San Antonio Independent School District v. Rodriguez/Dissent Marshall

Mr. Justice Marshall, with whom Mr. Justice Douglas concurs, dissenting.

The Court today decides, in effect, that a State may constitutionally vary the quality of education which it offers its children in accordance with the amount of taxable wealth located in the school districts within which they reside. The majority's decision represents an abrupt departure from the mainstream of recent state and federal court decisions concerning the unconstitutionality of state educational financing schemes dependent upon taxable local wealth. More unfortunately, though, the '''[p. 71]''' majority's holding can only be seen as a retreat from our historic commitment to equality of educational opportunity and as unsupportable acquiescence in a system which deprives children in their earliest years of the chance to reach their full potential as citizens. The Court does this despite the absence of any substantial justification for a scheme which arbitrarily channels educational resources in accordance with the fortuity of the amount of taxable wealth within each district.

In my judgment, the right of every American to an equal start in life, so far as the provision of a state service as important as education is concerned, is far too vital to permit state discrimination on grounds as tenuous as those presented by this record. Nor can I accept the notion that it is sufficient to remit these appellees to the vagaries of the political process which, contrary to the majority's suggestion, has proved singularly unsuited to the task of providing a remedy for this discrimination. I, for one, am unsatisfied with the hope of an ultimate "political" solution sometime in the indefinite future while, in the meantime, countless children unjustifiably receive inferior educations that "may affect their hearts [p. 72] and minds in a way unlikely ever to be undone." Brown v. Board of Education, 347 U.S. 483, 494 (1954). I must therefore respectfully dissent.

I
The Court acknowledges that "substantial interdistrict disparities in school expenditures" exist in Texas, ante, at 15, and that these disparities are "largely attributable to differences in the amounts of money collected through local property taxation," ante, at 16. But instead of closely examining the seriousness of these disparities and the invidiousness of the Texas financing scheme, the Court undertakes an elaborate exploration of the efforts Texas has purportedly made to close the gaps between its districts in terms of levels of district wealth and resulting educational funding. Yet, however praiseworthy Texas' equalizing efforts, the issue in this case is not whether Texas is doing its best to ameliorate the worst features of a discriminatory scheme but, rather, whether the scheme itself is in fact unconstitutionally discriminatory in the face of the Fourteenth Amendment's guarantee of equal protection of the laws. When the Texas financing scheme is taken as a whole, I do not think it can be doubted that it produces a discriminatory impact on substantial numbers of the schoolage children of the State of Texas.

A
Funds to support public education in Texas are derived from three sources: local ad valorem property taxes; the Federal Government; and the state government. It is enlightening to consider these in order.

'''[p. 73]''' Under Texas law, the only mechanism provided the local school district for raising new, unencumbered revenues is the power to tax property located within its boundaries. At the same time, the Texas financing scheme effectively restricts the use of monies raised by local property taxation to the support of public education within the boundaries of the district in which they are raised, since any such taxes must be approved by a majority of the property-taxpaying voters of the district.

The significance of the local property tax element of the Texas financing scheme is apparent from the fact that it provides the funds to meet some 40% of the cost of public education for Texas as a whole. Yet the amount of revenue that any particular Texas district can raise is dependent on two factors—its tax rate and its amount of taxable property. The first factor is determined by the property-taxpaying voters of the district. But, regardless of the enthusiasm of the local voters for public '''[p. 74]''' education, the second factor—the taxable property wealth of the district—necessarily restricts the district's ability to raise funds to support public education. Thus, even though the voters of two Texas districts may be willing to make the same tax effort, the results for the districts will be substantially different if one is property rich while the other is property poor. The necessary effect of the Texas local property tax is, in short, to favor property-rich districts and to disfavor property-poor ones.

The seriously disparate consequences of the Texas local property tax, when that tax is considered alone, are amply illustrated by data presented to the District Court by appellees. These data included a detailed study of a sample of 110 Texas school districts for the 1967-1968 school year conducted by Professor Joel S. Berke of Syracuse University's Educational Finance Policy Institute. Among other things, this study revealed that the 10 richest districts examined, each of which had more than $100,000 in taxable property per pupil, raised through local effort an average of $610 per pupil, whereas the four poorest districts studied, each of which had less than $10,000 in taxable property per pupil, were able '''[p. 75]''' to raise only an average of $63 per pupil. And, as the Court effectively recognizes, ante, at 27, this correlation between the amount of taxable property per pupil and the amount of local revenues per pupil holds true for the 96 districts in between the richest and poorest districts.

It is clear, moreover, that the disparity of per-pupil revenues cannot be dismissed as the result of lack of local effort—that is, lower tax rates—by property-poor districts. To the contrary, the data presented below indicate that the poorest districts tend to have the highest tax rates and the richest districts tend to have the lowest tax rates. Yet, despite the apparent extra effort being made by the poorest districts, they are unable even to begin to match the richest districts in terms of the production of local revenues. For example, the 10 richest districts studied by Professor Berke were able to produce $585 per pupil with an equalized tax rate of 31 cents '''[p. 76]''' on $100 of equalized valuation, but the four poorest districts studied, with an equalized rate of 70 cents on $100 of equalized valuation, were able to produce only $60 per pupil. Without more, this state-imposed system of educational funding presents a serious picture of widely varying treatment of Texas school districts, and thereby of Texas schoolchildren, in terms of the amount of funds available for public education.

Nor are these funding variations corrected by the other aspects of the Texas financing scheme. The Federal Government provides funds sufficient to cover only some 10% of the total cost of public education in Texas. Furthermore, while these federal funds are not distributed in Texas solely on a per-pupil basis, appellants do not here contend that they are used in such a way as to ameliorate significantly the widely varying consequences for Texas school districts and school children of the local property tax element of the state financing scheme.

State funds provide the remaining some 50% of the monies spent on public education in Texas. Technically, they are distributed under two programs. The first is the Available School Fund, for which provision is made in the Texas Constitution. The Available '''[p. 77]''' School Fund is composed of revenues obtained from a number of sources, including receipts from the state ad valorem property tax, one-fourth of all monies collected by the occupation tax, annual contributions by the legislature from general revenues, and the revenues derived from the Permanent School Fund. For the 1970-1971 school year the Available School Fund contained $296,000,000. The Texas Constitution requires that this money be distributed annually on a per capita basis to the local school districts. Obviously, such a flat grant could not alone eradicate the funding differentials attributable to the local property tax. Moreover, today the Available School Fund is in reality simply one facet of the second state financing program, the Minimum Foundation School Program, since each district's annual share of the Fund is deducted from the sum to which the district is entitled under the Foundation Program.

The Minimum Foundation School Program provides funds for three specific purposes: professional salaries, current operating expenses, and transportation expenses. The State pays, on an overall basis, for approximately 80% of the cost of the Program; the remaining 20% is distributed among the local school districts under the '''[p. 78]''' Local Fund Assignment. Each district's share of the Local Fund Assignment is determined by a complex "economic index" which is designed to allocate a larger share of the costs to property-rich districts than to property-poor districts. Each district pays its share with revenues derived from local property taxation.

The stated purpose of the Minimum Foundation School Program is to provide certain basic funding for each local Texas school district. At the same time, the Program was apparently intended to improve, to some degree, the financial position of property-poor districts relative to property-rich districts, since—through the use of the economic index—an effort is made to charge a disproportionate share of the costs of the Program to rich districts. It bears noting, however, that substantial criticism has been leveled at the practical effectiveness of the economic index system of local cost allocation. In theory, the index is designed to ascertain the relative ability of each district to contribute to the Local Fund Assignment from local property taxes. Yet the index is not developed simply on the basis of each district's taxable wealth. It also takes into account the district's relative income from manufacturing, mining, and agriculture, its payrolls, and its scholastic population. '''[p. 79]''' It is difficult to discern precisely how these latter factors are predictive of a district's relative ability to raise revenues through local property taxes. Thus, in 1966, one of the consultants who originally participated in the development of the Texas economic index adopted in 1949 told the Governor's Committee on Public School Education: "The Economic Index approach to evaluating local ability offers a little better measure than sheer chance, but not much."

Moreover, even putting aside these criticisms of the economic index as a device for achieving meaningful district wealth equalization through cost allocation, poor districts still do not necessarily receive more state aid than property-rich districts. For the standards which currently determine the amount received from the Foundation School Program by any particular district favor property-rich districts. Thus, focusing on the same '''[p. 80]''' Edgewood Independent and Alamo Heights School Districts which the majority uses for purposes of illustration, we find that in 1967-1968 property-rich Alamo Heights, which raised $333 per pupil on an equalized tax rate of 85 cents per $100 valuation, received $225 per pupil from the Foundation School Program, while property-poor Edgewood , which raised only $26 per pupil with an equalized tax rate of $1.05 per $100 valuation, received only $222 per pupil from the Foundation School Program. And, more recent data, which indicate that for the 1970-1971 school year Alamo Heights received $491 per pupil from '''[p. 81]''' the Program while Edgewood received only $356 per pupil, hardly suggest that the wealth gap between the districts is being narrowed by the State Program. To the contrary, whereas in 1967-1968 Alamo Heights received only $3 per pupil, or about 1%, more than Edgewood in state aid, by 1970-1971 the gap had widened to a difference of $135 per pupil, or about 38%. It was data of this character that prompted the District Court to observe that "the current [state aid] system tends to subsidize the rich at the expense of the poor, rather than the other way around." 337 F.Supp. 280, 282. And even the appellants go no further here than to venture that the Minimum Foundation School Program has "a mildly equalizing effect."

Despite these facts, the majority continually emphasizes how much state aid has, in recent years, been given '''[p. 82]''' to property-poor Texas school districts. What the Court fails to emphasize is the cruel irony of how much more state aid is being given to property-rich Texas school districts on top of their already substantial local property tax revenues. Under any view, then, it is apparent that the state aid provided by the Foundation School Program fails to compensate for the large funding variations attributable to the local property tax element of the Texas financing scheme. And it is these stark differences in the treatment of Texas school districts and school children inherent in the Texas financing scheme, not the absolute amount of state aid provided to any particular school district, that are the crux of this case. There can, moreover, be no escaping the conclusion that the local property tax which is dependent upon taxable district property wealth is an essential feature of the Texas scheme for financing public education.

B
The appellants do not deny the disparities in educational funding caused by variations in taxable district property wealth. They do contend, however, that whatever the differences in per-pupil spending among Texas districts, there are no discriminatory consequences for the children of the disadvantaged districts. They recognize that what is at stake in this case is the quality of the '''[p. 83]''' public education provided Texas children in the districts in which they live. But appellants reject the suggestion that the quality of education in any particular district is determined by money—beyond some minimal level of funding which they believe to be assured every Texas district by the Minimum Foundation School Program. In their view, there is simply no denial of equal educational opportunity to any Texas schoolchildren as a result of the widely varying per-pupil spending power provided districts under the current financing scheme.

In my view, though, even an unadorned restatement of this contention is sufficient to reveal its absurdity. Authorities concerned with educational quality no doubt disagree as to the significance of variations in per-pupil spending. Indeed, conflicting expert testimony was presented to the District Court in this case concerning the effect of spending variations on educational achievement. We sit, however, not to resolve disputes over educational theory but to enforce our Constitution. It is an inescapable fact that if one district has more funds available per pupil than another district, the '''[p. 84]''' former will have greater choice in educational planning than will the latter. In this regard, I believe the question of discrimination in educational quality must be deemed to be an objective one that looks to what the State provides its children, not to what the children are able to do with what they receive. That a child forced to attend an underfunded school with poorer physical facilities, less experienced teachers, larger classes, and a narrower range of courses than a school with substantially more funds—and thus with greater choice in educational planning—may nevertheless excel is to the credit of the child, not the State, cf. ''Missouri ex rel. Gaines v. Canada'', 305 U.S. 337, 349 (1938). Indeed, who can ever measure for such a child the opportunities lost and the talents wasted for want of a broader, more enriched education? Discrimination in the opportunity to learn that is afforded a child must be our standard.

Hence, even before this Court recognized its duty to tear down the barriers of state-enforced racial segregation in public education, it acknowledged that inequality in the educational facilities provided to students may be discriminatory state action as contemplated by the Equal Protection Clause. As a basis for striking down state-enforced segregation of a law school, the Court in Sweatt v. Painter, 339 U.S. 629, 633-634 (1950), stated:

"[W]e cannot find substantial equality in the educational opportunities offered white and Negro law students by the State. In terms of number of the faculty, variety of courses and opportunity for specialization, size of the student body, scope of the library, availability of law review and similar activities, the [whites-only] Law School is superior.... It is difficult to believe that one who had a free choice between these law schools would consider the question close."

'''[p. 85]' See also McLaurin v. Oklahoma State Regents for Higher Education'', 339 U.S. 637 (1950). Likewise, it is difficult to believe that if the children of Texas had a free choice, they would choose to be educated in districts with fewer resources, and hence with more antiquated plants, less experienced teachers, and a less diversified curriculum. In fact, if financing variations are so insignificant to educational quality, it is difficult to understand why a number of our country's wealthiest school districts, which have no legal obligation to argue in support of the constitutionality of the Texas legislation, have nevertheless zealously pursued its cause before this Court.

The consequences, in terms of objective educational input, of the variations in district funding caused by the Texas financing scheme are apparent from the data introduced before the District Court. For example, in 1968-1969, 100% of the teachers in the property-rich Alamo Heights School District had college degrees. By contrast, during the same school year only 80.02% of the teachers had college degrees in the property poor Edgewood Independent School District. Also, in 1968-1969, approximately 47% of the teachers in the Edgewood District were on emergency teaching permits, whereas only 11% of the teachers in Alamo Heights were on such permits. This is undoubtedly a reflection of the fact that the top of Edgewood's teacher salary scale was '''[p. 86]''' approximately 80% of Alamo Heights'. And, not surprisingly, the teacher-student ratio varies significantly between the two districts. In other words, as might be expected, a difference in the funds available to districts results in a difference in educational inputs available for a child's public education in Texas. For constitutional purposes, I believe this situation, which is directly attributable to the Texas financing scheme, raises a grave question of state-created discrimination in the provision of public education. Cf. Gaston County v. United States, 395 U.S. 285, 293-294 (1969).

At the very least, in view of the substantial interdistrict disparities in funding and in resulting educational inputs shown by appellees to exist under the Texas financing scheme, the burden of proving that these disparities do not in fact affect the quality of children's education must fall upon the appellants. Cf. Hobson v. Hansen, 327 F.Supp. 844, 860-861 (DC 1971). Yet appellants made no effort in the District Court to demonstrate that educational quality is not affected by variations in funding and in resulting inputs. And, in this Court, they have argued no more than that the relationship is ambiguous. This is hardly sufficient to overcome appellees' prima facie showing of state-created discrimination between the school children of Texas with respect to objective educational opportunity.

Nor can I accept the appellants' apparent suggestion that the Texas Minimum Foundation School Program effectively eradicates any discriminatory effects otherwise resulting from the local property tax element of the '''[p. 87]''' Texas financing scheme. Appellants assert that, despite its imperfections, the Program "does guarantee an adequate education to every child." The majority, in considering the constitutionality of the Texas financing scheme, seems to find substantial merit in this contention, for it tells us that the Foundation Program "was designed to provide an adequate minimum educational offering in every school in the State," ante, at 45, and that the Program "assur[es] a basic education for every child," ante, at 49. But I fail to understand how the constitutional problems inherent in the financing scheme are eased by the Foundation Program. Indeed, the precise thrust of the appellants' and the Court's remarks are not altogether clear to me.

The suggestion may be that the state aid received via the Foundation Program sufficiently improves the position of property-poor districts vis-à-vis property-rich districts—in terms of educational funds—to eliminate any claim of interdistrict discrimination in available educational resources which might otherwise exist if educational funding were dependent solely upon local property taxation. Certainly the Court has recognized that to demand precise equality of treatment is normally unrealistic, and thus minor differences inherent in any practical context usually will not make out a substantial equal protection claim. See, e.g., Mayer v. City of Chicago, 404 U.S. 189, 194-195 (1971); Draper v. Washington, 372 U.S. 487, 495-496 (1963); Bain Peanut Co. v. Pinson, 282 U.S. 499, 501 (1931). But, as has already been seen, we are hardly presented here with some de minimis claim of discrimination resulting from the play necessary in any functioning system; to the contrary, it is clear that the Foundation Program utterly fails to '''[p. 88]''' ameliorate the seriously discriminatory effects of the local property tax.

Alternatively, the appellants and the majority may believe that the Equal Protection Clause cannot be offended by substantially unequal state treatment of persons who are similarly situated so long as the State provides everyone with some unspecified amount of education which evidently is "enough." The basis for such a novel view is far from clear. It is, of course, true that the Constitution does not require precise equality in the treatment of all persons. As Mr. Justice Frankfurter explained:

"'The equality at which the 'equal protection' clause aims is not a disembodied equality. The Fourteenth Amendment enjoins 'the equal protection of the laws,' and laws are not abstract propositions.... The Constitution does not require things which are different in fact or opinion to be treated in law as though they were the same.' Tigner v. Texas, 310 U.S. 141, 147 (1940)."

See also Douglas v. California, 372 U.S. 353, 357 (1963); Goesaert v. Cleary, 335 U.S. 464, 466 (1948). '''[p. 89]''' But this Court has never suggested that because some "adequate" level of benefits is provided to all, discrimination in the provision of services is therefore constitutionally excusable. The Equal Protection Clause is not addressed to the minimal sufficiency but rather to the unjustifiable inequalities of state action. It mandates nothing less than that "all persons similarly circumstanced shall be treated alike." F.S. Royster Guano Co. v. Virginia, 253 U.S. 412, 415 (1920).

Even if the Equal Protection Clause encompassed some theory of constitutional adequacy, discrimination in the provision of educational opportunity would certainly seem to be a poor candidate for its application. Neither the majority nor appellants inform us how judicially manageable standards are to be derived for determining how much education is "enough" to excuse constitutional discrimination. One would think that the majority would heed its own fervent affirmation of judicial self-restraint before undertaking the complex task of determining at large what level of education is constitutionally sufficient. Indeed, the majority's apparent reliance upon the adequacy of the educational opportunity assured by the Texas Minimum Foundation School Program seems fundamentally inconsistent with its own recognition that educational authorities are unable to agree upon what makes for educational quality, see ante, at 42-43 and n. 86 and at 47 n. 101. If, as the majority stresses, such authorities are uncertain as to the impact of various levels of funding on educational quality, I fail to see where it finds the expertise to divine that the particular levels of funding provided by the Program assure an adequate educational opportunity—much less an education substantially equivalent in quality to that which a higher level of funding might provide. Certainly appellants' mere assertion before this Court of the adequacy of the education guaranteed by the Minimum '''[p. 90]''' Foundation School Program cannot obscure the constitutional implications of the discrimination in educational funding and objective educational inputs resulting from the local property tax—particularly since the appellees offered substantial uncontroverted evidence before the District Court impugning the now much touted "adequacy" of the education guaranteed by the Foundation Program.

In my view, then, it is inequality—not some notion of gross inadequacy—of educational opportunity that raises a question of denial of equal protection of the laws. I find any other approach to the issue unintelligible and without directing principle. Here appellees have made a substantial showing of wide variations in educational funding and the resulting educational opportunity afforded to the school children of Texas. This discrimination is, in large measure, attributable to significant disparities in the taxable wealth of local Texas school districts. This is a sufficient showing to raise a substantial question of discriminatory state action in violation of the Equal Protection Clause.

C
'''[p. 91]''' Despite the evident discriminatory effect of the Texas financing scheme, both the appellants and the majority raise substantial questions concerning the precise character of the disadvantaged class in this case. The District Court concluded that the Texas financing scheme draws "distinction between groups of citizens depending upon the wealth of the district in which they live" and thus creates a disadvantaged class composed of persons living in property-poor districts. See 337 F.Supp., at 282. See also id., at 281. In light of the data introduced before the District Court, the conclusion that the schoolchildren of property-poor districts constitute a sufficient class for our purposes seems indisputable to me.

Appellants contend, however, that in constitutional terms this case involves nothing more than discrimination against local school districts, not against individuals, since on its face the state scheme is concerned only with the provision of funds to local districts. The result of the Texas financing scheme, appellants suggest, is merely that some local districts have more available revenues for education; others have less. In that respect, '''[p. 92]''' they point out, the States have broad discretion in drawing reasonable distinctions between their political subdivisions. See Griffin v. County School Board of Prince Edward County, 377 U.S. 218, 231 (1964); McGowan v. Maryland, 366 U.S. 420, 427 (1961); Salsburg v. Maryland, 346 U.S. 545, 550-554 (1954).

But this Court has consistently recognized that where there is in fact discrimination against individual interests, the constitutional guarantee of equal protection of the laws is not inapplicable simply because the discrimination is based upon some group characteristic such as geographic location. See Gordon v. Lance, 403 U.S. 1, 4 (1971); Reynolds v. Sims, 377 U.S. 533, 565-566 (1964); Gray v. Sanders, 372 U.S. 368, 379 (1963). Texas has chosen to provide free public education for all its citizens, and it has embodied that decision in its constitution. Yet, having established public education for its citizens, the State, as a direct consequence of the variations in local property wealth endemic to Texas' financing scheme, has provided some Texas schoolchildren with substantially less resources for their education than others. Thus, while on its face the Texas scheme may merely discriminate between local districts, the impact of that discrimination falls directly upon the children whose educational opportunity is dependent upon where they happen to live. Consequently, the District Court correctly concluded that the Texas financing scheme discriminates, from a constitutional perspective, between schoolchildren on the basis of the amount of taxable property located within their local districts.

In my Brother Stewart's view, however, such a description of the discrimination inherent in this case is apparently not sufficient, for it fails to define the "kind of objectively identifiable classes" that he evidently perceives '''[p. 93]' to be necessary for a claim to be "cognizable under the Equal Protection Clause," ante'', at 62. He asserts that this is also the view of the majority, but he is unable to cite, nor have I been able to find, any portion of the Court's opinion which remotely suggests that there is no objectively identifiable or definable class in this case. In any event, if he means to suggest that an essential predicate to equal protection analysis is the precise identification of the particular individuals who compose the disadvantaged class, I fail to find the source from which he derives such a requirement. Certainly such precision is not analytically necessary. So long as the basis of the discrimination is clearly identified, it is possible to test it against the State's purpose for such discrimination—whatever the standard of equal protection analysis employed. This is clear from our decision only last Term in Bullock v. Carter, 405 U.S. 134 (1972), where the Court, in striking down Texas' primary filing fees as violative of equal protection, found no impediment to equal protection analysis in the fact that the members of the disadvantaged class could not be readily identified. The Court recognized that the filing-fee system tended "to deny some voters the opportunity to vote for a candidate of their choosing; at the same time it gives the affluent the power to place on the ballot their own names or the names of persons they favor." Id., at 144. The '''[p. 94]''' Court also recognized that "[t]his disparity in voting power based on wealth cannot be described by reference to discrete and precisely defined segments of the community as is typical of inequities challenged under the Equal Protection Clause..." Ibid. Nevertheless, it concluded that "we would ignore reality were we not to recognize that this system falls with unequal weight on voters...according to their economic status." Ibid. The nature of the classification in Bullock was clear, although the precise membership of the disadvantaged class was not. This was enough in Bullock for purposes of equal protection analysis. It is enough here.

It may be, though, that my Brother Stewart is not in fact demanding precise identification of the membership of the disadvantaged class for purposes of equal protection analysis, but is merely unable to discern with sufficient clarity the nature of the discrimination charged in this case. Indeed, the Court itself displays some uncertainty as to the exact nature of the discrimination and the resulting disadvantaged class alleged to exist in this case. See ante, at 19-20. It is, of course, essential to equal protection analysis to have a firm grasp upon the nature of the discrimination at issue. In fact, the absence of such a clear, articulable understanding of the nature of alleged discrimination in a particular instance may well suggest the absence of any real discrimination. But such is hardly the case here.

A number of theories of discrimination have, to be sure, been considered in the course of this litigation. Thus, the District Court found that in Texas the poor and minority group members tend to live in property-poor districts, suggesting discrimination on the basis of both personal wealth and race. See 337 F.Supp., at 282 and n. 3. The Court goes to great lengths to discredit the data upon which the District Court relied, and thereby its conclusion that poor people live in property-poor districts. '''[p. 95]''' Although I have serious doubts as to the correctness of the Court's analysis in rejecting the data submitted below, I have no need to join issue on these factual disputes.

'''[p. 96]''' I believe it is sufficient that the overarching form of discrimination in this case is between the schoolchildren of Texas on the basis of the taxable property wealth of the districts in which they happen to live. To understand both the precise nature of this discrimination and the parameters of the disadvantaged class it is sufficient to consider the constitutional principle which appellees contend is controlling in the context of educational financing. In their complaint appellees asserted that the Constitution does not permit local district wealth to be determinative of educational opportunity. This is simply another way of saying, as the District Court concluded, that consistent with the guarantee of equal protection of the laws, "the quality of public education may not be a function of wealth, other than the wealth of the state as a whole." 337 F.Supp., at 284. Under such a principle, the children of a district are excessively advantaged if that district has more taxable property per pupil than the average amount of taxable property per pupil considering the State as a whole. By contrast, the children of a district are disadvantaged if that district has less taxable property per pupil than the state average. The majority attempts to disparage such a definition of the disadvantaged class as the product of an "artificially defined level" of district wealth. Ante, at 28. But such is clearly not the case, for this is the '''[p. 97]''' definition unmistakably dictated by the constitutional principle for which appellees have argued throughout the course of this litigation. And I do not believe that a clearer definition of either the disadvantaged class of Texas schoolchildren or the allegedly unconstitutional discrimination suffered by the members of that class under the present Texas financing scheme could be asked for, much less needed. Whether this discrimination, against the schoolchildren of property-poor districts, inherent in the Texas financing scheme, is violative of the Equal Protection Clause is the question to which we must now turn.

II
To avoid having the Texas financing scheme struck down because of the interdistrict variations in taxable property wealth, the District Court determined that it was insufficient for appellants to show merely that the State's scheme was rationally related to some legitimate state purpose; rather, the discrimination inherent in the scheme had to be shown necessary to promote a "compelling state interest" in order to withstand constitutional scrutiny. The basis for this determination was twofold: first, the financing scheme divides citizens on a wealth basis, a classification which the District Court viewed as highly suspect; and second, the discriminatory scheme directly affects what it considered to be a "fundamental interest," namely, education.

This Court has repeatedly held that state discrimination which either adversely affects a "fundamental interest," see, e.g., Dunn v. Blumstein, 405 U.S. 330, 336-342 (1972); Shapiro v. Thompson, 394 U.S. 618, 629-631 (1969), or is based on a distinction of a suspect character, see, e.g., Graham v. Richardson, 403 U.S. 365, 372 '''[p. 98]' (1971); McLaughlin v. Florida'', 379 U.S. 184, 191-192 (1964), must be carefully scrutinized to ensure that the scheme is necessary to promote a substantial, legitimate state interest. See, e.g., Dunn v. Blumstein, supra, at 342-343; Shapiro v. Thompson, supra, at 634. The majority today concludes, however, that the Texas scheme is not subject to such a strict standard of review under the Equal Protection Clause. Instead, in its view, the Texas scheme must be tested by nothing more than that lenient standard of rationality which we have traditionally applied to discriminatory state action in the context of economic and commercial matters. See, e.g., McGowan v. Maryland, 366 U.S., at 425-426; Morey v. Doud, 354 U.S. 457, 465-466 (1957); F.S. Royster Guano Co. v. Virginia, 253 U.S., at 415; Lindsley v. Natural Carbonic Gas Co., 220 U.S. 61, 78-79 (1911). By so doing, the Court avoids the telling task of searching for a substantial state interest which the Texas financing scheme, with its variations in taxable district property wealth, is necessary to further. I cannot accept such an emasculation of the Equal Protection Clause in the context of this case.

A
To begin, I must once more voice my disagreement with the Court's rigidified approach to equal protection analysis. See Dandridge v. Williams, 397 U.S. 471, 519-521 (1970) (dissenting opinion); Richardson v. Belcher, 404 U.S. 78, 90 (1971) (dissenting opinion). The Court apparently seeks to establish today that equal protection cases fall into one of two neat categories which dictate the appropriate standard of review—strict scrutiny or mere rationality. But this Court's decisions in the field of equal protection defy such easy categorization. A principled reading of what this Court has done reveals that it has applied a spectrum of standards in reviewing discrimination allegedly violative of the Equal Protection '''[p. 99]''' Clause. This spectrum clearly comprehends variations in the degree of care with which the Court will scrutinize particular classifications, depending, I believe, on the constitutional and societal importance of the interest adversely affected and the recognized invidiousness of the basis upon which the particular classification is drawn. I find in fact that many of the Court's recent decisions embody the very sort of reasoned approach to equal protection analysis for which I previously argued—that is, an approach in which "concentration [is] placed upon the character of the classification in question, the relative importance to individuals in the class discriminated against of the governmental benefits that they do not receive, and the asserted state interests in support of the classification." Dandridge v. Williams, supra, at 520-521 (dissenting opinion).

I therefore cannot accept the majority's labored efforts to demonstrate that fundamental interests, which call for strict scrutiny of the challenged classification, encompass only established rights which we are somehow bound to recognize from the text of the Constitution itself. To be sure, some interests which the Court has deemed to be fundamental for purposes of equal protection analysis are themselves constitutionally protected rights. Thus, discrimination against the guaranteed right of freedom of speech has called for strict judicial scrutiny. See Police Dept. of Chicago v. Mosley, 408 U.S. 92 (1972). Further, every citizen's right to travel interstate, although nowhere expressly mentioned in the Constitution, has long been recognized as implicit in the premises underlying that document: the right "was conceived from the beginning to be a necessary concomitant of the stronger Union the Constitution created." United States v. Guest, 383 U.S. 745, 758 (1966). See also Crandall v. Nevada, 6 Wall. 35, 48 (1868). Consequently, the Court has required that a state classification affecting the constitutionally '''[p. 100]' protected right to travel must be "shown to be necessary to promote a compelling'' governmental interest." Shapiro v. Thompson, 394 U.S., at 634. But it will not do to suggest that the "answer" to whether an interest is fundamental for purposes of equal protection analysis is always determined by whether that interest "is a right...explicitly or implicitly guaranteed by the Constitution," ante, at 33-34.

I would like to know where the Constitution guarantees the right to procreate, Skinner v. Oklahoma, 316 U.S. 535, 541 (1942), or the right to vote in state elections, e.g., Reynolds v. Sims, 377 U.S. 533 (1964), or the right to an appeal from a criminal conviction, e.g., Griffin v. Illinois, 351 U.S. 12 (1956). These are instances in which, due to the importance of the interests at stake, the Court has displayed a strong concern with the existence of discriminatory state treatment. But the Court has never said or indicated that these are interests which independently enjoy full-blown constitutional protection.

Thus, in Buck v. Bell, 274 U.S. 200 (1927), the Court refused to recognize a substantive constitutional guarantee of the right to procreate. Nevertheless, in Skinner v. Oklahoma, supra, at 541, the Court, without impugning the continuing validity of Buck v. Bell, held that "strict scrutiny" of state discrimination affecting procreation "is essential," for "[m]arriage and procreation are fundamental to the very existence and survival of the race." Recently, in Roe v. Wade, 410 U.S. 113, 152-154 (1973), '''[p. 101]''' the importance of procreation has indeed been explained on the basis of its intimate relationship with the constitutional right of privacy which we have recognized. Yet the limited stature thereby accorded any "right" to procreate is evident from the fact that at the same time the Court reaffirmed its initial decision in Buck v. Bell. See Roe v. Wade, supra, at 154.

Similarly, the right to vote in state elections has been recognized as a "fundamental political right," because the Court concluded very early that it is "preservative of all rights." Yick Wo v. Hopkins, 118 U.S. 356, 370 (1886); see, e.g., Reynolds v. Sims, 377 U.S., at 561-562. For this reason, "this Court has made clear that a citizen has a constitutionally protected right to participate in elections on an equal basis with other citizens in the jurisdiction." Dunn v. Blumstein, 405 U.S., at 336 (emphasis added). The final source of such protection from inequality in the provision of the state franchise is, of course, the Equal Protection Clause. Yet it is clear that whatever degree of importance has been attached to the state electoral process when unequally distributed, the right to vote in state elections has itself never been accorded the stature of an independent constitutional guarantee. See Oregon v. Mitchell, 400 U.S. 112 (1970); Kramer v. Union School District, 395 U.S. 621, 626-629 (1969); ''Harper v. Virginia Bd. of Elections'', 383 U.S. 663, 665 (1966).

'''[p. 102]''' Finally, it is likewise "true that a State is not required by the Federal Constitution to provide appellate courts or a right to appellate review at all." Griffin v. Illinois, 351 U.S., at 18. Nevertheless, discrimination adversely affecting access to an appellate process which a State has chosen to provide has been considered to require close judicial scrutiny. See, e.g., Griffin v. Illinois, supra; Douglas v. California, 372 U.S. 353 (1963).

The majority is, of course, correct when it suggests that the process of determining which interests are fundamental is a difficult one. But I do not think the problem is insurmountable. And I certainly do not accept the view that the process need necessarily degenerate into an unprincipled, subjective "picking-and-choosing" between various interests or that it must involve this Court in creating "substantive constitutional rights in the name of guaranteeing equal protection of the laws," ante, at 33. Although not all fundamental interests are constitutionally guaranteed, the determination of which interests are fundamental should be firmly rooted in the text of the Constitution. The task in every case should be to determine the extent to which constitutionally guaranteed rights are dependent on interests not mentioned in the Constitution. As the nexus between the specific constitutional guarantee and the nonconstitutional interest draws closer, the nonconstitutional interest becomes '''[p. 103]''' more fundamental and the degree of judicial scrutiny applied when the interest is infringed on a discriminatory basis must be adjusted accordingly. Thus, it cannot be denied that interests such as procreation, the exercise of the state franchise, and access to criminal appellate processes are not fully guaranteed to the citizen by our Constitution. But these interests have nonetheless been afforded special judicial consideration in the face of discrimination because they are, to some extent, interrelated with constitutional guarantees. Procreation is now understood to be important because of its interaction with the established constitutional right of privacy. The exercise of the state franchise is closely tied to basic civil and political rights inherent in the First Amendment. And access to criminal appellate processes enhances the integrity of the range of rights implicit in the Fourteenth Amendment guarantee of due process of law. Only if we closely protect the related interests from state discrimination do we ultimately ensure the integrity of the constitutional guarantee itself. This is the real lesson that must be taken from our previous decisions involving interests deemed to be fundamental.

The effect of the interaction of individual interests with established constitutional guarantees upon the degree of care exercised by this Court in reviewing state discrimination affecting such interests is amply illustrated by our decision last Term in Eisenstadt v. Baird, 405 U.S. 438 (1972). In Baird, the Court struck down as violative of the Equal Protection Clause a state statute which denied unmarried persons access to contraceptive devices on the same basis as married persons. The Court '''[p. 104]''' purported to test the statute under its traditional standard whether there is some rational basis for the discrimination effected. Id., at 446-447. In the context of commercial regulation, the Court has indicated that the Equal Protection Clause "is offended only if the classification rests on grounds wholly irrelevant to the achievement of the State's objective." See, e.g., McGowan v. Maryland, 366 U.S., at 425; Kotch v. Board of River Port Pilot Comm'rs, 330 U.S. 552, 557 (1947). And this lenient standard is further weighted in the State's favor by the fact that "[a] statutory discrimination will not be set aside if any state of facts reasonably may be conceived [by the Court] to justify it." McGowan v. Maryland, supra, at 426. But in Baird the Court clearly did not adhere to these highly tolerant standards of traditional rational review. For although there were conceivable state interests intended to be advanced by the statute—e.g., deterrence of premarital sexual activity and regulation of the dissemination of potentially dangerous articles—the Court was not prepared to accept these interests on their face, but instead proceeded to test their substantiality by independent analysis. See 405 U.S., at 449-454. Such close scrutiny of the State's interests was hardly characteristic of the deference shown state classifications in the context of economic interests. See, e.g., Goesaert v. Cleary, 335 U.S. 464 (1948); Kotch v. Board of River Port Pilot Comm'rs, supra. Yet I think the Court's action was entirely appropriate, for access to and use of contraceptives bears a close relationship to the individual's constitutional right of privacy. See 405 U.S., at 453-454; id., at 463-464 (White, J., concurring in result). See also Roe v. Wade, 410 U.S., at 152-153.

A similar process of analysis with respect to the invidiousness of the basis on which a particular classification is drawn has also influenced the Court as to the '''[p. 105]''' appropriate degree of scrutiny to be accorded any particular case. The highly suspect character of classifications based on race, nationality , or alienage is well established. The reasons why such classifications call for close judicial scrutiny are manifold. Certain racial and ethnic groups have frequently been recognized as "discrete and insular minorities" who are relatively powerless to protect their interests in the political process. See Graham v. Richardson, 403 U.S., at 372; cf. United States v. Carolene Products Co., 304 U.S. 144, 152-153, n. 4 (1938). Moreover, race, nationality, or alienage is "'in most circumstances irrelevant' to any constitutionally acceptable legislative purpose, Hirabayashi v. United States, 320 U.S. 81, 100." McLaughlin v. Florida, 379 U.S., at 192. Instead, lines drawn on such bases are frequently the reflection of historic prejudices rather than legislative rationality. It may be that all of these considerations, which make for particular judicial solicitude in the face of discrimination on the basis of race, nationality, or alienage, do not coalesce—or at least not to the same degree—in other forms of discrimination. Nevertheless, these considerations have undoubtedly influenced the care with which the Court has scrutinized other forms of discrimination.

In James v. Strange, 407 U.S. 128 (1972), the Court held unconstitutional a state statute which provided for recoupment from indigent convicts of legal defense fees paid by the State. The Court found that the statute impermissibly differentiated between indigent criminals in debt to the State and civil judgment debtors, since criminal debtors were denied various protective exemptions '''[p. 106]''' afforded civil judgment debtors. The Court suggested that in reviewing the statute under the Equal Protection Clause, it was merely applying the traditional requirement that there be "'some rationality'" in the line drawn between the different types of debtors. Id., at 140. Yet it then proceeded to scrutinize the statute with less than traditional deference and restraint. Thus the Court recognized "that state recoupment statutes may betoken legitimate state interests" in recovering expenses and discouraging fraud. Nevertheless, Mr. Justice Powell, speaking for the Court, concluded that

"'these interests are not thwarted by requiring more even treatment of indigent criminal defendants with other classes of debtors to whom the statute itself repeatedly makes reference. State recoupment laws, notwithstanding the state interests they may serve, need not blight in such discriminatory fashion the hopes of indigents for self-sufficiency and self-respect.' Id., at 141-142."

The Court, in short, clearly did not consider the problems of fraud and collection that the state legislature might have concluded were peculiar to indigent criminal defendants to be either sufficiently important or at least sufficiently substantiated to justify denial of the protective exemptions afforded to all civil judgment debtors, to a class composed exclusively of indigent criminal debtors.

Similarly, in Reed v. Reed, 404 U.S. 71 (1971), the Court, in striking down a state statute which gave men '''[p. 107]''' preference over women when persons of equal entitlement apply for assignment as an administrator of a particular estate, resorted to a more stringent standard of equal protection review than that employed in cases involving commercial matters. The Court indicated that it was testing the claim of sex discrimination by nothing more than whether the line drawn bore "a rational relationship to a state objective," which it recognized as a legitimate effort to reduce the work of probate courts in choosing between competing applications for letters of administration. Id., at 76. Accepting such a purpose, the Idaho Supreme Court had thought the classification to be sustainable on the basis that the legislature might have reasonably concluded that, as a rule, men have more experience than women in business matters relevant to the administration of estate. 93 Idaho 511, 514, 465 P.2d 635, 638 (1970). This Court, however, concluded that "[t]o give a mandatory preference to members of either sex over members of the other, merely to accomplish the elimination of hearings on the merits, is to make the very kind of arbitrary legislative choice forbidden by the Equal Protection Clause of the Fourteenth Amendment..." 404 U.S., at 76. This Court, in other words, was unwilling to consider a theoretical and unsubstantiated basis for distinction—however reasonable it might appear—sufficient to sustain a statute discriminating on the basis of sex.

James and Reed can only be understood as instances in which the particularly invidious character of the classification caused the Court to pause and scrutinize with more than traditional care the rationality of state discrimination. Discrimination on the basis of past criminality and on the basis of sex posed for the Court the specter of forms of discrimination which it implicitly recognized to have deep social and legal roots without necessarily having any basis in actual differences. Still, '''[p. 108]''' the Court's sensitivity to the invidiousness of the basis for discrimination is perhaps most apparent in its decisions protecting the interests of children born out of wedlock from discriminatory state action. See Weber v. Aetna Casualty & Surety Co., 406 U.S. 164 (1972); Levy v. Louisiana, 391 U.S. 68 (1968).

In Weber, the Court struck down a portion of a state workmen's compensation statute that relegated unacknowledged illegitimate children of the deceased to a lesser status with respect to benefits than that occupied by legitimate children of the deceased. The Court acknowledged the true nature of its inquiry in cases such as these: "What legitimate state interest does the classification promote? What fundamental personal rights might the classification endanger?" Id., at 173. Embarking upon a determination of the relative substantiality of the State's justifications for the classification, the Court rejected the contention that the classifications reflected what might be presumed to have been the deceased's preference of beneficiaries as "not compelling...where dependency on the deceased is a prerequisite to anyone's recovery..." Ibid. Likewise, it deemed the relationship between the State's interest in encouraging legitimate family relationships and the burden placed on the illegitimates too tenuous to permit the classification to stand. Ibid. A clear insight into the basis of the Court's action is provided by its conclusion:

"'[I]mposing disabilities on the illegitimate child is contrary to the basic concept of our system that legal burdens should bear some relationship to individual responsibility or wrongdoing. Obviously, no child is responsible for his birth and penalizing the illegitimate child is an ineffectual—as well as an unjust—way of deterring the parent. Courts are powerless to prevent the social opprobrium suffered by these hapless children, but the Equal Protection [p. 109] Clause does enable us to strike down discriminatory laws relating to status of birth...' Id., at 175-176."

Status of birth, like the color of one's skin, is something which the individual cannot control, and should generally be irrelevant in legislative considerations. Yet illegitimacy has long been stigmatized by our society. Hence, discrimination on the basis of birth—particularly when it affects innocent children—warrants special judicial consideration.

In summary, it seems to me inescapably clear that this Court has consistently adjusted the care with which it will review state discrimination in light of the constitutional significance of the interests affected and the invidiousness of the particular classification. In the context of economic interests, we find that discriminatory state action is almost always sustained, for such interests are generally far removed from constitutional guarantees. Moreover, "[t]he extremes to which the Court has gone in dreaming up rational bases for state regulation in that area may in many instances be ascribed to a healthy revulsion from the Court's earlier excesses in using the Constitution to protect interests that have more than enough power to protect themselves in the legislative halls." Dandridge v. Williams, 397 U.S., at 520 (dissenting opinion). But the situation differs markedly when discrimination against important individual interests with constitutional implications and against particularly disadvantaged or powerless classes is involved. The majority suggests, however, that a variable standard of review would give this Court the appearance of a "superlegislature." Ante, at 31. I cannot agree. Such an approach seems to me a part of the guarantees of our Constitution and of the historic experiences with oppression of and discrimination against discrete, powerless minorities which underlie that document. In truth, '''[p. 110]''' the Court itself will be open to the criticism raised by the majority so long as it continues on its present course of effectively selecting in private which cases will be afforded special consideration without acknowledging the true basis of its action. Opinions such as those in Reed and James seem drawn more as efforts to shield rather than to reveal the true basis of the Court's decisions. Such obfuscated action may be appropriate to a political body such as a legislature, but it is not appropriate to this Court. Open debate of the bases for the Court's action is essential to the rationality and consistency of our decisionmaking process. Only in this way can we avoid the label of legislature and ensure the integrity of the judicial process.

Nevertheless, the majority today attempts to force this case into the same category for purposes of equal protection analysis as decisions involving discrimination affecting commercial interests. By so doing, the majority singles this case out for analytic treatment at odds with what seems to me to be the clear trend of recent decisions in this Court, and thereby ignores the constitutional importance of the interest at stake and the invidiousness of the particular classification, factors that call for far more than the lenient scrutiny of the Texas financing scheme which the majority pursues. Yet if the discrimination inherent in the Texas scheme is scrutinized with the care demanded by the interest and classification present in this case, the unconstitutionality of that scheme is unmistakable.

B
Since the Court now suggests that only interests guaranteed by the Constitution are fundamental for purposes of equal protection analysis, and since it rejects '''[p. 111]''' the contention that public education is fundamental, it follows that the Court concludes that public education is not constitutionally guaranteed. It is true that this Court has never deemed the provision of free public education to be required by the Constitution. Indeed, it has on occasion suggested that state supported education is a privilege bestowed by a State on its citizens. See ''Missouri ex rel. Gaines v. Canada'', 305 U.S., at 349. Nevertheless, the fundamental importance of education is amply indicated by the prior decisions of this Court, by the unique status accorded public education by our society, and by the close relationship between education and some of our most basic constitutional values.

The special concern of this Court with the educational process of our country is a matter of common knowledge. Undoubtedly, this Court's most famous statement on the subject is that contained in Brown v. Board of Education, 347 U.S., at 493:

"'Today, education is perhaps the most important function of state and local governments. Compulsory school attendance laws and the great expenditures for education both demonstrate our recognition of the importance of education to our democratic society. It is required in the performance of our most basic public responsibilities, even service in the armed forces. It is the very foundation of good citizenship. Today it is a principal instrument in awakening the child to cultural values, in preparing him for later professional training, and in helping him to adjust normally to his environment....'"

Only last Term, the Court recognized that "[p]roviding public schools ranks at the very apex of the function of a State." Wisconsin v. Yoder, 406 U.S. 205, 213 (1972). This is clearly borne out by the fact that in 48 '''[p. 112]''' of our 50 States the provision of public education is mandated by the state constitution. No other state function is so uniformly recognized as an essential element of our society's well-being. In large measure, the explanation for the special importance attached to education must rest, as the Court recognized in Yoder, id., at 221, on the facts that "some degree of education is necessary to prepare citizens to participate effectively and intelligently in our open political system...," and that "education prepares individuals to be self-reliant and self-sufficient participants in society." Both facets of this observation are suggestive of the substantial relationship which education bears to guarantees of our Constitution.

Education directly affects the ability of a child to exercise his First Amendment rights, both as a source and as a receiver of information and ideas, whatever interests he may pursue in life. This Court's decision in Sweezy v. New Hampshire, 354 U.S. 234, 250 (1957), speaks of the right of students "to inquire, to study and to evaluate, to gain new maturity and understanding..." Thus, we have not casually described the classroom as the "'marketplace of ideas.'" Keyishian v. Board of Regents, 385 U.S. 589, 603 (1967). The opportunity for formal education may not necessarily be the essential determinant of an individual's ability to enjoy throughout his life the rights of free speech and association '''[p. 113]''' guaranteed to him by the First Amendment. But such an opportunity may enhance the individual's enjoyment of those rights, not only during but also following school attendance. Thus, in the final analysis, "the pivotal position of education to success in American society and its essential role in opening up to the individual the central experiences of our culture lend it an importance that is undeniable."

Of particular importance is the relationship between education and the political process. "Americans regard the public schools as a most vital civic institution for the preservation of a democratic system of government." Abington School Dist. v. Schempp, 374 U.S. 203, 230 (1963) (Brennan, J., concurring). Education serves the essential function of instilling in our young an understanding of and appreciation for the principles and operation of our governmental processes. Education may instill the interest and provide the tools necessary for political discourse and debate. Indeed, it has frequently been suggested that education is the dominant factor affecting political consciousness and participation. A system of "[c]ompetition in ideas and governmental [p. 114] policies is at the core of our electoral process and of the First Amendment freedoms." Williams v. Rhodes, 393 U.S. 23, 32 (1968). But of most immediate and direct concern must be the demonstrated effect of education on the exercise of the franchise by the electorate. The right to vote in federal elections is conferred by Art. I, §2, and the Seventeenth Amendment of the Constitution, and access to the state franchise has been afforded special protection because it is "preservative of other basic civil and political rights," Reynolds v. Sims, 377 U.S., at 562. Data from the Presidential Election of 1968 clearly demonstrates a direct relationship between participation in the electoral process and level of educational attainment ; and, as this Court recognized in Gaston County v. United States, 395 U.S. 285, 296 (1969), the quality of education offered may '''[p. 115]''' influence a child's decision to "enter or remain in school." It is this very sort of intimate relationship between a particular personal interest and specific constitutional guarantees that has heretofore caused the Court to attach special significance, for purposes of equal protection analysis, to individual interests such as procreation and the exercise of the state franchise.

While ultimately disputing little of this, the majority seeks refuge in the fact that the Court has "never presumed to possess either the ability or the authority to guarantee to the citizenry the most effective speech or the most informed electoral choice." Ante, at 36. This serves only to blur what is in fact at stake. With due respect, the issue is neither provision of the most effective speech nor of the most informed vote. Appellees '''[p. 116]''' do not now seek the best education Texas might provide. They do seek, however, an end to state discrimination resulting from the unequal distribution of taxable district property wealth that directly impairs the ability of some districts to provide the same educational opportunity that other districts can provide with the same or even substantially less tax effort. The issue is, in other words, one of discrimination that affects the quality of the education which Texas has chosen to provide its children; and, the precise question here is what importance should attach to education for purposes of equal protection analysis of that discrimination. As this Court held in Brown v. Board of Education, 347 U.S., at 493, the opportunity of education, "where the state has undertaken to provide it, is a right which must be made available to all on equal terms." The factors just considered, including the relationship between education and the social and political interests enshrined within the Constitution, compel us to recognize the fundamentality of education and to scrutinize with appropriate care the bases for state discrimination affecting equality of educational opportunity in Texas' school districts —a conclusion '''[p. 117]''' which is only strengthened when we consider the character of the classification in this case.

C
The District Court found that in discriminating between Texas schoolchildren on the basis of the amount of taxable property wealth located in the district in which they live, the Texas financing scheme created a form of wealth discrimination. This Court has frequently recognized that discrimination on the basis of wealth may create a classification of a suspect character and thereby call for exacting judicial scrutiny. See, e.g., Griffin v. Illinois, 351 U.S. 12 (1956); Douglas v. California, 372 U.S. 353 (1963); McDonald v. Board of Election Commr's of Chicago, 394 U.S. 802, 807 (1969). The majority, however, considers any wealth classification in this case to lack certain essential characteristics which it contends are common to the instances of wealth discrimination that this Court has heretofore recognized. We are told that in every prior case involving a wealth classification, the members of the disadvantaged class have "shared two distinguishing characteristics: because [p. 118] of their impecunity they were completely unable to pay for some desired benefit, and as a consequence, they sustained an absolute deprivation of a meaningful opportunity to enjoy that benefit." Ante, at 20. I cannot agree. The Court's distinctions may be sufficient to explain the decisions in Williams v. Illinois, 399 U.S. 235 (1970); Tate v. Short, 401 U.S. 395 (1971); and even Bullock v. Carter, 405 U.S. 134 (1972). But they are not in fact consistent with the decisions in ''Harper v. Virginia Bd. of Elections, 383 U.S. 663 (1966), or Griffin v. Illinois, supra, or Douglas v. California, supra''.

In Harper, the Court struck down as violative of the Equal Protection Clause an annual Virginia poll tax of $1.50, payment of which by persons over the age of 21 was a prerequisite to voting in Virginia elections. In part, the Court relied on the fact that the poll tax interfered with a fundamental interest—the exercise of the state franchise. In addition, though, the Court emphasized that "[l]ines drawn on the basis of wealth or property...are traditionally disfavored." Id., at 668. Under the first part of the theory announced by the majority, the disadvantaged class in Harper, in terms of a wealth analysis, should have consisted only of those too poor to afford the $1.50 necessary to vote. But the Harper Court did not see it that way. In its view, the Equal Protection Clause "bars a system which excludes [from the franchise] those unable to pay a fee to vote or who fail to pay." Ibid. (Emphasis added.) So far as the Court was concerned, the "degree of the discrimination [was] irrelevant." Ibid. Thus, the Court struck down the poll tax in toto; it did not order merely that those too poor to pay the tax be exempted; complete impecunity clearly was not determinative of the limits of the disadvantaged class, nor was it essential to make an equal protection claim.

'''[p. 119]' Similarly, Griffin and Douglas'' refute the majority's contention that we have in the past required an absolute deprivation before subjecting wealth classifications to strict scrutiny. The Court characterizes Griffin as a case concerned simply with the denial of a transcript or an adequate substitute therefor, and Douglas as involving the denial of counsel. But in both cases the question was in fact whether "a State that [grants] appellate review can do so in a way that discriminates against some convicted defendants on account of their poverty." Griffin v. Illinois, 351 U.S., at 18 (emphasis added). In that regard, the Court concluded that inability to purchase a transcript denies "the poor an adequate appellate review accorded to all who have money enough to pay the costs in advance," ibid. (emphasis added), and that "the type of an appeal a person is afforded...hinges upon whether or not he can pay for the assistance of counsel," Douglas v. California, supra, at 355-356 (emphasis added). The right of appeal itself was not absolutely denied to those too poor to pay; but because of the cost of a transcript and of counsel, the appeal was a substantially less meaningful right for the poor than for the rich. It was on these terms that the Court found a denial of equal protection, and those terms clearly encompassed degrees of discrimination on the '''[p. 120]''' basis of wealth which do not amount to outright denial of the affected right or interest.

This is not to say that the form of wealth classification in this case does not differ significantly from those recognized in the previous decisions of this Court. Our prior cases have dealt essentially with discrimination on the basis of personal wealth. Here, by contrast, the '''[p. 121]''' children of the disadvantaged Texas school districts are being discriminated against not necessarily because of their personal wealth or the wealth of their families, but because of the taxable property wealth of the residents of the district in which they happen to live. The appropriate question, then, is whether the same degree of judicial solicitude and scrutiny that has previously been afforded wealth classifications is warranted here.

As the Court points out, ante, at 28-29, no previous decision has deemed the presence of just a wealth classification to be sufficient basis to call forth rigorous judicial scrutiny of allegedly discriminatory state action. Compare, e.g., Harper v. Virginia Bd. of Elections, supra, with, e.g., James v. Valtierra, 402 U.S. 137 (1971). That wealth classifications alone have not necessarily been considered to bear the same high degree of suspectness as have classifications based on, for instance, race or alienage may be explainable on a number of grounds. The "poor" may not be seen as politically powerless as certain discrete and insular minority groups. Personal poverty may entail much the same social stigma as historically attached to certain racial or ethnic groups. But personal poverty is not a permanent disability; its shackles may be escaped. Perhaps most importantly, though, personal wealth may not necessarily share the general irrelevance as a basis for legislative action that race or nationality is recognized to have. While the "poor" have frequently been a '''[p. 122]''' legally disadvantaged group, it cannot be ignored that social legislation must frequently take cognizance of the economic status of our citizens. Thus, we have generally gauged the invidiousness of wealth classifications with an awareness of the importance of the interests being affected and the relevance of personal wealth to those interests. See ''Harper v. Virginia Bd. of Elections, supra''.

When evaluated with these considerations in mind, it seems to me that discrimination on the basis of group wealth in this case likewise calls for careful judicial scrutiny. First, it must be recognized that while local district wealth may serve other interests, it bears no relationship whatsoever to the interest of Texas school children in the educational opportunity afforded them by the State of Texas. Given the importance of that interest, we must be particularly sensitive to the invidious characteristics of any form of discrimination that is not clearly intended to serve it, as opposed to some other distinct state interest. Discrimination on the basis of group wealth may not, to be sure, reflect the social stigma frequently attached to personal poverty. Nevertheless, insofar as group wealth discrimination involves wealth over which the disadvantaged individual has no significant control, it represents in fact a more serious basis of discrimination than does personal wealth. For such discrimination '''[p. 123]''' is no reflection of the individual's characteristics or his abilities. And thus—particularly in the context of a disadvantaged class composed of children—we have previously treated discrimination on a basis which the individual cannot control as constitutionally disfavored. Cf. Weber v. Aetna Casualty & Surety Co., 406 U.S. 164 (1972); Levy v. Louisiana, 391 U.S. 68 (1968).

The disability of the disadvantaged class in this case extends as well into the political processes upon which we ordinarily rely as adequate for the protection and promotion of all interests. Here legislative reallocation of the State's property wealth must be sought in the face of inevitable opposition from significantly advantaged districts that have a strong vested interest in the preservation of the status quo, a problem not completely dissimilar to that faced by underrepresented districts prior to the Court's intervention in the process of reapportionment, see Baker v. Carr, 369 U.S. 186, 191-192 (1962).

Nor can we ignore the extent to which, in contrast to our prior decisions, the State is responsible for the wealth discrimination in this instance. Griffin, Douglas, Williams, Tate, and our other prior cases have dealt with discrimination on the basis of indigency which was attributable to the operation of the private sector. But we have no such simple de facto wealth discrimination here. The means for financing public education in Texas are selected and specified by the State. It is the State that has created local school districts, and tied educational funding to the local property tax and thereby to local district wealth. At the same time, governmentally '''[p. 124]''' imposed land use controls have undoubtedly encouraged and rigidified natural trends in the allocation of particular areas for residential or commercial use, and thus determined each district's amount of taxable property wealth. In short, this case, in contrast to the Court's previous wealth discrimination decisions, can only be seen as "unusual in the extent to which governmental action is the cause of the wealth classifications."

In the final analysis, then, the invidious characteristics of the group wealth classification present in this case merely serve to emphasize the need for careful judicial scrutiny of the State's justifications for the resulting interdistrict discrimination in the educational opportunity afforded to the schoolchildren of Texas.

D
The nature of our inquiry into the justifications for state discrimination is essentially the same in all equal protection cases: We must consider the substantiality of the state interests sought to be served, and we must scrutinize the reasonableness of the means by which the State has sought to advance its interests. See Police Dept. of Chicago v. Mosley, 408 U.S., at 95. Differences in the application of this test are, in my view, a function of the constitutional importance of the interests at stake and the invidiousness of the particular classification. In terms of the asserted state interests, the Court has indicated that it will require, for instance, a "compelling," Shapiro v. Thompson, 394 U.S., at 634, or a "substantial" '''[p. 125]' or "important," Dunn v. Blumstein'', 405 U.S., at 343, state interest to justify discrimination affecting individual interests of constitutional significance. Whatever the differences, if any, in these descriptions of the character of the state interest necessary to sustain such discrimination, basic to each is, I believe, a concern with the legitimacy and the reality of the asserted state interests. Thus, when interests of constitutional importance are at stake, the Court does not stand ready to credit the State's classification with any conceivable legitimate purpose, but demands a clear showing that there are legitimate state interests which the classification was in fact intended to serve. Beyond the question of the adequacy of the State's purpose for the classification, the Court traditionally has become increasingly sensitive to the means by which a State chooses to act as its action affects more directly interests of constitutional significance. See, e.g., United States v. Robel, 389 U.S. 258, 265 (1967); Shelton v. Tucker, 364 U.S. 479, 488 (1960). Thus, by now, "less restrictive alternatives" analysis is firmly established in equal protection jurisprudence. See Dunn v. Blumstein, supra, at 343; Kramer v. Union School District, 395 U.S., at 627. It seems to me that the range of choice we are willing to accord the State in selecting the means by which it will act, and the care with which we scrutinize the effectiveness of the means which the State selects, also must reflect the constitutional importance of the interest affected and the invidiousness of the particular classification. Here, both the nature of the interest and the classification dictate close judicial scrutiny of the purposes which Texas seeks to serve with its present educational financing '''[p. 126]''' scheme and of the means it has selected to serve that purpose.

The only justification offered by appellants to sustain the discrimination in educational opportunity caused by the Texas financing scheme is local educational control. Presented with this justification, the District Court concluded that "[n]ot only are defendants unable to demonstrate compelling state interests for their classifications based upon wealth, they fail even to establish a reasonable basis for these classifications." 337 F.Supp., at 284. I must agree with this conclusion.

At the outset, I do not question that local control of public education, as an abstract matter, constitutes a very substantial state interest. We observed only last Term that "[d]irect control over decisions vitally affecting the education of one's children is a need that is strongly felt in our society." Wright v. Council of the City of Emporia, 407 U.S. 451, 469 (1972). See also id., at 477-478 (Burger, C.J., dissenting). The State's interest in local educational control—which certainly includes questions of educational funding—has deep roots in the inherent benefits of community support for public education. Consequently, true state dedication to local control would present, I think, a substantial justification to weigh against simply interdistrict variations in the treatment of a State's school children. But I need not now decide how I might ultimately strike the balance were we confronted with a situation where the State's sincere concern for local control inevitably produced educational inequality. For on this record, it is apparent that the State's purported concern with local control is offered primarily as an excuse rather than as a justification for interdistrict inequality.

In Texas, statewide laws regulate in fact the most minute details of local public education. For example, '''[p. 127]''' the State prescribes required courses. All textbooks must be submitted for state approval, and only approved textbooks may be used. The State has established the qualifications necessary for teaching in Texas public schools and the procedures for obtaining certification. The State has even legislated on the length of the school day. Texas' own courts have said:

"'As a result of the acts of the Legislature our school system is not of mere local concern but it is statewide. While a school district is local in territorial limits, it is an integral part of the vast school system which is coextensive with the confines of the State of Texas.' Treadaway v. Whitney Independent School District, 205 S.W.2d 97, 99 Tex. Ct. Civ. App. 1947)."

See also El Dorado Independent School District v. Tisdale, 3 S.W.2d 420, 422 (Tex. Comm'n App. 1928).

Moreover, even if we accept Texas' general dedication to local control in educational matters, it is difficult to find any evidence of such dedication with respect to fiscal matters. It ignores reality to suggest—as the Court does, ante, at 49-50—that the local property tax element of the Texas financing scheme reflects a conscious legislative effort to provide school districts with local fiscal control. If Texas had a system truly dedicated to local fiscal control, one would expect the quality of the educational opportunity provided in each district to vary with the decision of the voters in that district as '''[p. 128]''' to the level of sacrifice they wish to make for public education. In fact, the Texas scheme produces precisely the opposite result. Local school districts cannot choose to have the best education in the State by imposing the highest tax rate. Instead, the quality of the educational opportunity offered by any particular district is largely determined by the amount of taxable property located in the district—a factor over which local voters can exercise no control.

The study introduced in the District Court showed a direct inverse relationship between equalized taxable district property wealth and district tax effort with the result that the property-poor districts making the highest tax effort obtained the lowest per-pupil yield. The implications of this situation for local choice are illustrated by again comparing the Edgewood and Alamo Heights School Districts. In 1967-1968, Edgewood, after contributing its share to the Local Fund Assignment, raised only $26 per pupil through its local property tax, whereas Alamo Heights was able to raise $333 per pupil. Since the funds received through the Minimum Foundation School Program are to be used only for minimum professional salaries, transportation costs, and operating expenses, it is not hard to see the lack of local choice -- with respect to higher teacher salaries to attract more and better teachers, physical facilities, library books, and facilities, special courses, or participation in special state and federal matching funds programs—under which a property-poor district such as Edgewood is forced to labor. In fact, because of the difference in taxable local property wealth, Edgewood would have to tax itself almost nine times as heavily to obtain the same '''[p. 129]''' yield as Alamo Heights. At present, then, local control is a myth for many of the local school districts in Texas. As one district court has observed, "rather than reposing in each school district the economic power to fix its own level of per pupil expenditure, the State has so arranged the structure as to guarantee that some districts will spend low (with high taxes) while others will spend high (with low taxes)." Van Dusartz v. Hatfield, 334 F.Supp. 870, 876 (Minn. 1971).

In my judgment, any substantial degree of scrutiny of the operation of the Texas financing scheme reveals that the State has selected means wholly inappropriate to secure its purported interest in assuring its school districts local fiscal control. At the same time, appellees have pointed out a variety of alternative financing schemes which may serve the State's purported interest in local control as well as, if not better than, the present scheme without the current impairment of the educational opportunity of vast numbers of Texas schoolchildren. I see no need, however, to explore the practical or constitutional merits of those suggested alternatives at this time for, whatever their positive or negative features, experience '''[p. 130]''' with the present financing scheme impugns any suggestion that it constitutes a serious effort to provide local fiscal control. If, for the sake of local education control, this Court is to sustain interdistrict discrimination in the educational opportunity afforded Texas school children, it should require that the State present something more than the mere sham now before us.

III
In conclusion, it is essential to recognize that an end to the wide variations in taxable district property wealth inherent in the Texas financing scheme would entail none of the untoward consequences suggested by the Court or by the appellants.

First, affirmance of the District Court's decisions would hardly sound the death knell for local control of education. It would mean neither centralized decision making nor federal court intervention in the operation of public schools. Clearly, this suit has nothing to do with local decisionmaking with respect to educational policy or even educational spending. It involves only a narrow aspect of local control—namely, local control over the raising of educational funds. In fact, in striking down interdistrict disparities in taxable local wealth, the District Court took the course which is most likely to make true local control over educational decisionmaking a reality for all Texas school districts.

Nor does the District Court's decision even necessarily eliminate local control of educational funding. The District Court struck down nothing more than the continued interdistrict wealth discrimination inherent in the present property tax. Both centralized and decentralized plans for educational funding not involving such interdistrict discrimination have been put forward. The choice '''[p. 131]''' among these or other alternatives would remain with the State, not with the federal courts. In this regard, it should be evident that the degree of federal intervention '''[p. 132]''' in matters of local concern would be substantially less in this context than in previous decisions in which we have been asked effectively to impose a particular scheme upon the States under the guise of the Equal Protection Clause. See, e.g., Dandridge v. Williams, 397 U.S. 471 (1970); cf. Richardson v. Belcher, 404 U.S. 78 (1971).

Still, we are told that this case requires us "to condemn the State's judgment in conferring on political subdivisions the power to tax local property to supply revenues for local interests." Ante, at 40. Yet no one in the course of this entire litigation has ever questioned the constitutionality of the local property tax as a device for raising educational funds. The District Court's decision, at most, restricts the power of the State to make educational funding dependent exclusively upon local property taxation so long as there exists interdistrict disparities in taxable property wealth. But it hardly eliminates the local property tax as a source of educational funding or as a means of providing local fiscal control.

The Court seeks solace for its action today in the possibility of legislative reform. The Court's suggestions of legislative redress and experimentation will doubtless be of great comfort to the schoolchildren of Texas' disadvantaged districts, but considering the vested interests of wealthy school districts in the preservation of the status quo, they are worth little more. The possibility of legislative action is, in all events, no answer to this Court's duty under the Constitution to eliminate unjustified state discrimination. In this case we have been presented with an instance of such discrimination, in a particularly invidious form, against an individual interest of large constitutional and practical importance. To support the demonstrated discrimination in the provision '''[p. 133]''' of educational opportunity the State has offered a justification which, on analysis, takes on at best an ephemeral character. Thus, I believe that the wide disparities in taxable district property wealth inherent in the local property tax element of the Texas financing scheme render that scheme violative of the Equal Protection Clause.

I would therefore affirm the judgment of the District Court.

[Appendices I-IV are on immediately following pages.] '''[p. 134]'''

Appendix I to Opinion of Marshall, J., Dissenting
Based on Table V to affidavit of Joel S. Berke, App. 208, which was prepared on the basis of a sample of 110 selected Texas school districts from data for the 1967-1968 school year. '''[p. 135]'''

Appendix II to Opinion of Marshall, J., Dissenting
Based on Table II to affidavit of Joel S. Berke, App. 205, which was prepared on the basis of a sample of 110 selected Texas school districts from data for the 1967-1968 school year. '''[p. 136]'''

Appendix III to Opinion of Marshall, J., Dissenting
Based on Table XI to affidavit of Joel S. Berke, App. 220, which was prepared on the basis of a sample of six selected school districts located in Bexar County, Texas, from data for the 1967-1968 school year. '''[p. 137]'''

Appendix IV to Opinion of Marshall, J., Dissenting
Based on Table IX to affidavit of Joel S. Berke, App. 218, which was prepared on the basis of the 12 school districts located in Bexar County, Texas, from data from the 1967-1968 school year.