Westbrook v. Mihaly

2 Cal.3d 765 (1970) 471 P.2d 487 87 Cal. Rptr. 839

ELOUISE WESTBROOK et al., Petitioners,
v.
EMMERY MIHALY, as Registrar of Voters, etc., et al., Respondents. STEPHEN ADAMS, Petitioner,
v.
EMMERY MIHALY, as Registrar of Voters, etc., et al., Respondents. (Two Cases.)

Docket Nos. S.F. 22706, 22707.

Supreme Court of California. In Bank.

June 30, 1970.

COUNSEL

Stephen Adams, in pro. per., Adams & Adams, Philip Adams, Joseph L. Alioto, Peter J. Donnici, Lawrence E. Alioto and Joseph M. Alioto for Petitioners.

Paul N. Halvonik and Charles C. Marson as Amici Curiae on behalf of Petitioners in Nos. 22706 and 22707.

Donald L. Ungar as Amicus Curiae on behalf of Petitioners in No. 22706.

Thomas M. O'Connor, City Attorney, and Thomas J. Blanchard, Chief Deputy City Attorney, for Respondents.

Wilson, Jones, Morton & Lynch, Ernest A. Wilson, Pillsbury, Madison & Sutro, Francis R. Kirkham, Francis N. Marshall and Noble K. Gregory as Amici Curiae on behalf of Respondents.

OPINION

SULLIVAN, J.

We are presented in these cases with a common issue: whether that portion of former article XI, section 18 (present art. XIII, § 40) of the California Constitution which requires that general obligation bond proposals of counties, cities and school districts be approved by a two-thirds majority of the voters in a popular referendum violates the equal protection clause of the Fourteenth Amendment to the United States Constitution. The challenged section provides, in relevant part, "No county, city, town, township, board of education, or school district, shall incur any indebtedness or liability in any manner or for any purpose exceeding in any year the income and revenue provided for such year, without the assent of two-thirds of the qualified electors thereof, voting at an election to be held for that purpose, ..."^[1]

Two bond issue proposals were submitted to the voters of San Francisco at a special election held in November 1969. Proposition "A" sought authorization for the City and County of San Francisco^[2] to incur a bonded indebtedness of $9,998,000 for additions to and improvements of its park and recreation system. Proposition "B" sought authorization for a bonded indebtedness in the amount of $5,000,000 to provide new elementary schools and to modernize existing school facilities in the Hunters Point area of the city. Both propositions received the approval of a majority of the voters but less than the two-thirds required by article XI, section 18.^[3]

Petitioners in S.F. 22706 are residents of and registered voters in San Francisco. Some of them are property owners; others are parents of children attending public schools in the Hunters Point neighborhood. All voted in favor of both Proposition "A" and Proposition "B." Petitioner in S.F. 22707 is a resident of and registered voter in San Francisco who also voted in favor of both propositions. Respondents in both cases are various officials of the city and county who are charged with certain duties in connection with municipal bond elections.^[4] Through their respective attorneys, petitioners demanded of respondents that they certify both propositions as having been approved by the voters and that they take all further steps necessary to offer the bonds for sale. Respondents refused, taking the position that, since neither proposition had received the assent of two-thirds of the voters as required by article XI, section 18, they had no duties to perform in connection therewith other than that of certifying the actual results of the election to the San Francisco Board of Supervisors.

Petitioners then commenced these proceedings contending that the two-thirds vote requirement, by giving to each negative voter twice the voting power of each affirmative voter, substantially diminishes the effect of the votes of all persons who, like petitioners, favor passage of propositions authorizing the incurring of bonded indebtedness. Such "dilution" of voting power, it is claimed, denies their right to the equal protection of the laws. Invoking our original jurisdiction (Cal. Const., art. VI, § 10), petitioners in each case seek: (1) a declaration that article XI, section 18 of the state Constitution, insofar as it requires a two-thirds rather than a simple mathematical majority to approve the incurring of bonded indebtedness, is invalid under the Fourteenth Amendment to the United States Constitution; and (2) a peremptory writ of mandate commanding respondents to certify that the propositions at issue herein were duly approved and authorized by the qualified voters of the City and County of San Francisco and to proceed with all actions necessary to offer the bonds for sale. We issued alternative writs of mandate. Respondents have made returns thereto by demurrer and answer. (Code Civ. Proc., § 1089; Cal. Rules of Court, rule 56(c).) (1) By issuing the alternative writs we have necessarily determined that there is no adequate remedy in the ordinary course of law and that each case is a proper one for the exercise of our original jurisdiction. (County of Sacramento v. Hickman (1967) 66 Cal.2d 841, 845 [59 Cal. Rptr. 609, 428 P.2d 593]; see Cal. Rules of Court, rule 56(a).)

This court has considered the scope and applicability of article XI, section 18 on several occasions since it became a part of our state Constitution in 1879. In this case, however, we face for the first time a challenge to the provision on federal constitutional grounds. (2) When, as here, a state law is claimed to violate the equal protection clause, we must consider "the facts and circumstances behind the law, the interests which the State claims to be protecting, and the interests of those who are disadvantaged by the classification." (Fn. omitted.) (Williams v. Rhodes (1968) 393 U.S. 23, 30 [21 L.Ed.2d 24, 31, 89 S.Ct. 5].) We begin with a synopsis of the events leading up to the adoption of the challenged provision.

The California Constitution of 1849 contained no debt limits applicable to cities or other local governmental units. However, it did require that any state indebtedness in excess of $300,000 be approved by a majority of the voters in a statewide referendum (Cal. Const. of 1849, art VIII)^[5] and authorized the state Legislature to organize cities and restrict their powers of "... borrowing money, contracting debts, and loaning their credit, so as to prevent abuses...." (Cal. Const. of 1849, art. IV, § 37.) The Legislature, however, exercised this power only sporadically and such restrictions on municipal indebted as did exist were contained in the charters of individual cities rather than in general laws.^[6] Indeed, state regulation typically took the form of numerous special acts authorizing particular cities to incur specific indebtedness, often in total disregard of the local charter restrictions.^[7] In California, as elsewhere, municipal debt increased dramatically.^[8]

The bubble burst in the 1870's as California followed the rest of the nation into a severe financial depression. Banks collapsed, the securities markets declined, municipalities defaulted on their bonds, businesses closed and trade stagnated.^[9] As foreclosures and unemployment mounted, thousands drifted from the farms to the cities vainly seeking work.^[10] The general mood was one of disillusionment and anger and the state was swept by radical political movements. It was thus in an atmosphere of economic and political crisis that the delegates to the Constitutional Convention set to work in 1878.^[11]

A combination of state and local mismanagement, aggravated by the depression, had created widespread concern over excessive municipal indebtedness and the processes of local debt formation.^[12] Popular dissatisfaction had begun to focus on the special legislation authorizing local indebtedness since it was felt that this gave undue influence to narrow interest groups seeking public financing of projects from which they would reap disproportionate benefits. The practice was considered to be a major cause of the inordinate mass of debt with which many communities were burdened.^[13] It is, therefore, not surprising that when the delegates confronted the problem of regulating municipal indebtedness, one of their solutions was to strip the state Legislature of any role in its creation. This, inter alia, was accomplished by the absolute ban on special legislation contained in article IV, section 25 of the new Constitution.^[14]

Article XI, section 18, however, provided the principal instrument for the control of local borrowing. It is apparent, from a reading of the section, that one of its purposes was to establish the "pay as you go" principle as a cardinal rule of municipal finance. Accordingly, the section requires each year's expenditures to be satisfied from the income of that year, thus preventing a gradual year-by-year amassment of liabilities. As a result, the actual cost of municipal government is more closely reflected in the tax rate and the citizens are better able to make informed judgments on the performance of their elected officials. San Francisco Gas Co. v. Brickwedel (1882) 62 Cal. 641, the first case in which this court considered article XI, section 18, mentioned only this "pay as you go" restriction^[15] of the section, which it termed a "radical change" from previous conditions. This restriction is manifestly distinct from that portion of the section which establishes the percentage of voter approval requisite to authorize indebtedness in excess of current revenues.

It is also apparent that article XI, section 18 was intended to compel local legislative bodies to inform the public of projects necessitating long-term expenditures and to give the people the ultimate power of approving or rejecting them. However, we have been unable to discover from the records of the Constitutional Convention or from contemporary judicial opinions, why a two-thirds vote requirement, rather than that of a simple majority, was adopted.^[16] It may have been seen as a method by which most long-term indebtedness could be entirely prevented. (See Starner, op. cit. supra, at p. 7.) It may have been designed to protect property owners from the unrestrained desires of the landless for publicly financed projects. It may have reflected an unarticulated premise that only those proposals which could command such substantial consensus should be undertaken. It would be mere speculation on this record to designate one purpose as that for which the two-thirds figure was selected. There is, of course, no evidence indicating that the requirement was designed as a subterfuge for discriminating against the interests of racial, national or religious minorities. This, however, is by no means the end of our inquiry.

Nor do we consider this matter concluded by any of the cases from this and other jurisdictions to which respondents have referred us. While the cited cases rejected various attacks on state debt limitations, none has decided the question here presented.

Finally, we do not believe that, because an equal protection argument was not made by any of the parties in the cases upholding debt limitations, a challenge to article XI, section 18 on those grounds lacks merit today. It is not surprising that there was apparently no discussion of the possible conflict between a two-thirds vote requirement and the equal protection clause during the Constitutional Convention. Nor is it remarkable that the provision was not assailed on these grounds in the courts. The period in which section 18 was adopted was not greatly concerned about inequalities in the electoral process.^[17] For example even though the Fifteenth Amendment, forbidding denial or abridgement of the right to vote on racial grounds, had been ratified only eight years earlier, most of the delegates to the Constitutional Convention believed that Californians of Chinese ancestry should not be permitted to vote and only grudgingly limited the exclusion to "natives of China." (Gaylord, "History of the California Election Laws," West's Election Code, p. 41.) Much water has gone under the bridge since the 1870's and we may not resolve this problem by applying the standards of that era. (See, e.g., Brown v. Board of Education (1954) 347 U.S. 483, 492 [98 L.Ed. 873, 879, 74 S.Ct. 686, 38 A.L.R.2d 1180].) "Our understanding and conception of the rights guaranteed to the people by the ``stately admonitions' of the Fourteenth Amendment have deepened, and have resulted in a series of decisions, enriching the quality of our democracy, which certainly do not codify State's rights, governmental theories or conceptions of human liberties" as they existed in the nineteenth century. (Fn. omitted.) (Fortson v. Morris (1966) 385 U.S. 231, 247 [17 L.Ed.2d 330, 340, 87 S.Ct. 446]; Fortas, J., dissenting.) (3) Our task is to decide whether "the challenged provision is compatible with the demands of equal protection as they apply in contemporary society." (Castro v. State of California (1970) ante, pp. 223, 231 [85 Cal. Rptr. 20, 466 P.2d 244].) This means, since "notions of what constitutes equal treatment for purposes of the Equal Protection Clause do change" (Harper v. Virginia Board of Elections (1966) 383 U.S. 663, 669 [16 L.Ed.2d 169, 174, 86 S.Ct. 1079]), that the constitutionality of article XI, section 18 must be determined not only in light of the need for and effect of the provision under present conditions, but also in light of the greatly expanded protection we have afforded to the right to vote.

More than any decade since that immediately following the Civil War, the 1960's were a time of struggle to transform America's ancient promise of political equality into a reality. Ten years ago many of our citizens were barred from the polls for reasons related to their race, economic status or national background. Others, because of extreme disparities in population among electoral districts, were denied equal representation in Congress, in state legislatures, and in units of local government. Much of this has been changed, some by federal legislation^[18] and some by constitutional amendment.^[19] But the significant and stirring reforms have been effectuated by the United States Supreme Court which has insisted that our political structures and practices, no matter how venerable their origin, comport with the Fourteenth Amendment.

The court has entered the "political thicket" (Colegrove v. Green (1946) 328 U.S. 549, 556 [90 L.Ed. 1432, 1436, 66 S.Ct. 1198]) to enforce the equal protection clause in three principal contexts. The first such category is composed of cases involving state laws that exclude various groups from voting in all or in certain types of elections. The court has eliminated many of these barriers to the electoral process; it has invalidated residency requirements excluding servicemen,^[20] qualifying examinations excluding racial minorities,^[21] poll taxes excluding the poor and indigent,^[22] and property qualifications excluding the landless.^[23] The effect of these, and of related decisions,^[24] has been to expand sharply the number of persons entitled to participate actively in the election process. This court, for example, complying with principles established by the above line of cases, has limited the application of two voter qualification provisions contained in our state Constitution.^[25]

The second major area in which the high court has intervened involves state geographical districting systems which dilute the effectiveness of the franchise either directly or through the allocation of legislative representation. In order to insure that voting power remains precisely correlated with population, the court has abolished electoral systems which weighted votes in elections for statewide office;^[26] it has ordered that the districting of federal congressional seats^[27] and the apportionment of both houses of state legislatures^[28] be conducted on the basis of substantial population equality among such units. Subsequent cases have insisted upon a rule of absolute equality of population^[29] and have extended the principle of "one man, one vote" to various units of local government.^[30] The result of these decisions has been to equalize the effectiveness of participation in the political process.^[31]

The third category of which we speak is less neatly defined. In a general way, it deals with the extent to which a state, through the political process, may impose on the interests of one group burdens which are significantly more onerous than those it imposes on similar interests of other groups. In two cases decided last term, the high court struck down state and local laws placing unusual obstacles in the path of certain political and racial groups.^[32]

In our view, these three groups of decisions fashion the doctrinal structure and delineate the basic principles that must govern our decision in the case at bench. Turning now to the issue before us, we conceive the broad question to be whether California, after having directed that certain matters of local governmental policy be submitted to the people, may require the outcome to be determined by a decisional process other than that of a simple majority vote.^[33]

Petitioners contend that it may not. Their position may be fairly summarized as follows: article XI, section 18, classifies voters on the basis of the manner in which they vote and, by requiring a two-thirds majority, patently discriminates against those who vote in favor of bond issue propositions. This dilution of voting power denies such voters equal protection of laws, absent a showing that the provision is necessary to promote a compelling state interest. While economic conditions in the late Nineteenth Century may have justified such a provision, today it is no longer necessary to impair the right to vote in order to guarantee responsible municipal financial policies.

Respondents' position is more elaborate. First, respondents contend, for various reasons, that the equal protection clause has no application to this case. Second, assuming that the equal protection clause does apply, respondents urge, again for a variety of reasons, that we do not adopt the stringent standard of review for which petitioners argue, but instead apply the traditional equal protection test and inquire merely whether the two-thirds vote requirement bears any rational relationship to a valid state purpose which, they maintain, it does. Finally, assuming that the stricter standard of judicial review is applied, respondents assert that the constitutional provision is necessary to achieve a compelling state interest. We shall take up each contention in turn.

I

Respondents argue that petitioners' reliance on the equal protection clause is misplaced; three reasons are advanced in support of this position.

First, it is contended that since the purpose of the equal protection clause is to regulate the manner in which our laws classify citizens, it can have no application here because article XI, section 18, draws no lines between groups and creates no classifications. We do not agree. It appears obvious to us that section 18 implicitly creates two classes of voters: those who favor a proposed bond issue and those who oppose it. (4) It is not essential that a classification appear in explicit terms on the face of the law. Nor is it necessary that those who are allegedly disadvantaged by the classification either constitute a class which is "objectively identifiable" prior to the election or that they share in common other characteristics, such as race, economic status or residence. (5a) It is sufficient that the class emerge from its inchoate state at the time of the election and that it be defined by the act of voting affirmatively on a bond issue proposition. (6) The equal protection clause extends to the shared political interests of groups otherwise random and diverse. (Williams v. Rhodes, supra, 393 U.S. 23 [21 L.Ed.2d 24]; see also Carrington v. Rash, supra, 380 U.S. 89, 94 [13 L.Ed.2d 675, 679]: "``Fencing out' from the franchise a sector of the population because of the way it may vote is constitutionally impermissible." (Italics added.)^[34]

(5b) Next, we are advised that because section 18 deprives no one of his right to vote on bond propositions and because each vote is "fully counted," there is no discrimination, invidious or otherwise, against which the equal protection clause historically has been directed. It is always easier to assume away a problem than to resolve it. However, we think respondents' suggested assumption distorts reality. It does not require extended discussion to establish that the inevitable result of any extraordinary majority requirement is to give to one group of voters a greater influence on the outcome of an election than to another group of comparable size but opposite conviction. This is the functional equivalent of according to each person in the former group a vote of relatively greater weight precisely because of his position on the issue which is to be decided. In other words, a result identical to that produced by a two-thirds vote requirement could be achieved by requiring only a 50 percent affirmative vote for passage of bond issue propositions but providing that in tabulating the election results each negative vote would be multiplied by two while affirmative votes would be counted at face value. Alternatively, those citizens who opposed a measure could be permitted to vote twice while those who favored it were limited to only one ballot. The biased structure of both of these voting schemes is perhaps more obvious than the discrimination inherent in an extraordinary majority vote requirement. The effect of each, of course, is identical. "If a State in a statewide election weighted the male vote more heavily than the female vote or the white vote more heavily than the Negro vote, none could successfully contend that that discrimination was allowable. [Citation.]" (Gray v. Sanders, supra, 372 U.S. 368, 379 [9 L.Ed.2d 821, 829].)^[35] Although the question of whether such discrimination is allowable has not been settled by the Fifteenth or the Nineteenth Amendment, we think it equally apparent that when a state weights the negative vote more heavily than the affirmative vote, no one can successfully contend that there is no discrimination.

(7) Respondents, however, are unwilling to concede that article XI, section 18 (despite its creation of a classification with adverse consequences for those within it) constitutes governmental action to which the equal protection clause is applicable. They buttress this position primarily by the observation that there is no federal constitutional right in the people of a municipality or other local governmental entity to vote on such a matter as the incurring of bonded indebtedness. We do not understand petitioners to dispute this or to claim an inherent right to vote upon all issues concerning governmental expenditures. We emphatically reject, however, the inference, which respondents seek to draw from this initial premise, that voters have no right to equal protection in bond issue elections. To agree that there is no constitutionally granted right to vote on a particular issue is not to concede that such elections as may be held are therefore beyond the restraints of the United States Constitution. We are aware of no principle which required the California Constitution to provide that certain issues of local finance were to be entrusted to the voters. (See In re Pfahler (1906) 150 Cal. 71, 79 [88 P. 270].) Since it has done so, however, the voting procedures established must comply with equal protection. "[O]nce the franchise is granted to the electorate, lines may not be drawn which are inconsistent with the Equal Protection Clause of the Fourteenth Amendment." (Harper v. Virginia Board of Elections, supra, 383 U.S. at p. 665 [16 L.Ed.2d at p. 171].)

(8) Respondents also point out that there is no constitutional right to incur indebtedness. This may be correct; it is also wholly irrelevant since the requirements of due process and equal protection apply to governmental acts regardless of whether or not they affect rights specifically guaranteed by other constitutional provisions. (See, e.g., Griffin v. Illinois (1956) 351 U.S. 12 [100 L.Ed. 891, 76 S.Ct. 585]; Wieman v. Updegraff (1952) 344 U.S. 183 [97 L.Ed. 216, 73 S.Ct. 215].)

II

We now turn to consider what should be our standard of review in evaluating the classification created by article XI, section 18.

(9) As this court has previously noted,^[36] the United States Supreme Court has tended to employ a two-level test in reviewing legislative classifications under the equal protection clause. In the area of economic regulation, the high court has exercised restraint, investing legislation with a presumption of constitutionality and requiring merely that distinctions drawn by a challenged statute bear some rational relationship to a conceivable legitimate state purpose. (See McDonald v. Board of Election (1969) 394 U.S. 802, 809 [22 L.Ed.2d 739, 745-746, 89 S.Ct. 1404]; McGowan v. Maryland (1961) 366 U.S. 420, 425-426 [6 L.Ed.2d 393, 398-399, 81 S.Ct. 1101].)

(10) On the other hand, in cases involving "suspect classifications"^[37] or touching on "fundamental interests,"^[38] the court has adopted an attitude of active and critical analysis, subjecting the classification to strict scrutiny. (See Shapiro v. Thompson, supra, 394 U.S. 618, 638 [22 L.Ed.2d 600, 617]; Sherbert v. Verner (1963) 374 U.S. 398, 406 [10 L.Ed.2d 965, 971-972, 83 S.Ct. 1790]; Skinner v. Oklahoma, supra, 316 U.S. 535, 541 [86 L.Ed. 1655, 1660]; see also Developments in the Law — Equal Protection (1969) 82 Harv.L.Rev. 1064, 1120-1131.) Under the strict standard applied in such cases, the state bears the burden of establishing not only that it has a compelling interest which justifies the law but that the distinctions drawn by the law are necessary to further its purpose. The dual aspect of this standard is clearly formulated in the opinion of Justice Marshall, dissenting, in Hall v. Beals (1969) 396 U.S. 45, 52 [24 L.Ed.2d 214, 220, 90 S.Ct. 200]: "... once a State has determined that a decision is to be made by popular vote, it may exclude persons from the franchise only upon a showing of a compelling interest, and even then only when the exclusion is the least restrictive method of achieving the desired purpose. [Citations.]"

As the above quotation indicates the United States Supreme Court has included voting in the category of rights deemed "fundamental." Accordingly, state laws which permit or require differential treatment of those attempting to exercise this right have been "carefully and meticulously scrutinized" (Reynolds v. Sims, supra, 377 U.S. 533, 562 [12 L.Ed.2d 506, 527]) in order to determine whether the burdens imposed are "necessary to promote a compelling state interest." (Kramer v. Union School District, supra, 395 U.S. 621, 626-627 [23 L.Ed.2d 583, 589-590]; see also Cipriano v. City of Houma, supra, 395 U.S. 701; Castro v. State of California, supra, ante, p. 223; Otsuka v. Hite, supra, 64 Cal.2d 596, 602.)

Respondents, however, contend that the reapportionment and voting rights cases are not controlling and urge us to apply the more lenient standard, traditionally associated with review of fiscal and economic regulatory measures, in evaluating petitioners' equal protection claims. Two theories are advanced in support of this contention. Neither, we believe, has merit.

(11) First, it is argued that the right to vote (which is termed a "political" right) is not involved in this case, and that the reapportionment and voting rights cases, which did concern this "political" right, are inapposite. We are told that a vote in a bond issue election is merely a "procedural" step in an "administrative" process of authorizing a particular form of indebtedness. This, in turn, is said to be only an "economic" matter. We do not believe it would be either sensible or responsible for our decision to turn on whether bond issue elections are characterized as "political" or "procedural and administrative."

Voting for or against a candidate for public office is obviously not precisely identical to voting for or against a ballot proposition authorizing a local legislative body to incur a specific indebtedness to finance a particular project. But neither is it wholly dissimilar. Respondents have suggested no reasons (other than their conclusionary definitional distinctions) for us to ignore the salient fact that what might be termed an "administrative" decision is made, in a bond election, by a political method closely resembling direct democracy. Nor have any reasons been presented which indicate less need for judicial scrutiny in the context of such elections. Indeed, so far as appears, all that underlies this argument is the inaccurate observation that the Supreme Court has articulated the higher equal protection standard only in cases involving elections for public office. However, as the court itself has indicated, it is the nature of the right asserted, not the type of election, which is the critical determinant.^[39]

In any event, the argument has been foreclosed by Cipriano v. City of Houma, supra, 395 U.S. 701. Cipriano concerned an election whose purpose was not to select among candidates for a public office, but, as in the case at hand, to determine whether or not a proposed local government bond issue should be approved.^[40] The court perceived no distinction between the election in Cipriano and that in Kramer, which was an election to choose representatives on a policymaking body. Applying the identical standard, the court held that the exclusion of nonproperty owners violated equal protection because it had not been shown necessary to promote a compelling state interest. (395 U.S. at p. 706 [23 L.Ed.2d at pp. 651-652].)

(12) We find respondents' second argument no more persuasive. Pointing out that, unlike Harper, Kramer, and Cipriano, this case does not involve an absolute denial of the franchise, respondents contend that under McDonald v. Board of Elections, supra, 394 U.S. 802, the traditional equal protection test should be applied. Respondents' reliance on McDonald is misplaced. As the court noted in Kramer, the right to vote was not at issue in McDonald, which involved merely a "claimed right to an absentee ballot." (395 U.S. at p. 627, fn. 6 [23 L.Ed.2d at p. 589].) We have found no decision of the United States Supreme Court intimating that states may restrict or infringe voting rights, upon a mere showing of some rational relationship to a valid purpose, so long as they stop short of outright denial. To the contrary, the language of the relevant decisions is deliberately broad and encompassing. For example, in Harper, supra, the court declared: "We have long been mindful that where fundamental rights and liberties are asserted under the Equal Protection Clause, classifications which might invade or restrain them must be closely scrutinized and carefully confined. [Citations.]" (Italics added.) (383 U.S. at p. 670 [16 L.Ed.2d at p. 174].) In Williams v. Rhodes, supra, 393 U.S. 23 [21 L.Ed.2d 24], discussing state laws which imposed "burdens" on the right to vote, the court stated: "... the decisions of this Court have consistently held that ``only a compelling state interest....'" could justify them. (393 U.S. at p. 31 [21 L.Ed.2d at p. 32].) In short, state laws denying or diluting the right to vote are tested by the same stern standard. (See Kramer, supra, stating this proposition and citing reapportionment and voter qualification cases interchangeably.)

We are, therefore, satisfied that article XI, section 18, of the California Constitution is valid under the equal protection clause of the Fourteenth Amendment if, and only if, it can be shown necessary to promote a compelling state interest.^[41]

III

A.

(13) Primarily, respondents justify the two-thirds vote requirement as a protective device necessary to insure "municipal solvency," promote "fiscal responsibility," and forestall "ill-advised and excessive indebtedness."^[42] Stripped of rhetoric, their contention is that it is necessary to permit a minority to hold veto power over proposed bond issues approved by a majority of the local legislative body and a majority of their fellow citizens in order to prevent such bonded indebtedness from reaching levels that would necessitate either foreclosure or repudiation. Respondents imply that the two-thirds vote requirement is all that stands between the state and either immediate financial catastrophe^[43] or ultimate collapse of its political subdivisions under crushing burdens of debt foolishly incurred by reckless or malevolent popular majorities. There is no support for respondents' position. On the other hand, we find persuasive petitioners' argument that it is no longer necessary to restrict citizens' voting rights in order to protect society from the financially irresponsible accumulation of bonded indebtedness by local governments.

First, critics have characterized the two-thirds requirement as a "crude mechanical response to improvident overextension of public debt in the fiscally unsophisticated period of the latter Nineteenth Century." (See Cal. Const. Revision Com., supra, p. 370.) Petitioners point out that virtually every factor which may have been relevant during the chaotic period of its enactment has since been altered dramatically. One of the most significant changes is the "great improvement in the quality and integrity of governmental and financial administration. The existence of more responsible and responsive local legislative bodies, the development of a corps of professional administrators and finance officers, improvements in the management of local affairs, and the development and refinement of standards, procedures and techniques for financial administration — these are all important reasons why today's local governments are not likely to spend themselves into bankruptcy or to refuse to pay their debts." (Marini, Local Bond Elections in California: The Two-Thirds Majority Requirement (Univ. of Cal. Institute of Governmental Studies, 1963) p. 4.)

The institutions of the private sector have undergone a similar evolution. The bond market was a shambles when section 18 was written; now its well ordered operations discourage unsound bond offerings. "Numerous modern procedures of fiscal administration have been developed since 1879. A host of auditing and financial accounting requirements have been written into law. [See, e.g., Ed. Code, §§ 17199, 17206.] The bond market itself is subject to safeguards precluding most of the practices that the 1879 constitutional provision was probably intended to correct. For example, greatly improved reporting procedures are now in effect, assuring that full and accurate information on the financial and legal ramifications of individual bond issues is supplied to prospective buyers.^[44] Highly qualified bond consulting firms, specialized legal counsel, and careful bond issue evaluation by respected national rating agencies, have all built strong public confidence in most local bond offerings. The sense of security appears to be general, and is not limited to states requiring an extraordinary majority vote. This fact, in itself, strongly suggests that the many institutional and procedural improvements outlined here — and not the two-thirds requirement — are responsible for the soundness of modern local capital outlay finance." (Scott and Hamilton, op. cit. supra, at p. 2.)^[45]

Second, other provisions of article XI, section 18, itself serve to moderate any tendency to incur debt beyond the capacity of the issuing entity to service or beyond its capacity to service without curtailing essential governmental functions. Each general obligation bond issue is required to contain a provision "for the collection of an annual tax sufficient to pay the interest on such indebtedness as it falls due" and also "to constitute a sinking fund for the payment of the principle thereof...." (Cal. Const., art. XI, § 18)^[46] It is this tax, the total amount of the bond issue, and the purpose for which the bonds are issued and on which their proceeds are to be expended, which are submitted to the electorate. The requirement of an election itself may act as a deterrent to unnecessary borrowing, since the effort and expense entailed make it unlikely to be undertaken without serious thought and preparation. More important, the electoral mechanism permits open debate and discussion of the merits of each bond proposal by the residents of the local community. We consider it fanciful to argue, in the absence of any evidence, that a majority of this electorate, better educated and with access to far more sources of information than its counterpart of a century ago,^[47] is so incapable of mature judgment that it will bankrupt itself through indiscriminate borrowing.

Nor is there any evidence that California's two-thirds requirement (which is among the most restrictive in the nation),^[48] has produced greater fiscal security than may be achieved with a simple majority vote requirement. Although most states do not require a two-thirds vote, we have been referred to no data indicating more frequent bond defaults, lower credit ratings, or extravagant public projects in those states. As we have previously noted, since 1849 the State of California has been permitted to borrow in excess of its debt limit with the approval of 50 percent of those voting on a particular bond issue. (Cal. Const., art. XVI, § 1) Yet, State of California general obligation bonds are rated as high-grade bonds by the major investment advisory services. The judgment of the market place does not support respondents' argument.^[49] Indeed, the credit rating of a community may well be eventually impaired if, as a result of inability to secure a two-thirds consensus, necessary projects are deferred and the quality of life deteriorates.

Finally, there is impressive evidence that the two-thirds vote requirement has not so much limited total local government indebtedness as it has redirected borrowing into vehicles other than the general obligation bonds of regularly constituted political units. While the economic consequences of these alternate financing devices are substantially similar to those of voted general obligation bonds, they are not required to receive a two-thirds majority vote and, in some instances, permit long-term borrowing over which the public has no direct control at all.^[50]

This phenomenon has attracted the widespread, often critical, attention of legal and financial commentators.^[51] Their consensus, which we accept, is that the diversion of local borrowing into novel, more costly financing methods which often entirely evade submission to elections, renders any claim that the two-thirds vote requirement insures sound fiscal policy and popular control of long-term borrowing highly dubious at best.^[52]

We find nothing to support respondents' grim warnings of financial disaster in the absence of a two-thirds vote requirement. We must, therefore, conclude that that provision is not necessary to insure the solvency of our local governments.

B.

A variety of other interests are asserted to be served by the extraordinary majority vote requirement. It is claimed, for example, that the state should be allowed to insist, before permitting its political subdivisions to incur bonded debt, that "a strong popular and informed public sentiment be satisfied as to the priority, the wisdom, the financial ability of the community to absorb the cost, and the public interest [in] and necessity for the creation of indebtedness for the particular improvement." This public sentiment, it is contended, is not expressed in a majority vote; it is expressed in a two-thirds vote. We do not doubt that a two-thirds requirement is well suited to deter the forms of governmental action to which it applies, since it conditions such action on a degree of consensus unusual in our pluralistic society. This, however, merely emphasizes the impact and burden of section 18; it does not justify it.

It is explained that the necessity for this consensus stems from the nature of bonded indebtedness. We are told that long-term borrowing irrevocably commits the future revenues of a community and that such a drastic step should not be undertaken upon the will of a mere majority. However, many decisions of government at all levels are crucial and irreversible, yet this alone has never been thought a sufficient ground upon which to remove them from the democratic process. We are not persuaded that a decision to commit future tax revenues to the financing of projects whose scale and useful life make payment of their costs from current revenues impracticable or inequitable is so unique or so fraught with peril that it warrants this deprivation of voting rights.

We do not think respondents have demonstrated that bond-finance decisions are fundamentally different from many other political decisions made by a majority vote or by representatives elected by a majority vote. Even assuming, however, that respondents had done so we think their argument fails at another level. This justification for the extraordinary majority requirement rests on the premise that a decision to undertake a project such as the construction of schools and playgrounds is qualitatively different from a decision not to do so. This, in turn, is based on the assumption that spending money is a more serious matter than not spending it and, consequently, must be justified whereas frugality is self-justifying. A predisposition to thrift may serve a man well. It does not, however, justify governmental inertia, especially when government is faced with critical social problems demanding urgent and sometimes costly remedies. There is no presumption in favor of inaction, as the United States Supreme Court observed in Avery v. Midland County, supra, 390 U.S. 474, 484 [20 L.Ed.2d 45, 53]: "[W]e might point out that a decision not to exercise a function within [local government's] power — a decision, for example, not to build an airport or a library, or not to participate in the federal food stamp program — is just as much a decision affecting all citizens ... as an affirmative decision."

In sum, we do not believe that the nature of general obligation bond financing warrants diluting the votes of those who want action, in order to institutionalize a preference for the interests of those who want stasis. (Hunter v. Erickson, supra, 393 U.S. 385, 392-393 [21 L.Ed.2d 616, 622-623].)^[53]

Nor do we believe that the two-thirds requirement may be justified as a method of protecting real property owners from additional taxes necessary to service the higher level of bonded indebtedness which it is reasonable to assume would result were bonds to be issuable upon simple majority approval. We express no opinion as to whether or in what circumstances the state may structure its election laws to accommodate the financial interests of property owners. Assuming, arguendo, this to be a legitimate state interest, we do not believe that under contemporary conditions it is so compelling as to justify depriving petitioners of their right to a fully effective vote.

The argument that a two-thirds requirement is necessary to protect property owners rests on the following assumptions: (1) That the relationship between the level of bonded debt and the property tax rate is constantly proportional; (2) that bonded debt is voted by an electorate which, by and large, does not pay property taxes and is thus undeterred by the prospect of their increase; and (3) that maintenance of low property tax rate levels will invariably be in the best short and long-term interest of property owners. While possibly accurate in the 1870's, none of these assumptions has been shown to be valid today.

First, the close link between general obligation indebtedness and property taxes, characteristic of the nineteenth and early twentieth century, no longer exists.^[54] Not only has the percentage of municipal revenues generated by the property tax declined sharply,^[55] but in practice, many general obligation bonds have been redeemed through revenues derived from sources, such as excise taxes and enterprise activities, other than the property tax.^[56]

Second, the information available to us does not warrant a characterization of property owners as a beleaguered minority which would be forced to shoulder the total cost of bond-financed projects. There are indications that a substantial majority of the registered voters in California own their own homes.^[57] In addition, it is important to recognize that those who do not own real property nevertheless pay property taxes indirectly. Rental payments of lessees of taxable real property are to a material extent a function of the property tax.^[58] Those few who neither own nor rent taxable property must still purchase goods and services from commercial enterprises that do and which, to some extent at least, incorporate that tax in the prices charged consumers.^[59]

Thus, the factual predicates of respondents' contention do not exist. The probability of a property-owning minority being subjected to confiscatory taxation by the unbridled appetites of the propertyless masses is no more than "theoretically imaginable."^[60] (Mine Workers v. Illinois Bar Assn. (1967) 389 U.S. 217, 224 [19 L.Ed.2d 426, 432, 88 S.Ct. 353].) No such remote dangers can justify the "immediate and crippling impact" (Williams v. Rhodes, supra, 393 U.S. 23, 33 [21 L.Ed.2d 24, 33]) on the rights of all citizens (property owners or not) who favor certain proposed governmental action, to receive equal protection of the law.

Finally, respondents advance the argument that the compulsory referendum of section 18 stimulates development of a more enlightened citizenry and provides local residents with direct control over their government. This may well be true. It is, however, a purpose equally well served by a referendum whose outcome is determined by simple majority will. There is no causal relationship (much less a necessary one) between the benefits of "participatory democracy" and extraordinary majority vote requirements.

We conclude, therefore, that the two-thirds vote requirement is not necessary to the attainment of any compelling state interest.

IV

(14) We are unimpressed with respondents' remaining justifications proffered for the extraordinary majority requirement. Initially we find no significance in the fact that the two-thirds requirement of article XI, section 18, may be eliminated or modified by a simple majority of the voters in the state through a constitutional amendment. (See Cal. Const., art. IV, § 22; Cal Const., art. XVIII, § 1.) Our conclusion in this regard is squarely required by Lucas v. Colorado General Assembly (1964) 377 U.S. 713 [12 L.Ed.2d 632, 84 S.Ct. 1472]. The high court there held that the existence of a political remedy such as the initiative or referendum, while possibly justifying a court's decision to temporarily defer action, is of no constitutional significance. "Courts sit to adjudicate controversies involving alleged denials of constitutional rights ... [I]ndividual constitutional rights cannot be deprived, or denied judicial effectuation, because of the existence of a non-judicial remedy through which relief ..., which the individual voters seek, might be achieved." (377 U.S. 713, 736 [12 L.Ed.2d 632, 647].)^[61]

Nor is our conclusion altered by the fact that a proposed constitutional amendment to article XI, section 18 (Proposition 4), designed to reduce the required voter majority to 60 percent for school and library bond issues submitted to the voters at statewide primary or general elections, was defeated in November 1966. At the outset, we must observe that the proposed amendment might well have failed to command the support of many voters who opposed the two-thirds requirement yet found certain elements of Proposition 4 unacceptable.^[62] Thus, respondents' suggestion that the people of California have made a definite choice between two sharply contrasting alternatives and have indicated that they prefer, as a general rule, a minority veto power over local general obligation bond issue propositions is not factually supportable.

More fundamentally, popular approval of electoral systems which infringe an individual's constitutionally protected right to cast an equally weighted vote is irrelevant. "A citizen's constitutional rights can hardly be infringed simply because a majority of the people choose that it be. [Fn. omitted.]" (Lucas v. Colorado General Assembly, supra, 377 U.S. 713, 736-737 [12 L.Ed.2d 632, 647]; Jordan v. Silver (1965) 381 U.S. 415 [14 L.Ed.2d 689, 85 S.Ct. 1572] (concurring opinion).)^[63]

(15) We find no merit in the theoretical argument that there has been no violation of the equal protection clause because the elections here in issue represent an exercise of the legislative power retained by the people rather than of that delegated to the state Legislature. To describe in conceptual terms the origin of the power exercised at bond elections does not answer the question before us, which is to determine the restraints imposed on that power by the equal protection clause. The United States Supreme Court, however, has conclusively answered this argument: "[I]nsisting that a State may distribute legislative power as it desires and that the people may retain for themselves the power over certain subjects may generally be true, but these principles furnish no justification for a legislative structure which otherwise would violate the Fourteenth Amendment." (Hunter v. Erickson, supra, 393 U.S. 385, 392 [21 L.Ed.2d 616, 623].)

Respondents press a final argument upon us. They point out that the extraordinary majority requirement of article XI, section 18, is not unique and that there are many contexts in which governmental action is conditioned upon the ability of its proponents to secure the support of more than a bare majority. They observe that unqualified majoritarianism has not been the model for many of our institutions and that we have often afforded minorities power to preserve the status quo. They confront us with provisions of our federal and state Constitutions which sanction deviations from simple majority rule.^[64] They then admonish us that, since the two-thirds requirement cannot be distingiushed from any of these provisions, a decision invalidating it will inevitably undermine the constitutionality of all. We cannot agree.

First, there is no merit to the attempt to analogize section 18 to institutional arrangements of varied historical origin and specialized function, such as the electoral college, the allocation of two senate seats to each state regardless of population, or the unanimous jury verdict required for a criminal conviction. Efforts to justify state-imposed inequalities in voting power through reliance on analogies to other distinct institutions have met with no success in the United States Supreme Court. For example, in Gray v. Sanders, supra, 372 U.S. 368, 378 [9 L.Ed.2d 821, 829], the court declared: "We think the analogies to the electoral college, to districting and redistricting, and to other phases of the problems of representation in state or federal legislatures or conventions are inapposite. The inclusion of the electoral college in the Constitution, as the result of specific historical concerns, validated the collegiate principle despite its inherent numerical inequality, but implied nothing about the use of an analogous system by a State...." (Fns. omitted.) In Reynolds v. Sims, supra, 377 U.S. 533, 573 [12 L.Ed.2d 506, 534], the court observed that: "Attempted reliance on the federal analogy appears often to be little more than an after-the-fact rationalization offered in defense of maladjusted state apportionment arrangements." Our reaction to analogies to, for example, the treaty ratification procedure in the United States Senate is not dissimilar.

Second, many of the extraordinary majority provisions to which we are referred are readily distinguishable in that, since they apply solely to the internal procedures of legislative bodies, they involve no dilution of the individual exercise of the franchise which is in issue here. Some, such as the two-thirds required for conviction of impeachment (U.S. Const., art. I, § 3) or expulsion from Congress (U.S. Const., art. I, § 5) are designed to avoid precipitate action in areas of particular importance or sensitivity. Others, such as the two-thirds required to override a presidential (U.S. Const., art. I, § 7) or gubernatorial (Cal Const., art. IV, § 10(a)) veto or the two-thirds required to ratify a treaty (U.S. Const., art. II, § 2) represent those basic allocations of power between branches of government which are at the heart of that great system of "checks and balances" established by the founding fathers. In neither case is there any infringement of the individual citizen's right to vote such as that effected by section 18.

Finally, those extraordinary majority vote requirements which do apply outside the legislative process are by no means uniformly invalidated by our decision today. We emphasize that while it has not been demonstrated that a two-thirds vote requirement for approval of local general obligation bonds is necessary to promote a compelling state interest, similar provisions in other contexts may meet this standard. For example, it is common to insist upon a broad consensus before altering basic political documents. (See, e.g., U.S. Const., art. V.) Documents such as constitutions exist partly to provide continuity and stability to the affairs of state. This is a goal of fundamental importance and one which can be achieved only by protecting such charters from the will of temporary majorities. We see no a priori reason to assume that other extraordinary majority provisions cannot be shown to be necessary to attain "compelling" ends. Each must be judged on its particular facts. Those which serve no such end will fall, but our decision today commits us to no wholesale elimination of such laws. We, of course, express no opinion as to the validity of any of the foregoing provisions.

(16, 17) For the reasons already set forth, we hold that insofar as the constitutional provision engaging our attention (i.e., former art. XI, § 18, now renumbered as art. XIII, § 40) requires the assent of more than a simple majority to authorize the incurring of certain indebtedness by local governmental entities it violates the equal protection clause of the Fourteenth Amendment to the United States Constitution.^[65] (18) Since the requirement of a two-thirds vote is entirely distinct and severable from the other provisions of the section, the section remains, in all other aspects, valid and operative.

(19a) We turn now to the issue of affirmative relief. Although employing somewhat different language, petitioners in both S.F. 22706 and S.F. 22707 seek substantially the same relief. Both pray for a peremptory writ of mandate directed to respondents and commanding, generally, that Propositions "A" and "B" be certified as having been duly approved by the voters of San Francisco and that all the usual ministerial steps be taken in order to prepare the bonds for sale.

Respondents, in their return to the alternative writ, contend that, even assuming the two-thirds requirement of section 18 to be unconstitutional, the relief which petitioners request is improper. Respondents assert that, because of the procedures actually followed by the City and County of San Francisco, they have no present duties of a ministerial nature to perform in connection with either bond proposition which have not already been performed. We need not decide this question since we have determined that our decision today regarding the constitutionality of article XI, section 18 shall have prospective effect only and that accordingly no peremptory writs should issue.

(20) It is now beyond dispute that the United States Constitution permits state appellate courts to restrict the application of a newly announced rule of law to future cases. (England v. Medical Examiners (1964) 375 U.S. 411, 422 [11 L.Ed.2d 440, 449, 84 S.Ct. 461]; Great Northern Ry. Co. v. Sunburst Co. (1932) 287 U.S. 358, 364-366 [77 L.Ed. 360, 366-367, 53 S.Ct. 145, 85 A.L.R. 254]. See also the historical discussion in Linkletter v. Walker (1965) 381 U.S. 618, 622-629 [14 L.Ed.2d 601, 604-608, 85 S.Ct. 1731].) (21) "State courts have generally found state constitutions equally permissive and have frequently stated that the decision whether to apply an overruling decision retroactively or prospectively only turns on considerations of fairness and public policy. [Citations.]" (Forster Shipbuilding Co. v. County of Los Angeles (1960) 54 Cal.2d 450, 459 [6 Cal. Rptr. 24, 353 P.2d 736].) This is the rule we have adopted in California (Connor v. Great Western Sav. & Loan Assn. (1968) 69 Cal.2d 850, 868 [73 Cal. Rptr. 369, 447 P.2d 609]; County of Los Angeles v. Faus (1957) 48 Cal.2d 672, 681 [312 P.2d 680]) even though we have recognized that the purely prospective operation of a decision "temporarily preserves and applies a mistaken interpretation of the [state] Constitution. [Citation.]" (Forster Shipbuilding Co. v. County of Los Angeles, supra, 54 Cal.2d at p. 459.)

(19b) In recent years a substantial number of general obligation bond propositions have won the approval of a majority but less than two-thirds of the voters. (Scott and Hamilton, op. cit. supra, at pp. 3-5; Cal. Teachers' Assn., Research Bull. No. 228 (Oct. 1968), Results of Tax and Bond Elections in California School Districts 1967-1968, pp. 13-25; Cal. Teachers' Assn., Research Bull. No. 235 (Sept. 1969), Results of Tax and Bond Elections in California School Districts 1968-1969, pp. 17-32.) These bonds share the precise status of those for which Propositions "A" and "B" sought voter authorization. Thus, to declare that the bonds before us were authorized by the voters and to order that they be prepared for sale would logically commit us to the validation of hundreds of millions of dollars in bonded indebtedness voted upon at now-forgotten elections. Such a course might well impose severe and unforeseen hardships upon many Californians who, quite reasonably, have made significant personal, financial, and civic decisions in reliance upon the apparently settled declaration of election results. Additionally, since times and priorities change, were these same bond propositions resubmitted to the voters today, many might fail to secure even a majority vote. In such circumstances, we consider it fairer and wiser that only prospective effect be given to our decision and that its rule be applied only to elections held after the date upon which this decision becomes final in this court.

This course will settle immediately the status of bond propositions on a uniform, statewide basis. It will permit local governments to propose to the electorate new bond issues, drawn to meet present community needs and to conform to present economic conditions. It will also allow voters to conduct campaigns and cast their ballots on bond propositions with full understanding, in advance, of the rules by which the outcome of the elections will be determined.

The force of these and similar considerations of public policy have impressed other courts. Thus, we find ample authority for our refusal to disturb the results of an election conducted under ground rules later held invalid (Allen v. State Board of Elections (1969) 393 U.S. 544, 571-572 [22 L.Ed.2d 1, 20-21, 89 S.Ct. 817]; Rimarcik v. Johansen, supra, 310 F. Supp. 61, 71; Jenness v. Little (N.D. Ga. 1969) 306 F. Supp. 925, 929; appeal dismissed 397 U.S. 94 [25 L.Ed.2d 81, 90 S.Ct. 820]) and for our decision to limit the retroactive effect of judicial intervention in the bond election process. (Cipriano v. City of Houma, supra, 395 U.S. 701, 706 [23 L.Ed.2d 647, 652]; see also Douglass v. County of Pike (1879) 101 U.S. 677, 687 [25 L.Ed. 968, 971-972]; Gelpcke v. City of Dubuque (1863) 68 U.S. 175, 206 [17 L.Ed. 520, 525-526].)

In each of the proceedings before us (S.F. 22706 and S.F. 22707), the alternative writ of mandate is discharged and the petition for a peremptory writ is denied.

Wright, C.J., McComb, J., Tobriner, J., and Schauer, J.,^[*] concurred.

MOSK, J.

I concur in the court's conclusion that the two-thirds requirement for passage of bond issues is invalid, but I dissent from the majority's refusal to apply its decision to the litigants before the court.

The majority tell the parties who initiated these proceedings, who used their talents and resources to seek and obtain a hearing in this court, and who invoked the constitutional principles upon which we here rely, that they are undeniably right, their concept of the law is totally vindicated — but they lose, their petition for a peremptory writ denied. This is indeed a Pyrrhic victory, or as John Dryden wrote in the 17th century: "Even victors are by victories undone."

Few problems in both the criminal and civil fields have been fraught with more uncertainty than determination of the precise date on which newly discovered constitutional precepts become applicable. Justice Harlan referred to "the doctrinal confusion that has characterized our efforts" in this arena and the "many incompatible rules and inconsistent principles." (Desist v. United States (1969) 394 U.S. 244, 258 [22 L.Ed.2d 248, 260, 89 S.Ct. 1030] (dissenting opinion).) Legal scholars have been as hopelessly at odds over theories of retroactivity versus prospectivity. (See, e.g., Schaefer (1967) 42 N.Y.U.L. Rev. 631, as contrasted with Haddad (1969) 60 J.Crim.L. 417.)

The federal rule was originally settled, the Supreme Court adhering to the command of article III of the Constitution that it decide only existing controversies. "I know of no authority in this court to say," wrote Justice Holmes with typical grandeur, that decisions "shall make law only for the future." (Kuhn v. Fairmont Coal Co. (1910) 215 U.S. 349, 372 [54 L.Ed. 228, 239, 30 S.Ct. 140] (dissenting opinion).) Two decades later in Great Northern Ry. Co. v. Sunburst Oil & Refining Co. (1932) 287 U.S. 358 [77 L.Ed. 360, 53 S.Ct. 145, 85 A.L.R. 254], the legitimacy of the technique of prospective overruling was firmly established.

However, Sunburst made it clear that "A state in defining the limits of adherence to precedent may make a choice for itself between the principle of forward operation and that of relation backward. It may say that decisions of its highest court, though later overruled, are law nonetheless for intermediate transactions ... whenever injustice or hardship will thereby be averted." (Id. at p. 364 [77 L.Ed. at p. 366].)

Thus three alternatives are available to a state court which has fashioned a new rule. It may apply the rule: (1) to acts occurring subsequent to the announcement only; (2) to acts occurring subsequent to the announcement and also to the present litigants; or (3) to acts occurring subsequent to the announcement, to the present litigants, and also to acts which occurred prior to the announcement. (Note (1962) 71 Yale L.J. 909, 933.) In the instant case the majority choose the first alternative. I prefer the second.

Only the most amorphous guidelines are available to determine which of the alternatives is appropriate in an individual case. Justice Cardozo referred to "considerations of convenience, of utility, and of the deepest sentiments of justice." (Cardozo, The Nature of the Judicial Process (1921) 146-149). Justice Traynor spoke of "considerations of fairness and public policy." (Forster Shipbuilding Co. v. County of Los Angeles (1960) 54 Cal.2d 450, 459 [6 Cal. Rptr. 24, 353 P.2d 736].) A federal court indicated "that retroactivity should be determined from the circumstances of the particular case, having in mind the purpose which the new rule of law seeks to accomplish and the practical weighing of the comparative benefits and evils of retroactivity." (Hanover Shoe Inc. v. United Shoe Machinery Corp. (3d Cir.1967) 377 F.2d 776, 789.)

I am unimpressed by the "weighing of comparative benefits and evils," or the reasons — indeed, there is a paucity of reasons — given by the majority for denying to petitioners the fruits of their victory. The majority speak of bonds "voted upon at now-forgotten elections" (ante, p. 801). In addition to being a gratuitous reflection upon the delays of the judicial process — not markedly prolonged in this instance^[1] — the simple fact is that these specific bond elections are not now-forgotten. They have been kept very much alive in current memory by this well-publicized litigation; both proponents and opponents at the election are eagerly awaiting the results of this lawsuit, and have undoubtedly made appropriate plans for the eventuality of a determination pro or con their position.

To invoke a prospective-only technique results in an arbitrariness toward individual litigants that is unchararacteristic of a court of justice. As Justice Harlan wrote in his dissent in Desist, it is a "truism that it is the task of this court, like that of any other, to do justice to each litigant on the merits of his own case."

The United States Supreme Court in Stovall v. Denno (1967) 388 U.S. 293, 301 [18 L.Ed.2d 1199, 1206, 87 S.Ct. 1967], suggested that someone, i.e., the litigants before the court, must be given the benefit of decisions which create new rights. The high court held: "We recognize that Wade and Gilbert are, therefore, the only victims of pretrial confrontations in the absence of their counsel to have the benefit of the rules established in their cases. That they must be given that benefit is, however, an unavoidable consequence of the necessity that constitutional adjudications not stand as mere dictum. Sound policies of decision-making, rooted in the command of Article III of the Constitution that we resolve issues solely in concrete cases or controversies, and in the possible effect upon the incentive of counsel to advance contentions requiring a change in the law, militate against denying Wade and Gilbert the benefit of today's decisions. Inequity arguably results from according the benefit of a new rule to the parties in the case in which it is announced but not to other litigants similarly situated in the trial or appellate process who have raised the same issue. But we regard the fact that the parties involved are chance beneficiaries as an insignificant cost for adherence to sound principles of decision-making." (Fns. omitted.)

Other authorities have advanced the theory that prospective overruling destroys incentive for appeal, and if it becomes a frequent policy, it will tend to deter counsel from presenting "issues involving renovation of unsound or outmoded legal doctrines." (Mishkin, Foreword, The Supreme Court 1964 Term, 79 Harv.L.Rev. 56, 61.)

A number of state courts have adopted substantial changes in the law and applied the new doctrine to the case at hand while otherwise giving it only prospective application. (See, e.g., Balts v. Balts (1966) 273 Minn. 419 [142 N.W.2d 66, 75].) California adopted this procedure when it eliminated governmental immunity (Muskopf v. Corning Hospital Dist. (1961) 55 Cal.2d 211 [11 Cal. Rptr. 89, 359 P.2d 457]) and again when it imposed liability upon institutions financing tract developments. (Connor v. Great Western Sav. & Loan Assn. (1968) 69 Cal.2d 850 [73 Cal. Rptr. 369, 447 P.2d 609].)

In short, the majority subvert an otherwise irrefutable opinion by denying equable application of its benefits. The aggrieved parties who knocked on our door are turned away with approbation but without assistance. This, I believe, cannot be rationalized. One can only hope the petitioners find some solace in their service pro bono publico.

I would issue the peremptory writ.

Mr. Justice Peters has authorized me to say that he, too, would issue the writ.

NOTES

[1] Several constitutional amendments proposed by the Constitution Revision Commission were submitted to the voters at the election held on June 2, 1970. Proposition 2, the only such amendment to pass, contained provisions repealing article XI, section 18 and reenacting it, in identical language, as article XIII, section 40. The effect of the constitutional amendment was a simple renumbering of the former section. In view of this fact and since the parties in the instant cases and in each of the companion cases filed today, in their pleadings, briefs and oral arguments, all of which were presented to us before June 2, 1970, discussed the above constitutional provision in terms of its former numbering (art. XI, § 18) and since many of the authorities cited to us naturally referred only to such former section, we have decided, in the interest of convenience, to refer, in this and in each of the companion cases filed today, to the pertinent provision under its former numbering of article XI, section 18. We wish to make it clear, however, that what we say and hold herein is intended to apply to the provision under its new numbering of article XIII, section 40.

Towns and townships do not exist as local governmental units in California and boards of education are simply the governing bodies of school districts. (See Cal. Const. Revision Com., Background Study, Article XI: Local Government, p. 363.) Thus section 18 effectively applies only to counties, cities, and school districts, which are, however, the basic units of local government in the state.

In addition to requiring extraordinary majority voter approval, section 18 provides, as a condition to the incurring of long-term debt, that "provision shall be made for the collection of an annual tax sufficient to pay the interest on such indebtedness as it falls due, and also provision to constitute a sinking fund for the payment of the principal thereof, ..." The maximum term of such indebtedness is fixed at 40 years. Petitioners do not challenge these regulations of local borrowing.

[2] San Francisco is a consolidated city and county. It is subject to the provisions of section 18 by virtue of article XI, section 7 of the California Constitution.

[3] Proposition "A" received the affirmative votes of 56.8 percent of those voting on the proposition; Proposition "B" received the support of 52.3 percent of those so voting.

[4] Respondent Mihaly is the registrar of voters, respondent Dolan the clerk of the board of supervisors, and respondent Cooper the controller, of the City and County of San Francisco.

[5] The substance of the provision is retained in the present article XVI, section 1. During the early part of the nineteenth century, the eastern states borrowed heavily to finance capital improvements such as railroads, turnpikes, and canals. These ventures ended in financial disaster in 1837. "The midwestern states went through a similar cycle ... two decades later. Unable to meet their obligations, the states wound up these enterprises with great financial loss. Such ... experience led either to the abolition of the power to borrow for state improvements in the state legislature ... or to the introduction of various spending and financial restrictions in the state constitution." (Fn. omitted.) (L. Chermak, The Law of Revenue Bonds (1954), p. 53.)

The constitutional limit on state debt incurred without the approval of a popular majority came before this court in People v. Johnson (1856) 6 Cal. 499. We set forth at some length the opinion of Chief Justice Murray not simply for its description of the period but because it is a revealing example of the profound, moral reaction against financial extravagance on the part of government. He observed: "The framers of our State Constitution were mostly men fresh in the experience of the errors into which other States had fallen. They had witnessed the unhappy results that followed extravagant legislation, and were anxious to rear a bulwark here, which would protect us against similar disasters. They were thoroughly acquainted with the history of the gigantic schemes of internal improvements entered into by the South and West; of the establishment of banks; of the expansion of their circulating medium; of the inflation of everything; of the general demoralization, which resulted from an uncontrolled power of legislation; and, in fact, of the weakness of human nature itself. They had witnessed ruin like an avalanche overwhelm them, destroying visionary hopes of wealth and greatness, and leaving them to sink beneath the crushing weight of debt and taxation, or driving them to the more dishonorable alternative of repudiation....

"With these examples before them, in a country whose very atmosphere was heavy with gold, ... can it be wondered that the framers of our Constitution should have attempted to prevent the evil consequences which they foresaw, by constitutional checks and barriers? No one can doubt their intention, and it is only to be regretted that that which was so wisely begun, has been so rashly departed from....

"If we were at liberty to decide this case on grounds of public policy, we could not hesitate one moment between bankruptcy and ruin, or credit and prosperity. It is time that the axe should be laid at the root of this political evil; that this system of extravagance, which has been the curse of the State for the last five years, ... should be stopped; and that honesty, prudence and economy should preside over the legislation of the State, and the administration of her financial affairs." (6 Cal. 499, 503-505.)

[6] Restrictions varied from outright prohibition to authority to borrow without limit. Twenty-six out of sixty-eight charter cities required elections to incur certain types of indebtedness but only four required extraordinary majorities. (Starner, General Obligation Bond Financing by Local Governments: A Survey of State Controls (Univ. of Cal., Bureau of Public Administration 1961) p. 7.)

[7] One commentator has noted that "Much of the over-extended municipal debt of the mid-Nineteenth Century stemmed from the excessive zeal of local communities for transportation links, particularly for rail development." (Starner, op. cit. supra, at p. 7.) An example of this, and of the state Legislature's casualness in acceding to demands for special legislation, was an act passed in 1866 authorizing the City of Petaluma to issue $500,000 in bonds to construct locks, dams, and canals on the Petaluma Creek and to collect tolls from the vessels which, it was anticipated, would use the network. (See Beebe, et al., Joint Powers Authority Revenue Bonds (1968) 41 So.Cal.L.Rev. 19, 24.) Unsuccessful ventures such as this were among the "abuses" referred to in article IV, section 37 of the 1849 Constitution.

[8] "With the Federal Government remaining impotent and the state governments restricting their public works activities, the burden of furnishing internal improvements ... fell upon the local governments and private enterprise. Municipal debt rose from $20,000,000 in 1840 to $200,000,000 in 1860, being used primarily for the financing of the paving and sewering of streets, the constructing of water supply systems, and the making of railroad stock subscriptions. When the crisis of 1873-1874 found municipal debt vastly in excess of state debt, constitutional and statutory restrictions upon municipal borrowing were adopted...." (Chermak, op. cit. supra, at pp. 53-54.)

[9] D. Fehrenbacher, A Basic History of California (1964) at page 48.

[10] R. Roske, Every Man's Eden: A History of California (1968) pp. 378-379. There were 10,000 workingmen unemployed in San Francisco alone which, at that time, was a city of only 165,000.

[11] Mr. O'Donnell, a delegate from San Francisco and a member of the Workingmen's Party, described the social conditions which had led to the calling of the convention: "The people were suffering great social and political wrongs. There was sloth in the mart and schism in the temple; corruption in office and out of it. Murmurs of discontent arose among the people...." (1 Debates and Proceedings of the Constitutional Convention, p. 17.) Colonel Hoge, a representative of a more moderate faction, who was elected president of the convention, was less rhetorical but hardly more sanguine: "We have met, gentlemen, at not the most favorable time for making a Constitution. For many causes there seems to be very great depression in the business and industrial enterprises of the country. To some extent we have fallen upon evil times...." (Id. at p. 22.)

[12] Scott and Hamilton, Extraordinary Majority Voting Requirements, 10 Public Affairs Report No. 4 (Univ. of Cal., Institute of Governmental Studies, Aug. 1969), page 2.

[13] Starner, op. cit. supra, at page 6.

[14] The provision is now contained in article IV, section 16.

[15] We refer to the following language of section 18: "No county, ... shall incur any indebtedness or liability ... exceeding in any year the income and revenue provided for such year, ..." (Italics added.)

[16] Much of the recorded debate concerns the necessity and advisability of placing a maximum limit on local indebtedness. Proposals to accomplish this were introduced and a limit was included in initial drafts of section 18. Discussion on the floor centered on whether this limit should be expressed (as it was in the drafts) in the form of a ratio of debt to assessed value of taxable property, what the optimal debt to property ratio would be, whether certain types of projects should be exempted because of their social value and, if so, which types. Ultimately, no maximum limit was adopted, reliance apparently being placed on the other curbs on the incurring of debt. (2 Debates and Proceedings of the Constitutional Convention, supra, pp. 1068-1071; 3 Debates and Proceedings of the Constitutional Convention, supra, pp. 1381-1383.)

Nor do our cases shed any light on the purpose intended to be achieved by the two-thirds vote requirement. Two of the earliest cases, San Francisco Gas Co. v. Brickwedel, supra, 62 Cal. 641 and McBean v. City of Fresno (1896) 112 Cal. 159 [44 P. 358], discuss only the "pay as you go" aspect of the constitutional provision. City of Redondo Beach v. Taxpayers, Property Owners etc., City of Redondo Beach (1960) 54 Cal.2d 126 [5 Cal. Rptr. 10, 352 P.2d 170] does assign a purpose to the section as a whole but does not purport to explain the choice of a two-thirds majority: "The purpose ... is to safeguard the general funds and property of a municipality from a situation whereby the holders of an issue of bonds could, at some time after the issuance thereof, force an unconsented-to increase in the taxes of, or foreclosure on the general assets and property of the issuing public corporation...." (54 Cal.2d at p. 131.) This statement does not indicate what the relation between this purpose and an extraordinary majority vote requirement is, or was intended to be. Finally, while City of Palm Springs v. Ringwald (1959) 52 Cal.2d 620 [342 P.2d 898], does address itself to that portion of section 18 which requires a referendum, the case states only that it "is designed to afford the people who are required to pay the cost of providing such objects of public convenience and welfare an opportunity to express their approval or disapproval of a long-term indebtedness." (52 Cal.2d at p. 627.) This is hardly a satisfactory explanation for the two-thirds requirement, since the assigned purpose would be served equally well by a mandatory referendum whose outcome is determined by a simple majority vote.

[17] Thirty years were still to elapse before the Seventeenth Amendment sanctioned popular election of United States Senators, and it would be 40 years until the Nineteenth Amendment extended the franchise to women.

[18] See, e.g., The Voting Rights Act of 1965, 42 U.S.C.A. section 1971 et seq.

[19] United States Constitution, Amendment XXIV.

[20] Carrington v. Rash (1965) 380 U.S. 89 [13 L.Ed.2d 675, 85 S.Ct. 775], held unconstitutional a Texas law which barred all servicemen who had moved to Texas during their military service from voting in any state election.

[21] Louisiana v. United States (1965) 380 U.S. 145 [13 L.Ed.2d 709, 85 S.Ct. 817], banned the employment of voter qualification examinations with such discretionary standards as to permit their use as deliberate devices for racially motivated exclusions.

[22] Harper v. Virginia Board of Elections, supra, 383 U.S. 663, invalidated state laws requiring the payment of poll taxes as a condition to voting in state elections.

[23] Kramer v. Union School District (1969) 395 U.S. 621 [23 L.Ed.2d 583, 89 S.Ct. 1886], held that local school district elections could not, consistent with the equal protection clause, be limited to parents of school children, property owners and lessees of real property. Cipriano v. City of Houma (1969) 395 U.S. 701 [23 L.Ed.2d 647, 89 S.Ct. 1897], struck down a Louisiana statute confining the right to vote in municipal utility revenue bond elections to property taxpayers.

[24] See, e.g., Katzenbach v. Morgan (1966) 384 U.S. 641 [16 L.Ed.2d 828, 86 S.Ct. 1717], which upheld section 4(e) of the Voting Rights Act of 1965, invalidating state English literacy requirements as applied to certain foreign language literates.

[25] See Otsuka v. Hite (1966) 64 Cal.2d 596 [51 Cal. Rptr. 284, 414 P.2d 412], holding that the exclusion of those convicted of "infamous crimes" (Cal. Const., art. II, § 1) could not extend to all felons; Castro v. State of California, supra, ante, p. 223, holding that the requirement of literacy in English could not be applied to citizens who are literate in Spanish.

[26] Gray v. Sanders (1963) 372 U.S. 368 [9 L.Ed.2d 821, 83 S.Ct. 801], held that Georgia's "county unit system" (somewhat analogous in operation to the federal electoral college) unconstitutionally diluted the weight of some votes on the basis of geographical location. Gray stands for the proposition that "Once the geographical unit for which a representative is to be chosen is designated, all who participate in the election are to have an equal vote...." (372 U.S. at p. 379 [9 L.Ed.2d at p. 829].)

[27] Wesberry v. Sanders (1963) 376 U.S. 1 [11 L.Ed.2d 481, 84 S.Ct. 526], extended the rule of population equality to congressional districting. An apportionment of congressional seats which "contracts the value of some votes and expands that of others" is unconstitutional, Wesberry held since "the Federal Constitution intends that when qualified voters elect members of Congress each vote be given as much weight as any other vote...." (376 U.S. at p. 7 [11 L.Ed.2d at p. 486].)

[28] Reynolds v. Sims (1964) 377 U.S. 533 [12 L.Ed.2d 506, 84 S.Ct. 1362] applied the "one man, one vote" principle of Gray to state legislative apportionment. Every qualified resident, Reynolds held, has the right to a ballot for the election of state legislators of equal weight to the vote of every other resident and that right is infringed when legislators are elected from districts of substantially unequal population.

[29] Kirkpatrick v. Preisler (1969) 394 U.S. 526 [22 L.Ed.2d 519, 89 S.Ct. 1225]; Wells v. Rockefeller (1969) 394 U.S. 542 [22 L.Ed.2d 535, 89 S.Ct. 1234].

[30] Avery v. Midland County (1968) 390 U.S. 474 [20 L.Ed.2d 45, 88 S.Ct. 1114]; Hadley v. Junior College District (1970) 397 U.S. 50 [25 L.Ed.2d 45, 90 S.Ct. 791].

[31] California cases implementing the "one man, one vote" mandate include: Silver v. Reagan (1967) 67 Cal.2d 452 [62 Cal. Rptr. 424, 432 P.2d 26]; Wiltsie v. Board of Supervisors (1966) 65 Cal.2d 314 [55 Cal. Rptr. 222, 421 P.2d 70]; Miller v. Board of Supervisors (1965) 63 Cal.2d 343 [46 Cal. Rptr. 617, 405 P.2d 857] and Silver v. Brown (1965) 63 Cal.2d 270 [46 Cal. Rptr. 308, 405 P.2d 132].

[32] See Williams v. Rhodes, supra, 393 U.S. 23 [21 L.Ed.2d 24], holding unconstitutional under the equal protection clause Ohio election laws which imposed requirements to qualify for positions on the ballot giving a decided advantage to the established political parties; Hunter v. Erickson (1969) 393 U.S. 385 [21 L.Ed.2d 616, 89 S.Ct. 557], invalidating, on equal protection grounds, a city charter provision requiring that fair housing ordinances passed by the city council be submitted to the voters in a city-wide referendum.

[33] The question has been decided by the highest court of two states, West Virginia and Idaho. Lance v. Board of Education of County of Roane (1969) 153 W. Va. 559 [170 S.E.2d 783] held that a 60 percent vote requirement for the issuance of local general obligation bonds contravened the equal protection clause. Bogert v. Kinzer (1970) 93 Idaho 515 [465 P.2d 639], reached the opposite conclusion and upheld a provision of the state constitution requiring a two-thirds majority for the issuance of general obligation bonds. Neither opinion fully explains its conclusion nor why the United States Supreme Court voting rights and reapportionment cases are or are not apposite. Consequently, neither is of particular assistance to us. The United States Supreme Court on April 6, 1970, granted a writ of certiorari in Lance v. Board of Education of County of Roane, supra. (Docket No. 1244, Gordon v. Lance.)

[34] Respondents point out that an individual may vote in favor of some bond propositions and against others at the same election. We fail to see the relevance of this observation; it certainly does not demonstrate that section 18 works no classifications. It should be apparent that the section does not classify on the basis of characteristics which will define a physically cohesive, unchanging group. Rather, its lines are drawn on the basis of each person's preference on specific bond propositions in particular elections. Accordingly, the class so defined will vary to some extent from issue to issue, election to election. The fact remains, however, that each participant in the election must vote "yes" or "no" and section 18 classifies the voter on the basis of that expressed opinion. No more is required for the equal protection clause to apply.

[35] As stated by Justice Black, dissenting, in Colegrove v. Green, supra, 328 U.S. 549, 569-570 [90 L.Ed. 1432, 1445]: "No one would deny that the equal protection clause would ... prohibit a law that would expressly give certain citizens a half-vote and others a full vote.... [T]he constitutionally guaranteed right to vote and the right to have one's vote counted clearly imply the policy that state election systems, no matter what their form, should be designed to give approximately equal weight to each vote cast." See also the opinion of Justice Douglas, dissenting in South v. Peters (1950) 339 U.S. 276, 279 [94 L.Ed. 834, 838, 70 S.Ct. 641]: "There is more to the right to vote than the right to mark a piece of paper and drop it in a box or the right to pull a lever in a voting booth. The right to vote includes the right to have ... the vote counted at full value without dilution or discount [citation].... That federally protected right suffers substantial dilution ... [where a] favored group has full voting strength ... [and] [t]he groups not in favor have their votes discounted."

[36] Purdy & Fitzpatrick v. State of California (1969) 71 Cal.2d 566, 578-579 [79 Cal. Rptr. 77, 456 P.2d 645].

[37] Certain classifications, such as those based on race (McLaughlin v. Florida (1964) 379 U.S. 184 [13 L.Ed.2d 222, 85 S.Ct. 283]) or wealth (Douglas v. California (1963) 372 U.S. 353 [9 L.Ed.2d 811, 83 S.Ct. 814]), are "highly suspect" (McDonald, supra, at p. 807 [22 L.Ed.2d at p. 744]).

[38] Interests which have been identified as fundamental and therefore deserving of special treatment under the equal protection clause include procreation (Skinner v. Oklahoma (1942) 316 U.S. 535 [86 L.Ed. 1655, 62 S.Ct. 1110]) and interstate travel (Shapiro v. Thompson (1969) 394 U.S. 618 [22 L.Ed.2d 600, 89 S.Ct. 1322]).

[39] In Kramer, supra, the court stated: "Our exacting examination is not necessitated by the subject of the election; rather, it is required because some resident citizens are permitted to participate and some are not." (395 U.S. 621, 629 [23 L.Ed.2d 583, 591].) And, in Hadley v. Junior College District, supra, 397 U.S. 50, 54 [25 L.Ed.2d 45, 50], it held: "When a court is asked to decide whether a State is required by the Constitution to give each qualified voter the same power in an election open to all, there is no discernible, valid reason why constitutional distinctions should be drawn on the basis of the purpose of the election."

[40] The election in Cipriano involved revenue bonds; those at issue here involve general obligation bonds. The parties have correctly ascribed no relevance to this distinction for the purpose of determining the applicable equal protection standard. The United States Supreme Court has just held that there is no significant difference between the two types of municipal security insofar as the constitutionality of excluding voters on property-ownership grounds is concerned. Applying the standard utilized in Kramer and Cipriano, the high court held that state laws which restrict to property taxpayers the vote in elections to approve proposed general obligation bonds violate the equal protection clause. (Phoenix v. Kolodziejski (1970) 399 U.S. 204 [26 L.Ed.2d 523, 90 S.Ct. 1990].)

[41] Rimarcik v. Johansen (D. Minn. 1970) 310 F. Supp. 61 supports our conclusion. In that case a three-judge federal district court struck down as violative of the equal protection clause a Minnesota statute requiring 55 percent voter approval to adopt or amend city charter provisions governing liquor control laws. While recognizing that the reapportionment cases were factually distinguishable, the court thought the principle they vindicated — the right of a citizen to have his vote given equal weight to those of other voters — was applicable. The court applied the standard articulated in Kramer, which it considered controlling in all cases of denial or dilution of voting rights, and held the 55 percent vote requirement unconstitutional since it gave to those opposed to change in liquor laws a disproportionate voice in the outcome without a showing that this was necessary to promote a compelling state interest.

[42] There appear to be two causes for the usual vagueness with which these goals are described. The first is the scarcity of informative documentary evidence as to the intention of those who drafted and adopted the two-thirds requirement. (See pp. 14-15, ante.) To the extent this is the cause, the lack of specificity is unavoidable. The second cause appears to be a confusion between the general principle of debt control and the particular technique of an extraordinary majority vote requirement. We emphasize that the propriety of state imposed restrictions on local debt formation is not in issue. What is in issue is the legality of the two-thirds requirement and it is the connection between this device and municipal solvency which must be demonstrated.

[43] To the extent that respondents have specified their fears in this regard, we believe them to be unrealistic. If, as respondents assert, many presently authorized bonds cannot be sold because of the currently high market interest rates, we do not see how an increase in bond issue election successes as a result of a lower vote requirement could have seriously harmful consequences. The only effect would appear to be an increase in the amount of authorized but unsold bonds. In any event, we do not believe that current market conditions can justify the indefinite continuation of a device which so severely impairs the right to vote.

[44] Bonds which might jeopardize a community's financial stability would be unlikely to find a purchaser. Municipal bonds are valued by investors not because of their speculative possibilities but because of their great security and the favorable federal tax treatment accorded them under 26 U.S.C.A. section 103. In addition, bond purchasers are unusually sophisticated; in 1964 over 60 percent of all bondholders were institutional investors such as commercial banks and insurance companies. (Investment Bankers Association of America, Fundamentals of Municipal Bonds (4th ed. 1965) p. 21.)

[45] See also United States Advisory Commission on Intergovernmental Relations, A Commission Report: State Constitutional and Statutory Restrictions on Local Government Debt (1961) at p. 22.

[46] In City of Redondo Beach v. Taxpayers, Property Owners, etc. City of Redondo Beach, supra, 54 Cal.2d 126, we indicated that the purpose of requiring contemporaneous authorization of the necessary tax increment was to "safeguard the general funds and property of a municipality from a situation whereby the holders of an issue of bonds could, ... force an unconsented-to increase in the taxes of, or foreclose on the general assets and property of the issuing public corporation to obtain payment ...." (54 Cal.2d at p. 131.) It is obvious that a simple majority vote could authorize this tax quite as effectively as a two-thirds majority. Since this is so, the extraordinary majority requirement may not be justified on the ground it protects real property (and its owners) from the risk just described.

[47] Several provisions of our election laws seek to provide an accurate description of bond propositions and to insure that the voter has access to the viewpoints of both proponents and opponents. Thus, an analysis of the proposition made by public counsel (Elec. Code, § 3781) and a statement as to its tax cost (Elec. Code, § 5300 et seq.) are mailed to all voters. Opportunity is available for both favoring and opposing views to be included in ballot arguments and distributed to voters without cost. (Ed. Code, § 1460 et seq.; Elec. Code, §§ 3782, 3783, 5010 et seq.)

[48] Most states require some form of mandatory referendum as a condition precedent to incurring of local debt (Starner, op. cit. supra, p. 7). However, requirements vary, especially with regard to the majority required to approve a bond issue. By far the most common provision is for a simple majority. Only four states (California, Idaho, Kentucky and Missouri) require two-thirds as a general rule. (Cal. Const. Revision Com., op. cit. supra, at p. 372. Note, Judicial Activism and Municipal Bonds (1970) 56 Va. L. Rev. 295, 331-334.)

[49] A recent handbook published by the Investment Bankers Association of America, Fundamentals of Municipal Bonds (see fn. 44, ante) devotes a chapter to analyzing the credit of issuing municipalities. Factors ranging from the community's past attitude toward debt and the character of its officials, to its tax structure and the relationship of its current debt service requirements to its annual revenues are discussed. No reference is made in the credit analysis to the factor of whether the bonds were approved by two-thirds or a simple majority of the voters.

[50] Many ingenious organizational and financing arrangements have been devised to avoid the extraordinary majority barrier. For example, it is widely believed that the goal of avoiding the two-thirds vote requirement has stimulated formation of special districts which fall outside the scope of the constitutional provision. This, of course, simply increases the confusion and inefficiency which result from the artificial and complex structure of overlapping governmental units. (See Cal. Const. Revision Com., supra, pp. 367-368.) The principal financing devices which have been utilized to, in effect, incur long-term debt without securing a two-thirds vote are revenue bonds, various forms of lease-purchase financing, and joint powers authority bonds.

Revenue bonds are those which are dischargeable solely through revenues generated by the capital improvement system whose construction or enlargement they finance. They are not within the ambit of article XI, section 18 despite the fact that a pledge of the system's total revenues will diminish the financial resources of the entity that are available for general governmental expenditures, by reducing or eliminating operational revenues previously used for such purposes. However, their use is confined to the financing of projects which are income producing.

A second financing pattern which has developed is the lease of an improvement by a governmental entity from a public body that has constructed or purchased it. Initially, these lessor entities were separately managed public retirement systems which financed the improvement with retirement funds on hand. Subsequently, cities and counties wishing to construct nonvoted improvements would enter into leases with nonprofit corporations (which were often organized with the active encouragement of the lessee city or county). Such corporations were able to issue tax-exempt bonds without voter approval and apply the proceeds from the bond sales to the construction of the desired project. The project was then leased to the public entity, the nonprofit corporation pledging the lease rentals as security for its bonds.

Joint powers authority financing contemplates the formation of a separate and independent authority by joint action of two preexisting governmental units under the Joint Exercise of Powers Act. (Gov. Code, § 6500 et seq.) The authority constructs the desired project with funds derived from the sale of its bonds, which do not need extraordinary majority voter approval. It then leases the improvement to one or more of the public entities which formed it, using the lease rentals to retire the bonds.

[51] Beebe, et al., Joint Powers Authority Revenue Bonds, supra, 41 So.Cal.L.Rev. 19, contains an excellent history of state debt control and efforts to circumvent it in California. California Constitution Revision Commission, supra, pp. 360-368, provides a thorough compilation of case authority upon which the various financing techniques utilized to avoid the two-thirds requirement have been based. See also Rogers, Municipal Debt Restrictions and Lease Purchase Financing (1963) 49 A.B.A.J. 49; United States Advisory Commission on Intergovernmental Relations, op. cit. supra, at pp. 34-35; Marini, op. cit. supra, at pp. 11-16.

[52] In an address to the Extraordinary Majority Conference held on February 28, 1970, at the School of Law, University of California at Davis, under the joint sponsorship of the law school and the Institute of Governmental Affairs, Alan K. Browne, senior vice president of the Bank of America stated: "This is an era, for better or worse, of creative public finance. Where the need exists, a way will be found to fund it. The question is not whether and in what amount public improvements will be made, but rather how — by voted bonds or jerrybuilt financing vehicles; and by whom — the states and their localities, or by artificial creatures of public governments in masquerade." The full text of Browne's remarks will be published in the transcribed proceedings of the Conference on the Extraordinary Majority Requirement by the Institute of Governmental Affairs.

[53] See also Valtierra v. Housing Authority of San Jose (N.D. Cal. 1970) 313 F. Supp. 1, in which a three-judge federal district court held unconstitutional article XXXIV of the California Constitution. Relying on Hunter, the court held that the constitutional provision, which required that low-rent housing projects financed by government be approved by community residents in a mandatory referendum, violated the equal protection clause. On June 8, 1970, the United States Supreme Court noted probable jurisdiction. (38 U.S.L. Week 3485.) (Docket No. 1557, James v. Valtierra.)

[54] Marini, op. cit. supra, at page 6. Thus, while bonded indebtedness of all California cities has risen from $790,725,000 in 1954-1955 to $1,994,781,860 in 1966-1967 (California Controller, Annual Report of Financial Transactions Concerning California Cities (1966-1967) p. XI), the average city property tax rate has remained stable, increasing only from $1.83 (per $100 assessed valuation) in 1957-1958 to $1.84 (per $100 assessed valuation) in 1966-1967). (Id. at p. III.) Respondents are officials of a combined city and county whose consolidated tax rate is, of course, higher. (See fn. 2, ante.)

[55] In 1912 the property tax accounted for 82 percent of the revenue of California cities; in 1960 it accounted for only 40 percent (Marini, op. cit. supra, at pp. 5-6.) In 1966-1967 this had declined further to only 34 percent (California Controller, op. cit. supra, at p. VII).

[56] California Controller, op. cit. supra, at page XI.

[57] Marini, op. cit. supra, at page 6, footnote 4.

[58] Id., at pages 6-7; Scott and Hamilton, op. cit. supra, at page 3.

[59] In addition, these citizens directly pay a variety of local excise taxes which are available for commitment to debt service. (Cal. Const. Revision Com., op. cit. supra, at p. 375.) Of course, all citizens, whether or not they own property, share a subjective interest in and are affected by the developmental policies of their community and the projects for which bond financing is sought.

[60] It is by no means obvious that a two-thirds vote requirement would be constitutionally permissible even were this a more realistic possibility. Underlying respondents' argument is a normative assumption that, for each individual, the benefits derived from governmental action should correspond to the burdens imposed by the taxing system; i.e., that those who use public facilities should pay for them. This assumption has never been thought to describe our actual tax structure which is based, instead, on an "ability to pay" theory.

[61] In addition, the fact that a majority of the voters of the State of California may have a practicably available political remedy does not mean that petitioners or the voters of San Francisco (who represent only a small minority of the voters in the state) have one.

[62] The text of Proposition 4 as well as the arguments for and against the measure are reprinted in Beebe, op. cit. supra, at pp. 29-30, fn. 38. It is apparent from an examination of these materials that voters who would have supported a total elimination of extraordinary majority requirements for bond elections may have considered Proposition 4 a piecemeal, halfway reform measure and withheld their support on that ground. Others may have opposed the amendment because they could see no reason for singling out library bonds for preferential treatment. Still others may have felt that the consequences of the amendment's limitation of a reduced voter requirement to issues submitted at statewide elections would be undesirable.

[63] "``The protection of constitutional rights is not to be approached either pragmatically or expediently, and though the fact of enactment of a constitutional provision by heavy vote of the electorate produces pause and generates restraint we can not, true to our oath, uphold such legislation in the face of palpable infringement of rights. Thus, state racial legislation would unquestionably enjoy overwhelming electorate approval in certain of our states, yet no one would argue that this factor could compensate for manifest inequality. It is too clear for argument that constitutional law is not a matter of majority vote. Indeed, the entire philosophy of the Fourteenth Amendment teaches that it is personal rights which are to be protected against the will of the majority. The rights which are here asserted are the rights of the individual plaintiffs to have their votes counted equally with those of other voters.... [T]o say that a majority of the voters today indicate a desire to be governed by a minority, is to avoid the issue which this court is asked to resolve. It is no answer to say that the approval of the polling place necessarily evidences a rational plan. The plaintiffs have a right to expect that the cause will be determined in relation to the standards of equal protection. Utilization of other or different standards denies them full measure of justice.'" (Lisco v. Love (1963) 219 F. Supp. 922, 944 (dissenting opinion); quoted in Lucas v. Colorado General Assembly, supra, 377 U.S. 713, 737, fn. 30 [12 L.Ed.2d 632, 647-648].)

[64] Respondents include in their compilation the following extraordinary majority provisions:

A) United States Constitution: article 1, section 3 (conviction in an impeachment proceeding); article 1, section 5 (expulsion from Congress); article 1, section 7 (overriding of an executive veto); article II, section 2 (ratification of treaties); and article V (constitutional amendment process).

B) California Constitution: article IV, section 8(a) (elimination of 30-day waiting period before consideration of bill); article IV, section 8(d) (passage of emergency legislation); article IV, section 10(a) (override of executive veto of general bills); article IV, section 10(b) (override of executive reduction of appropriations bill); article IV, section 12(d) (appropriations from the general fund); article XI, section 7 1/2 (surrender and annulment of county charter); article XI, section 2 (removal of county seat); article XII, section 22 (removal from office of Public Utilities Commission member); article XIII, section 14-4/5 (i) (change in rate of taxation on insurance companies); article XIII, section 16.3 (imposition of tax on banks); article XIII, section 14 (classification of personal property for taxation and assessment purposes); article XVI, section 1 (submission of state bond issue to voters); article XVIII, section 1 (submission of proposed constitutional amendment to voters).

An indication of the varied contexts in which extraordinary majority vote requirements are found is contained in Comment, Extraordinary Majority Voting Requirements (1969) 58 Geo. L.J. 411, 412-413, fns. 10-15.

[65] Our decision herein extends, of course, to all statutory provisions which implement the constitutional directive. We thus declare unconstitutional those portions of Government Code section 43614 and Education Code section 21754 which incorporate the two-thirds vote requirement in their direction of procedural steps to be followed in the issuance of, respectively, municipal and school district bonds. On the other hand, we need not deal directly with the relevant sections of the Charter of the City and County of San Francisco, since they will automatically correspond to the effect of our decision on article XI, section 18.

[*] Retired Associate Justice of the Supreme Court sitting under assignment by the Chairman of the Judicial Council.

[1] The majority conclusion and the timing of this opinion are particularly unfortunate. There were 26 bond issues voted upon this very month, at the June 2 primary election. Some of them received overwhelming, but not two-thirds, approval. (E.g., Proposition B (light bonds) in San Francisco was favored by a vote of 112,575 to 62,261; Proposition C (fire equipment bonds) in San Francisco received a favorable vote of 107,136 to 65,357.) It seems palpably unfair to doom those bond propositions to defeat when, had this opinion been rendered a month earlier or the election held a month later, the majority vote would have been deemed sufficient for passage of the measures. Despite such fortuity, the same Constitution was in effect on June 2 that governs us today.