dissenting.
Instead of deciding this case in accordance with controlling precedent, the Court today delivers a disconcerting lecture about the evils of governmental racial classifications. For its text the Court has selected three propositions, represented by the bywords “skepticism,” “consistency,” and “congruence.” See ante, at 223-224. I shall comment on each of these propositions, then add a few words about stare deci-sis, and finally explain why I believe this Court has a duty to affirm the judgment of the Court of Appeals.
HH
The Court’s concept of skepticism is, at least m principle, a good statement of law and of common sense. Undoubtedly, a court should be wary of a governmental decision that relies upon a racial classification. “Because racial characteristics so seldom provide a relevant basis for disparate treatment, and because classifications based on race are potentially so harmful to the entire body politic,” a reviewing court must satisfy itself that the reasons for any such classification are “clearly identified and unquestionably legitimate.” Fullilove v. Klutznick, 448 U. S. 448, 533-535 (1980) (Stevens, J., dissenting). This principle is explicit in Chief Justice Burger’s opinion, id., at 480; in Justice Powell’s concurrence, id., at 496; and in my dissent in Fullilove, id., at 533-534. I welcome its renewed endorsement by the Court today. But, as the opinions in Fullilove demonstrate, substantial agreement on the standard to be applied in deciding difficult cases does not necessarily lead to agreement on how those cases actually should or will be resolved. In my judgment, because uniform standards are often anything but uniform, we should evaluate the Court’s comments on. “consistency,” “congruence,” and stave decisis with the same type of skepticism that the Court advocates for the underlying issue.
*243II
The Court s concept of consistency assumes that there is no significant difference between a decision by the majority to impose a special burden on the members of a minority race and a decision by the majority to provide a benefit to certain members of that minority notwithstanding its incidental burden on some members of the majority. In my opinion that assumption is untenable. There is no moral or constitutional equivalence between a policy that is designed to perpetuate a caste system and one that seeks to eradicate racial subordination. Invidious discrimination is an engine of oppression, subjugating a disfavored group to enhance or maintain the power of the majority. Remedial race-based preferences reflect the opposite impulse: a desire to foster equality in society. No sensible conception of the Government’s constitutional obligation to “govern impartially,” Hampton v. Mow Sun Wong, 426 U. S. 88, 100 (1976), should ignore this distinction.1
*244To illustrate the point, consider our cases addressing the Federal Government’s discrimination against Japanese-Americans during World War II, Hirabayashi v. United States, 320 U. S. 81 (1943), and Korematsu v. United States, 323 U. S. 214 (1944). The discrimination at issue in those cases was invidious because the Government imposed special burdens — a curfew and exclusion from certain areas on the West Coast2 — on the members of a minority class defined by racial and ethnic characteristics. Members of the same racially defined class exhibited exceptional heroism in the service of our country during that war. Now suppose Congress decided to reward that service with a federal program that gave all Japanese-American veterans an extraordinary preference in Government employment. Cf. Personnel Administrator of Mass. v. Feeney, 442 U. S. 256 (1979). If Congress had done so, the same racial characteristics that motivated the discriminatory burdens in Hirabayashi and Korematsu would have defined the preferred class of veterans. Nevertheless, “consistency” surely would not require us to describe the incidental burden on everyone else in the country as “odious” or “invidious” as those terms were used in those cases. We should reject a concept of “consistency” that would view the special preferences that the National Government has provided to Native Americans since 18343 *245as comparable to the official discrimination against African-Americans that was prevalent for much of our history.
The consistency that the Court espouses would disregard the difference between a “No Trespassing” sign and a welcome mat. It would treat a Dixiecrat Senator’s decision to vote against Thurgood Marshall’s confirmation in order to keep African-Americans off the Supreme Court as on a par with President Johnson’s evaluation of his nominee’s race as a positive factor. It would equate a law that made black citizens ineligible for military service with a program aimed at recruiting black soldiers. An attempt by the majority to exclude members of a minority race from a regulated market is fundamentally different from a subsidy that enables a relatively small group of newcomers to enter that market. An interest in “consistency” does not justify treating differences as though they were similarities.
The Court’s explanation for treating dissimilar race-based decisions as though they were equally objectionable is a supposed inability to differentiate between “invidious” and “benign” discrimination. Ante, at 225-226. But the term “affirmative action” is common and well understood. Its presence in everyday parlance shows that people understand the. difference between good intentions and bad. As with any legal concept, some cases may be difficult to classify,4 but our equal protection jurisprudence has identified a critical difference between state action that imposes burdens on a *246disfavored few and state action that benefits the few “in spite of” its adverse effects on the many. Feeney, 442 U. S., at 279.
Indeed, our jurisprudence has made the standard to be applied in cases of invidious discrimination turn on whether the discrimination is “intentional,” or whether, by contrast, it merely has a discriminatory “effect.” Washington v. Davis, 426 U. S. 229 (1976). Surely this distinction is at least as subtle, and at least as difficult to apply, see id., at 253-254 (concurring opinion), as the usually obvious distinction between a measure intended to benefit members of a particular minority race and a measure intended to burden a minority race. A state actor inclined to subvert the Constitution might easily hide bad intentions in the guise of unintended “effects”; but I should think it far more difficult to enact a law intending to preserve the majority’s hegemony while casting it plausibly in the guise of affirmative action for minorities.
Nothing is inherently wrong with applying a single standard to fundamentally different situations, as long as that standard takes relevant differences into account. For example, if the Court in all equal protection cases were to insist that differential treatment be justified by relevant characteristics of the members of the favored and disfavored classes that provide a legitimate basis for disparate treatment, such a standard would treat dissimilar cases differently while still recognizing that there is, after all, only one Equal Protection Clause. See Cleburne v. Cleburne Living Center, Inc., 473 U. S. 432, 451-455 (1985) (Stevens, J., concurring); San Antonio Independent School Dist. v. Rodriguez, 411 U. S. 1, 98-110 (1973) (Marshall, J., dissenting). Under such a standard, subsidies for disadvantaged businesses may be constitutional though special taxes on such businesses would be invalid. But a single standard that purports to equate remedial preferences with invidious discrimination cannot be defended in the name of “equal protection.”
*247Moreover, the Court may find that its new “consistency” approach to race-based classifications is difficult to square with its insistence upon rigidly separate categories for discrimination against different classes of individuals. For example, as the law currently stands, the Court will apply “intermediate scrutiny” to cases of invidious gender discrimination and “strict scrutiny” to cases of invidious race discrimination, while applying the same standard for benign classifications as for invidious ones. If this remains the law, then today’s lecture about “consistency” will produce the anomalous result that the Government can more easily enact affirmative-action programs to remedy discrimination against women than it can enact affirmative-action programs to remedy discrimination against African-Americans — even though the primary purpose of the Equal Protection Clause was to end discrimination against the former slaves. See Associated General Contractors of Cal., Inc. v. San Francisco, 813 F. 2d 922 (CA9 1987) (striking down racial preference under strict scrutiny while upholding gender preference under intermediate scrutiny). When a court becomes preoccupied with abstract standards, it risks sacrificing common sense at the altar of formal consistency.
As a matter of constitutional and democratic principle, a decision by representatives of the majority to discriminate against the members of a minority race is fundamentally different from those same representatives’ decision to impose incidental costs on the majority of their constituents in order to provide a benefit to a disadvantaged minority.5 Indeed, *248as I have previously argued, the former is virtually always repugnant to the principles of a free and democratic society, whereas the latter is, in some circumstances, entirely consistent with the ideal of equality. Wygant v. Jackson Bd. of Ed., 476 U. S. 267, 316-317 (1986) (Stevens, J., dissenting).6 *249By insisting on a doctrinaire notion of “consistency” in the standard applicable to all race-based governmental actions, the Court obscures this essential dichotomy.
III
The Court s concept of congruence assumes that there is no significant difference between a decision by the Congress of the United States to adopt an affirmative-action program and such a decision by a State or a municipality. In my opinion that assumption is untenable. It ignores important practical and legal differences between federal and state or local decisionmakers.
These differences have been identified repeatedly and consistently both in opinions of the Court and in separate opinions authored by Members of today’s majority. Thus, in Metro Broadcasting, Inc. v. FCC, 497 U. S. 547 (1990), in which we upheld a federal program designed to foster racial diversity in broadcasting, we identified the special “institu*250tional competence” of our National Legislature. Id., at 563. “It is of overriding significance in these cases,” we were careful to emphasize, “that the FCC’s minority ownership programs have been specifically approved — indeed, mandated — by Congress.” Ibid. We recalled the several opinions in Fullilove that admonished this Court to “ ‘approach our task with appropriate deference to the Congress, a coequal branch charged by the Constitution with the power to “provide for the . . . general Welfare of the United States” and “to enforce, by appropriate legislation,” the equal protection guarantees of the Fourteenth Amendment.’ [Fullilove, 448 U. S.], at 472; see also id., at 491; id., at 510, and 515-516, n. 14 (Powell, J., concurring); id., at 517-520 (Marshall, J., concurring in judgment).” 497 U. S., at 563. We recalled that the opinions of Chief Justice Burger and Justice Powell in Fullilove had “explained that deference was appropriate in light of Congress’ institutional competence as the National Legislature, as well as Congress’ powers under the Commerce Clause, the Spending Clause, and the Civil War Amendments.” 497 U. S., at 563 (citations and footnote omitted).
The majority in Metro Broadcasting and the plurality in Fullilove were not alone in relying upon a critical distinction between federal and state programs. In his separate opinion in Richmond v. J. A. Croson Co., 488 U. S. 469, 520-524 (1989), Justice Scalia discussed the basis for this distinction. He observed that “it is one thing to permit racially based conduct by the Federal Government — whose legislative powers concerning matters of race were explicitly enhanced by the Fourteenth Amendment, see U. S. Const., Arndt. 14, § 5 — and quite another to permit it by the precise entities against whose conduct in matters of race that Amendment was specifically directed, see Arndt. 14, § 1.” Id., at 521-522. Continuing, Justice Scalia explained why a “sound distinction between federal and state (or local) action based on race rests not only upon the substance of the *251Civil War Amendments, but upon social reality and governmental theory.” Id., at 522.
“What the record shows, in other words, is that racial discrimination against any group finds a more ready expression at the state and local than at the federal level. To the children of the Founding Fathers, this should come as no surprise. An acute awareness of the heightened danger of oppression from political factions in small, rather than large, political units dates to the very beginning of our national history. See G. Wood, The Creation of the American Republic, 1776-1787, pp. 499-506 (1969). As James Madison observed in support of the proposed Constitution’s enhancement of national powers:
“ ‘The smaller the society, the fewer probably will be the distinct parties and interests composing it; the fewer the distinct parties and interests, the more frequently will a majority be found of the same party; and the smaller the number of individuals composing a majority, and the smaller the compass within which they are placed, the more easily will they concert and execute their plan of oppression. Extend the sphere and you take in a greater variety of parties and interests; you make it less probable that a majority of the whole will have a common motive to invade the rights of other citizens; or if such a common motive exists, it will be more difficult for all who feel it to discover their own strength and to act in unison with each other.’ The Federalist No. 10, pp. 82-84 (C. Rossiter ed. 1961).” Id., at 523 (opinion concurring in judgment).
In her plurality opinion in Croson, Justice O’Connor also emphasized the importance of this distinction when she responded to the city’s argument that Fullilove was controlling. She wrote:
*252“What appellant ignores is that Congress, unlike any State or political subdivision, has a specific constitutional mandate to enforce the dictates of the Fourteenth Amendment. The power to ‘enforce’ may at times also include the power to define situations which Congress determines threaten principles of equality and to adopt prophylactic rules to deal with those situations. The Civil War Amendments themselves worked a dramatic change in the balance between congressional and state power over matters of race.” 488 U. S., at 490 (joined by Rehnquist, C. J., and White, J.) (citations omitted).
An additional reason for giving greater deference to the National Legislature than to a local lawmaking body is that federal affirmative-action programs represent the will of our entire Nation’s elected representatives, whereas a state or local program may have an impact on nonresident entities who played no part in the decision to enact it. Thus, in the state or local context, individuals who were unable to vote for the local representatives who enacted a race-conscious program may nonetheless feel the effects of that program. This difference recalls the goals of the Commerce Clause, U. S. Const., Art. I, § 8, cl. 3, which permits Congress to legislate on certain matters of national importance while denying power to the States in this area for fear of undue impact upon out-of-state residents. See Southern Pacific Co. v. Arizona ex rel. Sullivan, 325 U. S. 761, 767-768, n. 2 (1945) (“[T]o the extent that the burden of state regulation falls on interests outside the state, it is unlikely to be alleviated by the operation of those political restraints normally exerted when interests within the state are affected”).
Ironically, after all of the time, effort, and paper this Court has expended in differentiating between federal and state affirmative action, the majority today virtually ignores the issue. See ante, at 230-231. It provides not a word of direct explanation for its sudden and enormous departure from *253the reasoning in past cases. Such silence, however, cannot erase the difference between Congress’ institutional competence and constitutional authority to overcome historic racial subjugation and the States’ lesser power to do so.
Presumably, the majority is now satisfied that its theory of “congruence” between the substantive rights provided by the Fifth and Fourteenth Amendments disposes of the objection based upon divided constitutional powers. But it is one thing to say (as no one seems to dispute) that the Fifth Amendment encompasses a general guarantee of equal protection as broad as that contained within the Fourteenth Amendment. It is another thing entirely to say that Congress’ institutional competence and constitutional authority entitles it to no greater deference' when it enacts a program designed to foster equality than the deference due a state legislature.7 The latter is an extraordinary proposition; and, as the foregoing discussion demonstrates, our precedents have rejected it explicitly and repeatedly.8
*254Our opinion in Metro Broadcasting relied on several constitutional provisions to justify the greater deference we owe to Congress when it acts with respect to private individuals. 497 U. S., at 563. In the programs challenged in this case, Congress has acted both with respect to private individuals and, as in Fullilove, with respect to the States themselves.9 When Congress does this, it draws its power directly from §5 of the Fourteenth Amendment.10 That section reads: *255“The Congress shall have power to enforce, by appropriate legislation, the provisions of this article.” One of the “provisions of this article” that Congress is thus empowered to enforce reads: “No State shall make or enforce any law which shall abridge the privileges or immunities of citizens of the United States; nor shall any State deprive any person of life, liberty, or property, without due process of law; nor deny to any person within its jurisdiction the equal protection of the laws.” U. S. Const., Arndt. 14, § 1. The Fourteenth Amendment directly empowers Congress at the same time it expressly limits the States.11 This is no accident. It represents our Nation’s consensus, achieved after hard experience throughout our sorry history of race relations, that the Federal Government must be the primary defender of racial minorities against the States, some of which may be inclined to oppress such minorities. A rule of “congruence” that ignores a purposeful “incongruity” so fundamental to our system of government is unacceptable.
In my judgment, the Court’s novel doctrine of “congruence” is seriously misguided. Congressional deliberations about a matter as important as affirmative action should be accorded far greater deference than those of a State or municipality.
IV
The Court’s concept of stare decisis treats some of the language we have used in explaining our decisions as though it *256were more important than our actual holdings. In my opinion that treatment is incorrect.
This is the third time in the Court’s entire history that it has considered the constitutionality of a federal affirmative-action program. On each of the two prior occasions, the first in 1980, Fullilove v. Klutznick, 448 U. S. 448, and the second in 1990, Metro Broadcasting, Inc. v. FCC, 497 U. S. 547, the Court upheld the program. Today the Court explicitly overrules Metro Broadcasting (at least in part), ante, at 227, and undermines Fullilove by recasting the standard on which it rested and by calling even its holding into question, ante, at 235. By way of explanation, Justice O’Connor advises the federal agencies and private parties that have made countless decisions in reliance on those cases that “we do not depart from the fabric of the law; we restore it.” Ante, at 234. A skeptical observer might ask whether this pronouncement is a faithful application of the doctrine of stare decisis.12 A brief comment on each of the two ailing cases may provide the answer.
In the Court’s view, our decision in Metro Broadcasting was inconsistent with the rule announced in Richmond v. J. A. Croson Co., 488 U. S. 469 (1989). Ante, at 225-226. But two decisive distinctions separate those two cases. First, Metro Broadcasting involved a federal program, whereas Croson involved a city ordinance. Metro Broadcasting thus drew primary support from Fullilove, which predated Croson and which Croson distinguished on the grounds of the federal-state dichotomy that the majority today discredits. Although Members of today’s majority trumpeted the importance of that distinction in Croson, they now reject it in the name of “congruence.” It is therefore *257quite wrong for the Court to suggest today that overruling Metro Broadcasting merely restores the status quo ante, for the law at the time of that decision was entirely open to the result the Court reached. Today's decision is an unjustified departure from settled law.
Second, Metro Broadcasting’s holding rested on more than its application of “intermediate scrutiny.” Indeed, I have always believed that, labels notwithstanding, the Federal Communications Commission (FCC) program we upheld in that case would have satisfied any of our various standards in affirmative-action cases — including the one the majority fashions today. What truly distinguishes Metro Broadcasting from our other affirmative-action precedents is the distinctive goal of the federal program in that case. Instead of merely seeking to remedy past discrimination, the FCC program was intended to achieve future benefits in the form of broadcast diversity. Reliance on race as a legitimate means of achieving diversity was first endorsed by Justice Powell in Regents of Univ. of Cal. v. Bakke, 438 U. S. 265, 311-319 (1978). Later, in Wygant v. Jackson Bd. of Ed., 476 U. S. 267 (1986), I also argued that race is not always irrelevant to governmental decisionmaking, see id., at 314-315 (Stevens, J., dissenting); in response, Justice O’Connor correctly noted that, although the school board had relied on an interest in providing black teachers to serve as role models for black students, that interest “should not be confused with the very different goal of promoting racial diversity among the faculty.” Id., at 288, n. She then added that, because the school board had not relied on an interest in diversity, it was not “necessary to discuss the magnitude of that interest or its applicability in this case.” Ibid.
Thus, prior to Metro Broadcasting, the interest in diversity had been mentioned in a few opinions, but it is perfectly clear that the Court had not yet decided whether that interest had sufficient magnitude to justify a racial classification. Metro Broadcasting, of course, answered that question in the *258affirmative. The majority today overrules Metro Broadcasting only insofar as it is “inconsistent with [the] holding” that strict scrutiny applies to “benign” racial classifications promulgated by the Federal Government. Ante, at 227. The proposition that fostering diversity may provide a sufficient interest to justify such a program is not inconsistent with the Court’s holding today — indeed, the question is not remotely presented in this case — and I do not take the Court’s opinion to diminish that aspect of our decision in Metro Broadcasting.
The Court’s suggestion that it may be necessary in the future to overrule Fullilove in order to restore the fabric of the law, ante, at 235, is even more disingenuous than its treatment of Metro Broadcasting. For the Court endorses the “strict scrutiny” standard that Justice Powell applied in Bakke, see ante, at 224, and acknowledges that he applied that standard in Fullilove as well, ante, at 218-219. Moreover, Chief Justice Burger also expressly concluded that the program we considered in Fullilove was valid under any of the tests articulated in Bakke, which of course included Justice Powell’s. 448 U. S., at 492. The Court thus adopts a standard applied in Fullilove at the same time it questions that case’s continued vitality and accuses it of departing from prior law. I continue to believe that the Fullilove case was incorrectly decided, see id., at 532-554 (Stevens, J., dissenting), but neither my dissent nor that filed by Justice Stewart, id., at 522-532, contained any suggestion that the issue the Court was resolving had been decided before.13 As was true *259of Metro Broadcasting, the Court in Fullilove decided an important, novel, and difficult question. Providing a different answer to a similar question today cannot fairly be characterized as merely “restoring” previously settled law.
V
The Court’s holding in Fullilove surely governs the result in this case. The Public Works Employment Act of 1977 (1977 Act), 91 Stat. 116, which this Court upheld in Fullilove, is different in several critical respects from the portions of the Small Business Act (SBA), 72 Stat. 384, as amended, 15 U. S. C. § 631 et seq., and STURAA, 101 Stat. 132, challenged in this case. Each of those differences makes the current program designed to provide assistance to DBE’s significantly less objectionable than the 1977 categorical grant of $400 million in exchange for a 10% set-aside in public contracts to “a class of investors defined solely by racial characteristics.” Fullilove, 448 U. S., at 532 (Stevens, J., dissenting). In no meaningful respect is the current scheme more objectionable than the 1977 Act. Thus, if the 1977 Act was constitutional, then so must be the SBA and STURAA. Indeed, even if my dissenting views in Fullilove had prevailed, this program would be valid.
Unlike the 1977 Act, the present statutory scheme does not make race the sole criterion of eligibility for participation in the program. Race does give rise to a rebuttable presumption of social disadvantage which, at least under STURAA,14 gives rise to a second rebuttable presumption *260of economic disadvantage. 49 CFR §23:62 (1994). But a small business may qualify as a DBE, by showing that it is both socially and economically disadvantaged, even if it receives neither of these presumptions. 13 CFR §§ 124.105(c), 124.106 (1995); 48 CFR § 19.703 (1994); 49 CFR pt. 23, subpt. D., Apps. A and C (1994). Thus, the current preference is more inclusive than the 1977 Act because it does not make race a necessary qualification.
More importantly, race is not a sufficient qualification. Whereas a millionaire with a long history of financial successes, who was a member of numerous social clubs and trade associations, would have qualified for a preference under the 1977 Act merely because he was an Asian-American or an African-American, see Fullilove, 448 U. S., at 537-538, 540, 543-544, and n. 16, 546 (Stevens, J., dissenting), neither the SBA nor STURAA creates any such anomaly. The DBE program excludes members of minority races who are not, in fact, socially or economically disadvantaged.15 13 CFR § 124.106(a)(ii) (1995); 49 CFR § 23.69 (1994). The presumption of social disadvantage reflects the unfortunate fact that irrational racial prejudice — along with its lingering effects— still survives.16 The presumption of economic disadvantage *261embodies a recognition that success in the private sector of the economy is often attributable, in part, to social skills and relationships. Unlike the 1977 set-asides, the current preference is designed to overcome the social and economic disadvantages that are often associated with racial characteristics. If, in a particular case, these disadvantages are not present, the presumptions can be rebutted. 13 CFR §§124.601-124.610 (1995); 49 CFR §23.69 (1994). The program is thus designed to allow race to play a part in the decisional process only when there is a meaningful basis for assuming its relevance. In this connection, I think it is particularly significant that the current program targets the negotiation of subcontracts between private firms. The 1977 Act applied entirely to the award of public contracts, an area of the economy in which social relationships should be irrelevant and in which proper supervision of government contracting officers should preclude any discrimination against particular bidders on account of their race. In this case, in contrast, the program seeks to overcome barriers of prejudice between private parties — specifically, between general contractors and subcontractors. The SBA and STURAA embody Congress’ recognition that such barriers may actually handicap minority firms seeking business as subcontractors from established leaders in the industry that have a history of doing business with their golfing partners. Indeed, minority subcontractors may face more obstacles than direct, intentional racial prejudice: They may face particular barriers simply because they are more likely to be new in the business and less likely to know others in the business. Given such difficulties, Congress could reasonably find that a minority subcontractor is less likely to receive favors from the entrenched businesspersons who award subcontracts only to people with whom — or with whose friends — they have an existing relationship. This program, then, if in part a remedy for past discrimination, is most importantly a *262forward-looking response to practical problems faced by minority subcontractors.
The current program contains another forward-looking component that the 1977 set-asides did not share. Section 8(a) of the SBA provides for periodic review of the status of DBE’s, 15 U.S.C. §§637(a)(BMQ (1988 ed., Supp. V); 13 CFR § 124.602(a) (1995),17 and DBE status can be challenged by a competitor at any time under any of the routes to certification. 13 CFR §124.603 (1995); 49 CFR §23.69 (1994). Such review prevents ineligible firms from taking part in the program solely because of their minority ownership, even when those firms were once disadvantaged but have since become successful. The emphasis on review also indicates the Administration’s anticipation that after their presumed disadvantages have been overcome, firms will “graduate” into a status in which they will be able to compete for business, including prime contracts, on an equal basis. 13 CFR § 124.208 (1995). As with other phases of the statutory policy of encouraging the formation and growth of small business enterprises, this program is intended to facilitate entry and increase competition in the free market.
Significantly, the current program, unlike the 1977 set-aside, does not establish any requirement — numerical or otherwise — that a general contractor must hire DBE subcontractors. The program we upheld in Fullilove required that 10% of the federal grant for every federally funded project be expended on minority business enterprises. In contrast, the current program contains no quota. Although it provides monetary incentives to general contractors to hire DBE subcontractors, it does not require them to hire DBE’s, *263and they do not lose their contracts if they fail to do so. The importance of this incentive to general contractors (who always seek to offer the lowest bid) should not be underestimated; but the preference here is far less rigid, and thus more narrowly tailored, than the 1977 Act. Cf. Bakke, 438 U. S., at 319-320 (opinion of Powell, J.) (distinguishing between numerical set-asides and consideration of race as a factor).
Finally, the record shows a dramatic contrast between the sparse deliberations that preceded the 1977 Act, see Fullilove, 448 U. S., at 549-550 (Stevens, J., dissenting), and the extensive hearings conducted in several Congresses before the current program was developed.18 However we might *264evaluate the benefits and costs — both fiscal and social — of this or any other affirmative-action program, our obligation to give deference to Congress’ policy choices is much more demanding in this case than it was in Fullilove. If the 1977 program of race-based set-asides satisfied the strict scrutiny dictated by Justice Powell’s vision of the Constitution — a vision the Court expressly endorses today — it must follow as night follows the day that the Court of Appeals’ judgment upholding this more carefully crafted program should be affirmed.
VI
My skeptical scrutiny of the Court’s opinion leaves me in dissent. The majority’s concept of “consistency” ignores a difference, fundamental to the idea of equal protection, between oppression and assistance. The majority’s concept of “congruence” ignores a difference, fundamental to our constitutional system, between the Federal Government and the States. And the majority’s concept of stare decisis ignores the force of binding precedent. I would affirm the judgment of the Court of Appeals.
As Justice Ginsburg observes, post, at 275-276, the majority’s “flexible” approach to “strict scrutiny” may well take into account differences between benign and invidious programs. The majority specifically notes that strict scrutiny can accommodate “ ‘relevant differences,’ ” ante, at 228; surely the intent of a government actor and the effects of a program are relevant to its constitutionality. See Missouri v. Jenkins, ante, at 112 (O’Connor, J., concurring) (“[Tjime and again, we have recognized the ample authority legislatures possess to combat racial injustice .... It is only by applying strict scrutiny that we can distinguish between unconstitutional discrimination and narrowly tailored remedial programs that legislatures may enact to further the compelling governmental interest in redressing the effects of past discrimination”).
Even if this is so, however, I think it is unfortunate that the majority insists on applying the label “strict scrutiny” to benign race-based programs. That label has usually been understood to spell the death of any governmental action to which a court may apply it. The Court suggests today that “strict scrutiny” means something different — something less strict — when applied to benign racial classifications. Although I agree that benign programs deserve different treatment than invidious programs, there is a danger that the fatal language of “strict scrutiny” will *244skew the analysis and place well-crafted benign programs at unnecessary risk.
These were, of course, neither the sole nor the most shameful burdens the Government imposed on Japanese-Americans during that War. They were, however, the only such burdens this Court had occasion to address in Hirabayashi and Korematsu. See Korematsu, 323 U. S., at 223 (“Regardless of the true nature of the assembly and relocation centers ... we are dealing specifically with nothing but an exclusion order”).
See Morton v. Mancari, 417 U. S. 535, 541 (1974). To be eligible for the preference in 1974, an individual had to “ ‘be one fourth or more degree Indian blood and be a member of a Federally-recognized tribe.’ ” Id., at 553, n. 24, quoting 44 BIAM 335, 3.1 (1972). We concluded that the classi*245fication was not “racial” because it did not encompass all Native Americans. 417 U. S., at 553-554. In upholding it, we relied in part on the plenary power of Congress to legislate on behalf of Indian tribes. Id., at 551-552. In this case respondents rely, in part, on the fact that not all members of the preferred minority groups are eligible for the preference, and on the special power to legislate on behalf of minorities granted to Congress by § 5 of the Fourteenth Amendment.
For example, in Richmond v. J. A. Croson Co., 488 U. S. 469 (1989), a majority of the members of the city council that enacted the race-based set-aside were of the same race as its beneficiaries.
In his concurrence, Justice Thomas argues that the most significant cost associated with an affirmative-action program is its adverse stigmatic effect on its intended beneficiaries. Ante, at 240-241. Although I agree that this cost may be more significant than many people realize, see Fullilove v. Klutznick, 448 U. S. 448, 545 (1980) (Stevens, J., dissenting), I do not think it applies to the facts of this case. First, this is not an argument that petitioner Adarand, a white-owned business, has standing to advance. No beneficiaries of the specific program under attack today have challenged its constitutionality — perhaps because they do not find the prefer-*248enees stigmatizing, or perhaps because their ability to opt out of the program provides them all the relief they would need. Second, even if the petitioner in this ease were a minority-owned business challenging the stigmatizing effect of this program, I would not find Justice Thomas’ extreme proposition — that there is a moral and constitutional equivalence between an attempt to subjugate and an attempt to redress the effects of a caste system, ante, at 240 — at all persuasive. It is one thing to question the wisdom of affirmative-action programs: There are many responsible arguments against them, including the one based upon stigma, that Congress might find persuasive when it decides whether to enact or retain race-based preferences. It is another thing altogether to equate the many well-meaning and intelligent lawmakers and their constituents— whether members of majority or minority races — who have supported affirmative action over the years, to segregationists and bigots.
Finally, although Justice Thomas is more concerned about the potential effects of these programs than the intent of those who enacted them (a proposition at odds with this Court’s jurisprudence, see Washington v. Davis, 426 U. S. 229 (1976), but not without a strong element of common sense, see id., at 252-266 (Stevens, J., concurring); id., at 256-270 (Brennan, J., dissenting)), I am not persuaded that the psychological damage brought on by affirmative action is as severe as that engendered by racial subordination. That, in any event, is a judgment the political branches can be trusted to make. In enacting affirmative-action programs, a legislature intends to remove obstacles that have unfairly placed individuals of equal qualifications at a competitive disadvantage. See Fullilove, 448 U. S., at 521 (Marshall, J., concurring in judgment). I do not believe such action, whether wise or unwise, deserves such an invidious label as “racial paternalism,” ante, at 240 (opinion of Thomas, J.). If the legislature is persuaded that its program is doing more harm than good to the individuals it is designed to benefit, then we can expect the legislature to remedy the problem. Significantly, this is not true of a government action based on invidious discrimination.
As I noted in Wygant:
“There is ... a critical difference between a decision to exclude a member of a minority race because of his or her skin color and a decision *249to include more members of the minority in a school faculty for that reason.
“The exclusionary decision rests on the false premise that differences in race, or in the color of a person’s skin, reflect real differences that are relevant to a person’s right to share in the blessings of a free society. As noted, that premise is ‘utterly irrational,’ Cleburne v. Cleburne Living Center, 473 U. S. 432, 452 (1985), and repugnant to the principles of a free and democratic society. Nevertheless, the fact that persons of different races do, indeed have differently colored skin, may give rise to a belief that there is some significant difference between such persons. The inclusion of minority teachers in the educational process inevitably tends to dispel that illusion whereas their exclusion could only tend to foster it. The inclusionary decision is consistent with the principle that all men are created equal; the exclusionary decision is at war with that principle. One decision accords with the Equal Protection Clause of the Fourteenth Amendment; the other does not. Thus, consideration of whether the consciousness of race is exclusionary or inclusionary plainly distinguishes the Board’s valid purpose in this case from a race-conscious decision that would reinforce assumptions of inequality.” 476 U. S., at. 316-317 (dissenting opinion).
Despite the majority’s reliance on Korematsu v. United States, 323 U. S. 214 (1944), ante, at 214-215, that case does not stand for the proposition that federal remedial programs are subject to strict scrutiny. Instead, Korematsu specifies that “all legal restrictions which curtail the civil rights of a single racial group are immediately suspect.” 323 U. S., at 216, quoted ante, at 214 (emphasis added). The programs at issue in this case (as in most affirmative-action cases) do not “curtail the civil rights of a single racial group”; they benefit certain racial groups and impose an indirect burden on the majority.
We have rejected this proposition outside of the affirmative-action context as well. In Hampton v. Mow Sun Wong, 426 U. S. 88, 100 (1976), we held:
“The federal sovereign, like the States, must govern impartially. The concept of equal justice under law is served by the Fifth Amendment’s guarantee of due process, as well as by the Equal Protection Clause of the Fourteenth Amendment. Although both Amendments require the same type of analysis, see Buckley v. Valeo, 424 U. S. 1, 93 [(1976)], the Court of Appeals correctly stated that the two protections are not always coextensive. Not only does the language of the two Amendments differ, but more importantly, there may be overriding national interests which justify *254selective federal legislation that would be unacceptable for an individual State. On the other hand, when a federal rule is applicable to only a limited territory, such as the District of Columbia, or an insular possession, and when there is no special national interest involved, the Due Process Clause has been construed as having the same significance as the Equal Protection Clause.”
The funding for the preferences challenged in this case comes from the Surface Transportation and Uniform Relocation Assistance Act of 1987 (STURAA), 101 Stat. 132, in which Congress has granted funds to the States in exchange for a commitment to foster subcontracting by disadvantaged business enterprises, or “DBE’s.” STURAA is also the source of funding for DBE preferences in federal highway contracting. Approximately 98% of STURAA’s funding is allocated to the States. Brief for Respondents 38, n. 34. Moreover, under STURAA States are empowered to certify businesses as “disadvantaged” for purposes of receiving subcontracting preferences in both state and federal contracts. STURAA § 106(c)(4), 101 Stat. 146.
In this case, Adarand has sued only the federal officials responsible for implementing federal highway contracting policy; it has not directly challenged DBE preferences granted in state contracts funded by STURAA. It is not entirely clear, then, whether the majority’s “congruence” rationale would apply to federally regulated state contracts, which may conceivably be within the majority’s view of Congress’ § 5 authority even if the federal contracts are not. See Metro Broadcasting, 497 U. S., at 603-604 (O’Con-nor, J., dissenting). As I read the majority’s opinion, however, it draws no distinctions between direct federal preferences and federal preferences achieved through subsidies to States. The extent to which STURAA intertwines elements of direct federal regulations with elements of federal conditions on grants to the States would make such a distinction difficult to sustain.
Because Congress has acted with respect to the States in enacting STURAA, we need not revisit today the difficult question of § 5’s application to pure federal regulation of individuals.
We have read § 5 as a positive grant of authority to Congress, not just to punish violations, but also to define and expand the scope of the Equal Protection Clause. Katzenbach v. Morgan, 384 U. S. 641 (1966). In Katz-enbach, this meant that Congress under §5 could require the States to allow non-English-speaking citizens to vote, even if denying such citizens a vote would not have been an independent violation of § 1. Id., at 648-651. Congress, then, can expand the coverage of § 1 by exercising its power under §5 when it acts to foster equality. Congress has done just that here; it has decided that granting certain preferences to minorities best serves the goals of equal protection.
Our skeptical observer might also notice that Justice O’Connor’s explanation for departing from settled precedent is joined only by Justice Kennedy. Ante, at 204. Three Members of the majority thus provide no explanation whatsoever for their unwillingness to adhere to the doctrine of stare decisis.
Of course, Justice Stewart believed that his view, disapproving of racial classifications of any kind, was consistent with this Court’s precedents. See ante, at 234-236, citing 448 U. S., at 623-626. But he did not claim that the question whether the Federal Government could engage in race-conscious affirmative action had been decided before Fullilove. The fact that a Justice dissents from an opinion means that he disagrees with the result; it does not usually mean that he believes the decision so departs from the fabric of the law that its reasoning ought to be repudiated at the *259next opportunity. Much less does a dissent bind or authorize a later majority to reject a precedent with which it disagrees.
STURAA accords a rebuttable presumption of both social and economic disadvantage to members of racial minority groups. 49 CFR §23.62 (1994). In contrast, §8(a) of the SBA accords a presumption only of social disadvantage, 13 CFR § 124.105(b) (1995); the applicant has the burden of demonstrating economic disadvantage, id., § 124.106. Finally, §8(d) of the SBA accords at least a presumption of social disadvantage, but it is ambiguous as to whether economic disadvantage is presumed or *260must be shown. See 15 U. S. C. § 637(d)(3) (1988 ed. and Supp. V); 13 CFR §124.601 (1995).
The Government apparently takes this exclusion seriously. See Autek Systems Corp. v. United States, 835 F. Supp. 13 (DC 1993) (upholding Small Business Administration decision that minority business owner’s personal income disqualified him from DBE status under § 8(a) program), aff’d, 43 F. 3d 712 (CADC 1994).
“The unhappy persistence of both the practice and the lingering effects of racial discrimination against minority groups in this country is an unfortunate reality, and government is not disqualified from acting in response to it.” Ante, at 237.
“Our findings clearly state that groups such as black Americans, Hispanic Americans, and Native Americans, have been and continue to be discriminated against and that this discrimination has led to the social disadvantagement of persons identified by society as members of those groups.” 124 Cong. Rec. 34097 (1978)
The Department of Transportation strongly urges States to institute periodic review of businesses certified as DBE’s under STURAA, 49 CFR pt. 23, subpt. D, App. A (1994), but it does not mandate such review. Respondents point us to no provisions for review of § 8(d) certification, although such review may be derivative for those businesses that receive § 8(d) certification as a result of § 8(a) or STURAA certification.
Respondents point us to the following legislative history:
H. R. 5612, To amend the Small Business Act to Extend the current SBA 8(a) Pilot Program: Hearing on H. R. 5612 before the Senate Select Committee on Small Business, 96th Cong., 2d Sess. (1980); Small and Minority Business in the Decade of the 1980’s (Part 1): Hearings before the House Committee on Small Business, 97th Cong., 1st Sess. (1981); Minority Business and Its Contribution to the U. S. Economy: Hearing before the Senate Committee on Small Business, 97th Cong., 2d Sess. (1982); Federal Contracting Opportunities for Minority and Women-Owned Businesses— An Examination of the 8(d) Subcontracting Program: Hearings before the Senate Committee on Small Business, 98th Cong., 1st Sess. (1983); Women Entrepreneurs — Their Success and Problems: Hearing before the Senate Committee on Small Business, 98th Cong., 2d Sess. (1984); State of Hispanic Small Business in America: Hearing before the Subcommittee on SBA and SBIC Authority, Minority Enterprise, and General Small Business Problems of the House Committee on Small Business, 99th Cong., 1st Sess. (1985); Minority Enterprise and General Small Business Problems: Hearing before the Subcommittee on SBA and SBIC Authority, Minority Enterprise, and General Small Business Problems of the House Committee on Small Business, 99th Cong., 2d Sess. (1986); Disadvantaged Business Set-Asides in Transportation Construction Projects: Hearings before the Subcommittee on Procurement, Innovation, and Minority Enterprise Development of the House Committee on Small Business, 100th Cong., 2d Sess. (1988); Barriers to Full Minority Participation in Federally Funded Highway Construction Projects: Hearing before a Subcommittee of the House Committee on Government Operations, 100th Cong., 2d Sess. (1988); *264Surety Bonds and Minority Contractors: Hearing before the Subcommittee on Commerce, Consumer Protection, and Competitiveness of the House Committee on Energy and Commerce, 100th Cong., 2d Sess. (1988); Small Business Problems: Hearings before the House Committee on Small Business, 100th Cong., 1st Sess. (1987). See Brief for Respondents 9-10, n. 9.