack vocational training. All released ex-offenders, irrespective of gender, encounter enormous difficulty in securing employment. Plaintiffs are entitled to the same opportunity as their male counterparts to learn marketable skills which will enable them to obtain employment upon release from prison.
Further, the district court found that NCW and NSP house inmates with comparable custody classifications:
[S]inee NCW spans the security continuum and houses maximum custody inmates, and since NSP also houses maximum custody inmates, it is likely that the court will be comparing populations, aside from gender, which are classified as roughly comparable by the State of Nebraska. In this regard, since there is no segregation between minimum custody inmates and maximum custody inmates at NCW, minimum custody inmates receive the same type of treatment as maximum custody inmates. Thus, it is not inconsistent to treat both institutions as having roughly comparable populations. Stated another way, female minimum custody inmates are treated essentially the same as female maximum custody inmates. Therefore, if it is appropriate to compare institutions housing female maximum custody inmates with institutions housing male maximum custody inmates, then the existence of the lower custody status inmates at NCW is irrelevant.
Klinger v. Nebraska Dep’t of Corrections, 824 F.Supp. 1374, 1389 (D.Neb.1993) (Klinger).
Nonetheless, the majority opinion holds that “NCW inmates and NSP inmates are not similarly situated for purposes of prison programs and services” because of different characteristics of male and female inmates and different circumstances surrounding their incarceration. Maj. op. at 731-32. The reasons cited by the majority opinion for the purported dissimilarity of male and female inmates include factors such as the shorter average length of inmate stay at NCW (13 months) as compared to NSP (54 months). A close examination of the record belies using length of inmate stay as a ground for differentiating between NCW and NSP for purposes of comparing programs and services. Regardless of length of stay, male inmates at NSP, unlike their female counterparts at NCW, are afforded the opportunity to participate in a wide gamut of vocational programs and enroll in post-secondary educational courses that can ultimately lead to a college degree. As the district court concluded, “it is difficult to understand the rationale for denying females [these opportunities] based upon a factor that was not employed for males.” Klinger, 824 F.Supp. at 1401. No evidence was presented that NSP provided the superior programs and services only to “long-time” male inmates or that, conversely, NSP’s “short-time” male inmates did not benefit from the training. The *736“length-of-stay” argument is further belief by the district court’s observation that:
the difference in “length of stay” between NCW inmates and “medium” custody NSP inmates was small. The “medium” custody male group at NSP was probably “most like” the females at NCW. For example, in June, 1990, (Ex. 1385), the data showed: (a) inmates staying 24 months or less: NSP (medium) 70.7% vs. NCW 82.4%; (b) inmates staying 18 months or less: NSP (medium) 61.5% vs. NCW 74.3%; (c) inmates staying 12 months or less: NSP (medium) 50.6% vs. NCW 66.9%. Even if one looks at the maximum custody inmates at NSP, one out of five inmates stayed 12 months or less. Id. Indeed, nearly 40% of the NSP maximum custody inmates stayed 24 months or less. Since college course work was offered to these male inmates regardless of “length of stay,” it seems fair to conclude that “length of stay” was not a factor in offering college work at NSP.
Id.
The majority opinion also finds dissimilarity because “[ajmong Nebraska’s five single-sex prisons, NCW houses the fewest inmates ... NSP, in contrast, houses about six times as many inmates as NCW.” Maj. op. at 731. The majority opinion’s analysis is driven by the fact that the Department is operating under “the restrictions of a limited budget ... and the difference ... [in] institution size” of NCW and NSP make them dissimilar for purposes of comparing programs and services. Maj. op. at 731-32.
In Bukhari v. Hutto, 487 F.Supp. 1162 (E.D.Va.1980) {Bukhari), the plaintiff argued that her conditions of confinement in an all-female prison were inferior to those of male inmates in an all-male institution. The court recognized that the defendants “must deal with the fiscal reality that providing for a wide range of programs for a smaller number of prisoners entails a greater cost.” Id. at 1172. Nonetheless, the court opined that “such seemingly practical considerations may not be used to ‘justify official inaction or unwillingness to operate a prison system in a constitutional manner.’ ” Id. (citations omitted); see also Califano v. Goldfarb, 430 U.S. 199, 217, 97 S.Ct. 1021, 1032, 51 L.Ed.2d 270 (1977) (savings in time, money and effort do not justify gender-based discrimination).
In West v. Virginia Dep’t of Corrections, 847 F.Supp. 402 (W.D.Va.1994) (West), the plaintiff, a female inmate at the Virginia Correctional Center for Women, raised an equal protection clause challenge to the establishment of a pilot boot camp program, which offered academic education, vocational assessment and life skills training, at a male-only prison, while no such pilot program existed at the female-only prison. In response to the defendants’ argument that the program in the men’s prison was justified because of limited resources, and because problems were more pressing in male prisons, and it was more cost-effective to address those problems, the court held “[i]f defendants’ argument were carried to its logical extension, then the same argument could be used to deny women inmates the opportunity for education, vocational training or rehabilitation. Surely such an inequitable distribution of resources is not contemplated by the Fourteenth Amendment.” Id. at 407.
It is well-settled that financial hardship is not a defense to sex discrimination in prisons. See Shapiro v. Thompson, 394 U.S. 618, 633, 89 S.Ct. 1322, 1330-31, 22 L.Ed.2d 600 (1969) (while states have a valid interest in preserving the fiscal integrity of programs, it may not accomplish such a purpose by invidious distinctions between classes of citizens). Cost-driven differences in institution size and inmate length of stay do not constitutionally justify limiting NCW inmates’ training to domestics and other “women’s” occupations such as “clerical arts” and homemaking, providing inferior inmate pay, pre-employment training and prison law libraries, while denying them post-secondary education or vocational courses that could lead to a college degree or college credit, and providing NSP inmates with training in higher-paying, more technical, and more varied occupations, more sophisticated and accessible prison law libraries, and educational and vocational courses leading to attainment of a college degree or college credit.
“To be ‘similarly situated,’ groups need not be identical in makeup, they need only share *737commonalities that merit similar treatment.” Betts v. McCaughtry, 827 F.Supp. 1400, 1405 (W.D.Wis.1993). In the present case, female inmates at NCW do in fact “share commonalities” with male inmates at NSP that merit similar treatment in terms of their right and ability to receive educational and vocational programs and services. NCW inmates are equally as capable of participating in, and benefitting from, the academic, vocational and rehabilitative opportunities offered to the NSP inmates. NCW and NSP inmates are both incarcerated as the result of a criminal conviction, and the Department views the purpose of incarceration for both classes of inmates to be the same regardless of gender. Further, NCW and NSP reflect comparable custody classifications, and the record reveals absolutely no evidence that concerns for institutional security justify offering different programs and services at the respective institutions. “On the contrary, if anything, the relative lack of security concerns at NCW, compared with NSP, makes at least some of the differences in programs inexplicable if based upon security concerns.” Klinger, 824 F.Supp. at 1390. Nonetheless, due to their smaller numbers, predominantly non-violent, non-predatory nature, and the fact that they are generally incarcerated for less serious offenses and for shorter time periods than their male counterparts, the majority opinion chooses to sanction constitutionally unequal treatment of female inmates at NCW and to neglect the rights guaranteed to them by the Equal Protection Clause of the Fourteenth Amendment. The fact of incarceration and the particular exigencies of prison administration cannot justify programs and services provided female inmates that are markedly inferior to those provided male inmates.
The majority opinion further confuses basic tenets of constitutional law in holding fatal to the plaintiffs’ claim of gender discrimination the district court’s failure to make “findings on whether the plaintiffs carried their burden of proving intentional discrimination” and failure to “mention that the plaintiffs carried this burden.” Maj. op. at 734. Along this same vein, the majority opinion mistakenly cites Personnel Adm’r of Mass. v. Feeney, 442 U.S. 256, 274, 99 S.Ct. 2282, 2293-94, 60 L.Ed.2d 870 (1979) (Feeney), to support its holding.
Equal protection clause analysis distinguishes between facial classifications and non-facial classifications. Non-facial classifications, such as the veteran-preference policy at issue in Feeney, are subject to heightened scrutiny only upon proof that they have a disparate impact on a protected group and are motivated by a discriminatory intent. Id. at 272-74, 99 S.Ct. at 2292-94. By contrast, overt gender classifications of similarly situated people, such as the gender-based prison policy in the present case, are subjected to the heightened scrutiny set forth in Craig v. Boren, 429 U.S. 190, 97 S.Ct. 451, 50 L.Ed.2d 397 (1976), under which the government bears the burden of establishing that the gender classification has an important purpose and that the relationship between purpose and classification is substantial. Intentional discrimination can be presumed where the government fails to meet this burden. In Feeney, the Supreme Court opined that purposeful discrimination is “ ‘the condition that offends the Constitution,’ ” 442 U.S. at 274, 99 S.Ct. at 2293 (citations omitted), at least where a covert or overt gender-based classification is not in issue. In other words, “purposeful discrimination is the subject of the inquiry absent a discernible, non-neutral legislative or administrative classification subject to the important governmental interest/substantial relation test.” Marshall v. Kirkland, 602 F.2d 1282, 1299 (8th Cir.1979) (emphasis added). Cases involving an overt or covert gender-based prison policy do not require explicit findings of intentional or purposeful discrimination, but rather are correctly analyzed by employing the important governmental interest/substantial relation test to implicitly establish discriminatory intent. See, e.g., West, 847 F.Supp. at 405-07; McCoy v. Nevada Dep’t of Prisons, 776 F.Supp. 521, 523 (D.Nev.1991); Dawson v. Kendrick, 527 F.Supp. 1252, 1372 (S.D.W.Va.1981); Bukhari, 487 F.Supp. at 1171.2 The *738existence of intentional discrimination is intertwined with the heightened scrutiny framework. “[E]vidence that invidious discrimination — the ultimate object of heightened scrutiny — produced or underlies the classification at hand would doom the [contested] prisoner policy [by] demonstrating its lack of direct and substantial relation to an important governmental interest.” Pitts v. Thornburgh, 866 F.2d 1450, 1459 (D.C.Cir.1989) (emphasis added).
The majority opinion correctly notes that the female inmates of NCW do not challenge the classification by “Nebraska prisons in that male and female inmates are segregated by institution.” Maj. op. at 734. However, the majority opinion then erroneously concludes that, because the gender-based segregation of inmates is not directly challenged, a facially neutral policy is at issue, and Feeney and its requirement of express findings of discriminatory intent must control.
Plaintiffs, as noted above, do not challenge the policy of incarcerating men and women in separate institutions. They do assert, however, “that the policy of housing men and women in separate institutions has been applied in such a manner as to relegate women to an inferior opportunity to take advantage of academic, vocational, and training opportunities.” Canterino v. Wilson, 546 F.Supp. 174, 211 (W.D.Ky.1982), as amended, 562 F.Supp. 106, (1983) aff'd, 875 F.2d 862 (6th Cir.), cert. denied, 493 U.S. 991, 110 S.Ct. 539, 107 L.Ed.2d 536 (1989). To state otherwise, the consequence of the gender-based prison segregation system in Nebraska is a gross inferiority in a host of opportunities for female inmates at NCW as compared to males at NSP. As such, Nebraska must show that the disparate treatment of female inmates is substantially related to an important governmental interest. ‘Where the actual consequence of a seemingly harmless classification reveals disparate treatment, there is ample justification to treat the classification and its consequence together in order to determine the ‘fit’ between this end result and the [governmental interest].” Id. 546 F.Supp. at 211 (quoting Glover v. Johnson, 478 F.Supp. 1075, 1080 (E.D.Mich.1979) (Glover)).
In the present ease, the district court astutely concluded that a facially gender-based policy existed:
[T]here is no doubt that women are incarcerated at NCW because of their gender and not for some other reason. There is only one prison for women in the State of Nebraska. Whether a woman is convicted of writing bad checks or of murder, she will find herself at NCW when sentenced to prison. Men, on the other hand, could be sent to a variety of institutions depending upon a variety of penological goals— from security to programming. As a consequence, despite the fact that nongender-related reasons may exist to justify treating female prisoners differently from male prisoners, the fact remains that female prisoners, as a result of their gender alone, can receive only the programs available at NCW. Therefore, if the programming at NCW is comparatively inferior to programs at NSP, the female recipients may receive poor programming because of their gender and not for some nongender-related reason.
Klinger, 824 F.Supp. at 1388 (emphasis added); accord Bukhari, 487 F.Supp. at 1171 (sex-based, disparate treatment analysis applied to plaintiffs claim, inter alia, that the limited opportunity for education for female maximum security prisoners at all-female prison compared to those opportunities for male maximum security prisoners at major male prisons violated equal protection clause); Batton v. North Carolina, 501 F.Supp. 1173, 1176 (E.D.N.C.1980) (where female inmates were incarcerated at all-female institution, rather than any other prison operated by the Department, solely because of their gender, court applied facial-classification analysis to female inmates’ equal protection claim that they allegedly received vocational training opportunities only in low-paying, traditionally female jobs while male in*739mates were afforded training in a wide variety of occupations); Glover, 478 F.Supp. at 1080 (given the inferior programs offered at female prison, the practical effect of housing women separately is to deny them access to the range and quality of programs offered male inmates based solely on their gender).3
In the present case, the district court correctly reasoned that the offering of educational, vocational, and recreational programs only in the men’s prison, without offering equivalent programs in the women’s prison, constituted a denial of benefits of a program because of sex. In a voluminous and comprehensive opinion, the district court weighed and analyzed in great detail the abundance of evidence which showed that the gender-based allocation of programs and services sprang from indifference, bias, and stereotypical notions of women’s role in society. Contrary to the majority opinion’s assertion, the district court did not assume any inferior program or service at NOW was the product of intentional discrimination. Rather, the framework employed for determining whether an Equal Protection Clause violation existed was to determine
whether there are programs (means) afforded (in quantity or quality) to male inmates at NSP but not female inmates at NCW which place a “substantial burden” on female inmates (disparity); if so, ascertain whether there is an important governmental objective (objective) which underlies the “means”; if so, finally decide whether the “means” chosen to further the governmental “objective” are directly and substantially related to the “objective” (relationship of means to objective) such that the “disparity” may be overlooked.
Klinger, 824 F.Supp. at 1391.
In short, the district court correctly considered whether or not the facially gender-based allocation of programs and services constituted unconstitutional discrimination as defined by such leading eases as Craig v. Boren and Mississippi Univ. for Women v. Hogan, 458 U.S. 718, 102 S.Ct. 3331, 73 L.Ed.2d 1090 (1981) (Hogan). By analyzing whether or not the Department of Corrections’ unequal allocation of programs and services responded substantially to an important governmental interest, the district couit completed its task of determining whether or not invidious discrimination existed. The district court did not impermissibly shift “the burden to the defendants to justify differences in prison programming on sex-neutral grounds and thus disprove discriminatory purpose,” as argued by the majority opinion, Maj. op. at 734, but rather, the district court correctly required the defendants to “carry the burden of showing an ‘exceedingly persuasive justification’ for the classification.” Hogan, 458 U.S. at 724, 102 S.Ct. at 3336 (citing Kirchberg v. Feenstra, 450 U.S. 455, 461, 101 S.Ct. 1195, 1199, 67 L.Ed.2d 428 (1981); Feeney, 442 U.S. at 273, 99 S.Ct. at 2293). As noted above, the district court carefully reviewed the defendants’ evidence and found that the defendants failed to show any exceedingly persuasive justification for the sex-based difference in treatment.
I also agree with the district court’s well-reasoned analysis regarding the remaining issues presented on appeal, and, accordingly, I would affirm the judgment of the district court in all respects.
. The challenged programs and services encompass employment and economic concerns, educational and vocational training courses, concerns regarding the law library, medical and dental services, recreational opportunities, and concerns regarding visitation.
. Eighth Circuit analysis is in accord. See, e.g., Navedo v. Preisser, 630 F.2d 636, 641 (8th Cir.1980) (state statute which prohibited males over twenty-five years old from engaging in sexual intercourse with females sixteen years old, without punishing female over twenty-five years of *738age for engaging in sexual intercourse with males sixteen years old violated equal protection clause "[b]ecause the state has failed to show that its gender-based classification substantially furthers the [governmental objectives of] prevention of physical injury, emotional trauma, or pregnancy caused by sexual intercourse with an older person”).
. As observed by the court in Glover:
[A] female in the State of Michigan will be sent to Huron Valley by reason of her gender alone, and will necessarily have access only to these programs currently available at that location. A male prisoner, on the other hand, can be classified or later transferred to a wide variety of prison facilities in the State and will have access to more program opportunities than his female counterpart. I conclude, therefore, that because of these limitations women as a group are treated differently them men as a group, and that these differences in treatment are directly related to gender.
Glover, 478 F.Supp. at 1078.