*588Opinion

TOBRINER, J.

We face today the question of the validity of an initiative ordinance enacted by the voters of the City of Livermore which prohibits issuance of further residential building permits until local educational, sewage disposal, and water supply facilities comply with specified standards.1 Plaintiff, an association of contractors, subdividers, and other persons interested in residential construction in Livermore, brought this suit to enjoin enforcement of the ordinance. The superior court issued a permanent injunction, and the city appealed.

In Hurst v. City of Burlingame (1929) 207 Cal. 134 [277 P. 308], we held that statutes requiring notice and hearing to precede enactment of municipal zoning and land use ordinances applied to initiatives, a holding which effectively denied voters of general law cities the power to enact such legislation by initiative. In accord with that precedent, the trial court here held that Livermore, as a general law city, lacked authority to enact the initiative ordinance at issue. We have concluded, however, that Hurst was incorrectly decided; the statutory notice and hearing provisions govern only ordinances enacted by city council action and do not limit the power of municipal electors, reserved to them by the state Constitution, to enact legislation by initiative. We therefore reverse the trial court holding on this issue.

We also reject the trial court’s alternative holding that the ordinance is unconstitutionally vague. By interpreting the ordinance to incorporate standards established by the Livermore Valley Joint School District and the Regional Water Quality Control Board, we render its terms sufficiently specific to comply with constitutional requisites. The failure of the ordinance to designate the person or agency who determines when its standards have been fulfilled does not make it unconstitutionally vague; the duty to enforce the ordinance reposes in the city’s building inspector, whose decisions are subject to judicial review by writ of mandamus.

Finally, we reject plaintiff’s suggestion that we sustain the trial court’s injunction on the ground that the ordinance unconstitutionally attempts *589to bar immigration to Livermore. Plaintiff’s contention symbolizes the growing conflict between the efforts of suburban communities to check disorderly development, with its concomitant problems of air and water pollution and inadequate public facilities, and the increasing public need for adequate housing opportunities. We take this opportunity, therefore, to reaffirm and clarify the principles which govern validity of land use ordinances which substantially limit immigration into a community; we hold that such ordinances need not be sustained by a compelling state interest, but are constitutional if they are reasonably related to the welfare of the region affected by the ordinance. Since on the limited record before us plaintiff has not demonstrated that the Livermore ordinance lacks a reasonable relationship to the regional welfare, we cannot hold the ordinance unconstitutional under this standard.

1. Summary of proceedings.

The initiative ordinance in question was enacted by a majority of the voters at the Livermore municipal election of April 11, 1972, and became effective on April 28, 1972. The ordinance, set out in full in the margin,2 *590states that it was enacted to further the health, safety, and welfare of the citizens of Livermore and to contribute to the solution of air pollution. Finding that excessive issuance of residential building permits has caused school overcrowding, sewage pollution, and water rationing, the ordinance prohibits issuance of further permits until three standards are met: “1. Educational Facilities—No double sessions in the schools nor overcrowded classrooms as determined by the California Education Code. 2. Sewage—The sewage treatment facilities and capacities meet the standards set by the Regional Water Quality Control Board. 3. Water Supply—No rationing of water with respect to human consumption or irrigation and adequate water reserves for fire protection exist.”

Plaintiff association filed suit to enjoin enforcement of the ordinance and for declaratoiy relief. After the city filed its answer, all parties moved for judgment on the pleadings and stipulated that the court, upon the pleadings and other documents submitted, could determine the merits of the cause. On the basis of that stipulation the court rendered findings and entered judgment for plaintiff. The city appeals from that judgment.

2. The enactment of the Livermore ordinance by initiative does not violate the state zoning law.

The superior court found that the initiative ordinance was adopted “without complying with the statutes . . . governing general law cities,” specifically Government Code sections 65853 through 65857. These sections provide that any ordinance which changes zoning ór imposes a land use restriction listed in Government Code section 65850 can be enacted only after noticed hearing before the city’s planning commission and legislative body.3 The superior court concluded that notice and *591hearing must precede enactment of any ordinance regulating land use. Since Livermore passed its ordinance pursuant to the procedures specified in the statutes governing municipal initiatives (Elec. Code, § 4000 et seq.), which do not provide for hearings before the city planning commission or council, the court held the ordinance invalid.

The amendment of the California Constitution in 1911 to provide for the initiative and referendum signifies one of the outstanding achievements of the progressive movement of the early 1900’s.4 Drafted in light of the theory that all power of government ultimately resides in the people, the amendment speaks of the initiative and referendum, not as a right granted the people, but as a power reserved by them.5 Declaring it “the duty of the courts to 'jealously guard this right of the people” (Martin v. Smith (1959) 176 Cal.App.2d 115, 117 [1 Cal.Rptr. 307]), the courts have described the initiative and referendum as articulating “one of the most precious rights of our democratic process” (Mervynne v. Acker, supra, 189 Cal.App.2d 558, 563). “[I]t has long been our judicial policy to apply a liberal construction to this power wherever it is challenged in order that the right be not improperly annulled. If doubts can reasonably be resolved in favor of the use of this reserve power, courts will preserve it.” (Mervynne v. Acker, supra, 189 Cal.App.2d 558, 563-564; Gayle v. Hamm, supra, 25 Cal.App.3d 250, 258.)6

The 1911 amendment, in reserving the right of initiative to electors of counties and cities, authorized the Legislature to establish procedures to facilitate the exercise of that right.7 Accordingly the Legislature enacted *592statutes, now codified as sections 4000-4023 of the Election Code, providing for the circulation of petitions, the calling of elections, and other procedures required to enact an initiative measure.

The 1911 amendment was first applied to zoning matters in 1927 in Dwyer v. City Council, supra, 200 Cal. 505, in which the court mandated the Berkeley City Council to submit a zoning ordinance to referendum. The opinion reasoned that since the city council had the legislative authority to enact zoning ordinances, the people had the power to do so by initiative or referendum. Rejecting an argument that the referendum procedure denied affected persons the right, granted them by municipal ordinance, to appear before the city council and state their views on the ordinance, the court replied that “the matter has been removed from the forum of the Council to the forum of the electorate. The proponents and opponents are given all the privileges and rights to express themselves in an open election that a democracy or republican form of government can afford to its citizens. ... It is clear that the constitutional right reserved by the people to submit legislative questions to a direct vote cannot be abridged by any procedural requirement....” (200 Cal. at p. 516.)

Two years later the court decided Hurst v. City of Burlingame, supra, 207 Cal. 134, the decision on which the trial court in the instant case based its ruling. The City of Burlingame had enacted by initiative a city-wide zoning ordinance which classified as residential the property where plaintiff had a retail store. Contending that he had been denied the right to a public hearing established in the Zoning Act of 1917 (Stats. 1917, ch. 734, p. 1419), plaintiff sued to enjoin enforcement of the ordinance. Beginning with the premise that “an ordinance proposed by the electors of a county or of a city in this state under the initiative law must constitute such legislation as the legislative body of such county or city has the power to enact . . . .” (207 Cal. at p. 140), the Hurst court reasoned that since the board of trustees of the City of Burlingame could not lawfully enact a zoning ordinance without complying with the hearing requirement of the state law, the voters could not adopt such an ordinance by initiative.

*593Responding to the argument that the enactment of the ordinance complied with the state initiative law, the court stated that “The initiative law and the zoning law are hopelessly inconsistent and in conflict as to the manner of the preparation and adoption of a zoning ordinance. The Zoning Act is a special statute dealing with a particular subject and must be deemed to be controlling over the initiative, which is general in its scope.” (P. 141.) Finally, the court distinguished Dwyer v. City Council, supra, 200 Cal. 505, on the ground that Dwyer upheld a referendum, and thus persons affected by the referendum had already been granted a right to notice and hearing' at the time of the original enactment of the ordinance. (See p. 142.)

Although Hurst thus held the Burlingame initiative invalid for noncompliance with the state zoning law, the court added a constitutional dictum, asserting that “the statutory notice and hearing . . . becomes necessary in order to satisfy the requirements of due process. . . .” (P. 141.) In later years this constitutional dictum overshadowed the statutory holding of Hurst. Courts and commentators alike questioned Hurst’s statutory holding,8 but reexamination of that holding seemed pointless if the landowner’s right to notice and hearing derived from constitutional compulsion independent of statute.9

Two years ago, however, in San Diego Bldg. Contractors Assn. v. City Council (1974) 13 Cal.3d 205, 216 [118 Cal.Rptr. 146, 529 P.2d 570, 72 A.L.R.3d 973] (app. dism. (1976) 427 U.S. 901 [49 L.Ed.2d 1195, 96 S.Ct. 3184]) we expressly disapproved the constitutional dictum of Hurst and later decisions. We held that a city violates no constitutional prohibition in enacting a zoning ordinance without notice and hearing to landowners, and hence may do so by initiative. (13 Cal.3d at pp. 217-218.) That decision clears the way for a long-needed reconsideration of the actual holding of Hurst that bars a general law city from enacting a zoning ordinance by initiative.

At first glance it becomes apparent that something must be wrong with the reasoning in Hurst. Starting from a premise of equálity—that the voters possess only the same legislative authority as does the city *594 council—Hurst arrived at the conclusion that only the council and not the voters had the authority to enact zoning measures. Thus in the name of equality Hurst decrees inequality. The errors which lead to this non sequitur appear after further analysis.

First, Hurst, erroneously contriving a conflict between state zoning statutes and the initiative law, set out to resolve that presumed conflict.10 No conflict occurs, however; the Legislature never intended the notice and hearing requirements of the zoning law to apply to the enactment of zoning initiatives. (See Comment, The Initiative and Referendum’s Use in Zoning (1976) 64 Cal.L.Rev. 74, 104-105.) The Legislature plainly drafted the questioned provisions of the zoning law with a view to ordinances adopted by vote of the city council; the provisions merely add certain additional procedural requirements to those already specified in Government Code sections 36931-36939 for the enactment of ordinances in general. Procedural requirements which govern council action, however, generally do not apply to initiatives,11 any more than the provisions of the initiative law govern the enactment of ordinances in council. No one would contend, for example, that an initiative of the people failed because a quorum of councilmen had not voted upon it, any more than one would contend that an ordinary ordinance of a council failed because a majority of voters had not voted upon it.

In the second place, Hurst, in treating the case as one involving a conflict between two statutes of equal status—the zoning law and the initiative law—overlooked a crucial distinction: that although the procedures for exercise of the right of initiative are spelled out in the *595initiative law, the right itself is guaranteed by the Constitution. The 1911 constitutional amendment, in reserving the right of initiative on behalf of municipal voters, stated that “This section is self-executing, but legislation may be enacted to facilitate its operation, but in no way limiting or restricting either the provisions of this section or the powers herein reserved(Former Cal. Const., art. IV, § 1.) (Italics added.)12 Although the Legislature can specify the manner in which general law cities enact ordinances restricting land use,13 legislation which permits council action but effectively bars initiative action may run afoul of the 1911 amendment. (See Comment, op. cit., supra, 64 Cal.L.Rev. 74, 102.) Thus the notice and hearing provisions of the state zoning law, if interpreted to bar initiative land use ordinances, would be of doubtful constitutionality.; all such doubt dissolves in the light of an interpretation which limits those requirements to ordinances enacted by city councils.

The fact that the zoning law is a special statute will not support Hurst; special legislation is still subject to constitutional limitations. If, for example, a “special” statute were enacted prohibiting criticism of a named official, such as the Vice-President, it would not be deemed controlling over the First Amendment on the ground that the latter is “general in its scope.” Indeed if the constitutional power reserved by the people can be abridged by special statutes, then by enacting a host of special statutes the Legislature could totally abrogate that power.

Finally, Hurst erred in distinguishing Dwyer v. City Council, supra, 200 Cal. 505, on the ground that Dwyer involved a referendum on a zoning ordinance; as Dwyer itself pointed out, “if the right of referendum can be invoked, the corollary right to initiate legislation must be conceded to exist.” (200 Cal. at p. 511.)

*596Thus both precedent and established principles of judicial construction dictate the conclusion that Hurst erred in holding the notice and hearing provisions of the Zoning Act of 1917 applied to zoning ordinances enacted by initiative. Resting upon the precepts that statutes which are apparently in conflict should, if reasonably possible, be reconciled (see, e.g., Warne v. Harkness (1963) 60 Cal.2d 579, 588 [35 Cal.Rptr. 601, 387 P.2d 377]; Pacific Motor Transport Co. v. State Bd. of Equalization (1972) 28 Cal.App.3d 230, 235 [104 Cal.Rptr. 558]); that a statute should be construed to “eliminate . . . doubts as to the provision’s constitutionality” (In re Kay (1970) 1 Cal.3d 930, 942 [83 Cal.Rptr. 686, 464 P.2d 142]); that the initiative power must be broadly construed, resolving all doubts in favor of the reserved power (see cases cited pp. 591-592, ante, and fn. 6), we resolve that Hurst v. Burlingame, supra, 207 Cal. 134, was incorrectly decided and is therefore overruled.14

The notice and hearing provisions of the present zoning law (Gov. Code, §§ 65853-65857), like the provisions of the 1911 law before the Hurst court, make no mention of zoning by initiative. The procedures they prescribe refer only to action by the city council, and are inconsistent with the regulations that the Legislature has established to govern enactment of initiatives. For the reasons stated in our discussion of Hurst v. Burlingame, supra, we conclude that sections 65853-65857 do not apply to initiative action, and that the Livermore ordinance is not invalid for noncompliance with those sections.

3. The Livermore ordinance is not voidfor vagueness.

The trial court found the ordinance unconstitutionally vague on two grounds: (1) that the ordinance did not contain sufficiently specific *597standards for the issuance or denial of building permits, and (2) that it did not specify what person or agency was empowered to determine if the ordinance’s standards have been met. W,e disagree with both rationales and find the ordinance sufficiently specific to fulfill constitutional requirements.

The controversy concerning the specificity of the ordinance centers upon the standard as to education. The ordinance prohibits issuance of residential building permits until a “satisfactory solution” has been evolved to the problem of “Educational Facilities;” it. defines a satisfactory solution as one characterized by “No double sessions in the schools nor overcrowded classrooms as determined by the California Education Code.”

The term “double sessions” is sufficiently specific; as stated by Professor Deutsch, it “can be defined by reference to common practice, since the term is frequently used to refer to a situation where different groups of students in the same grade are attending the same, school at different times of the day because of a lack of space.” (Deutsch, op. cit., supra, pp. 22-23.) The phrase “overcrowded classrooms as determined by the California Education Code,” however, is less clear, since nowhere in the Education Code does there appear a definition of “overcrowded classrooms.”

The City of Livermore, however, points out that the ordinance does not refer to a definition of “overcrowded classrooms” contained in the Education Code, but to a determination of that subject. The language, it contends—and plaintiff does not dispute the contention—was intended to refer to resolution 3220, adopted by the board of the Livermore Valley Joint School District on January 18, 1972, in which that board, pursuant to authority granted it by Education Code section 1052, established clear and specific standards for determining whether schools are overcrowded.15

*598Rather than interpret the ordinance in a manner which would expose it to the charge of unconstitutional vagueness, we adopt the suggestion of the city and construe the ordinance’s standard on education to incorporate the specific guidelines established in board resolution 3220. In so doing we conform to the rule that enactments should be interpreted when possible to uphold their validity (see San Francisco Unified School Dist. v. Johnson (1971) 3 Cal.3d 937, 948 [92 Cal.Rptr. 309, 479 P.2d 669]), and the corollary principle that courts should construe enactments to give specific content to terms that might otherwise be unconstitutionally vague. (See Bloom v. Municipal Court (1976) 16 Cal.3d 71 [127 Cal.Rptr. 317, 545 P.2d 229]; In re Kay, supra, 1 Cal.3d 930.)

Our decision in Braxton v. Municipal Court (1973) 10 Cal.3d 138 [109 Cal.Rptr. 897, 514 P.2d 697], illustrates the principle and provides a close analogy to the present case. In Braxton, we construed Penal Code section 626.4, which authorized a state college or university to bar from its campus anyone who had “disrupted” the orderly operation of the campus. Defendants argued that the term “disrupted” was unconstitutionally vague. We determined, however, that the Legislature had intended to authorize banishment only of persons who had violated other more specific criminal statutes. Although section 626.4 did not expressly refer to such other statutes, we interpreted section 626.4 to incorporate the specific standards set out in those statutes in order to uphold the constitutionality of the section. (10 Cal.3d at p. 152.)

Following the course suggested by Braxton, we construe the Livermore ordinance to incorporate the standards for determining the *599overcrowded condition of schools contained in the school board resolution of January 18, 1972. So construed, the ordinance provides a clear and ascertainable educational standard to guide the issuance or denial of a building permit, and is not void for vagueness.

The ordinance’s standards relating to sewage and water supply present no constitutional difficulties. The sewage provision incorporates the “standards set by the Regional Water Quality Control Board;” that agency has in fact established specific and detailed standards of water purification and sewage disposal.16 The water supply provision describes a “satisfactory solution” as one in which water is not rationed, and “adequate water reserves for fire protection exist.” The existence of rationing is an objective fact which can be ascertained by inquiry to the agencies having authority to ration.17 Although individuals may differ as to the adequacy of reserves for fire protection, the considered judgment of the agencies responsible for fire protection would provide a reliable guide.

Although we háve determined that the ordinance’s standards meet constitutional requirements of certainty, plaintiff argues, and the trial court held, that the ordinance is void because it fails to designate what agency or person determines whether these standards have been achieved. We question plaintiff’s underlying assumption that an ordinance or statute is void if it does not specify on its face the agency that is to adjudge disputes concerning its application; by such a test most of the civil and criminal laws of this state would be invalidated. In any event, we believe that the Livermore ordinance, read in the light of the structure of Livermore’s city government and the applicable judicial decisions, does indicate the method by which disagreements concerning the ordinance’s standards are resolved.

The Livermore ordinance establishes standards to govern the issuance or denial of residential building permits. These standards must be *600directed in the first instance to the city building inspector, the official charged with the duty of issuing or denying such permits. Since the duties of this official are ministerial in character, his decisions can be reviewed by writ of mandamus. (McCombs v. Larson (1959) 176 Cal.App.2d 105, 107 [1 Cal.Rptr., 140]; Palmer v. Fox (1953) 118 Cal.App.2d 453 [258 P.2d 30].) Thus the ultimate decision as to compliance with the standards will be rendered by the courts. (See generally Hagman et al., Cal. Zoning Practice (Cont.Ed.Bar 1969) § 12.4.)

4. On the limited record before us, plaintiff cannot demonstrate that the Livermore ordinance is riot a constitutional exercise of the city’s police power.

Plaintiff urges that we affirm the trial court’s injunction on a ground which it raised below, but upon which the trial court did not rely. Plaintiff contends that the ordinance proposes, and will cause, the prevention of nonresidents from migrating to Livermore, and that the ordinance therefore attempts an unconstitutional exercise of the police power, both because no compelling state interest justifies its infringement upon the migrant’s constitutionally protected right to travel, and because it exceeds the police power of the municipality.18

The ordinance on its face imposes no absolute prohibition or limitation upon population -growth or residential construction. It does provide that no building permits will issue unless standards for educational facilities, water supply and sewage disposal have been met, but plaintiff presented no evidence to show that the ordinance’s standards were unreasonable or unrelated to their apparent objectives of protecting the public health and welfare. Thus, we do not here confront the question of the constitutionality of an ordinance which limits or bars population growth either directly in express language or indirectly by the imposition of prohibitory standards; we adjudicate only the validity of *601an ordinance limiting building permits in accord with standards that reasonably measure the adequacy of public services.

As we shall explain, the limited record here prevents us from resolving that constitutional issue. We deal here with a case in which a land use ordinance is challenged solely on the ground that it assertedly exceeds the municipality’s authority under the police power; the challenger eschews any claim that the ordinance discriminates on a basis of race or wealth. Under such circumstances, we view the past decisions of this court and the federal courts as establishing the following standard: the land use restriction withstands constitutional attack if it is fairly debatable that the restriction in fact bears a reasonable relation to the general welfare. For the guidance of the trial court we point out that if a restriction significantly affects residents of surrounding communities, the constitutionality of the restriction must be measured by its impact not only upon the welfare of the enacting community, but upon the welfare of the surrounding region. We explain the process by which the court can determine whether or not such a restriction reasonably relates to the regional welfare. Since the record in the present case is limited to the pleadings and stipulations, and is devoid of evidence concerning the probable impact and duration of the ordinance’s restrictions, we conclude that we cannot now adjudicate the constitutionality of the ordinance. Thus we cannot sustain the trial court judgment on the ground that the ordinance exceeds the city’s authority under the police power; that issue can be resolved only after trial.

We turn now to consider plaintiff’s arguments in greater detail. Seeking to capitalize upon the absence of an evidentiary record, plaintiff contends that the challenged ordinance must be subjected to strict judicial scrutiny; that it can be sustained only upon a showing of a compelling interest, and that the city has failed to make that showing.

Many writers have contended that exclusionary land use ordinances tend primarily to exclude racial minorities and the poor, and on that account should be subject to strict judicial scrutiny. (See, e.g., Davidoff & Davidoff, Opening the Suburbs: Toward Inclusionary Land Use Controls (1971) 22 Syracuse L.Rev. 509; Sager, Tight Little Islands: Exclusionary Zoning, Equal Protection, and the Indigent (1969) 21 Stan.L.Rev. 767; Note, Phased Zoning: Regulation of the Tempo and Sequence of Land Development, 26 Stan.L.Rev. 585, 597, fn. 45 and authorities there cited; *602Note, The Equal Protection Clause and Exclusionary Zoning after Valtierra and Dandridge (1971) 81 Yale L.J. 61.) These writers, however, are concerned primarily with ordinances which ban or limit less expensive forms of housing while permitting expensive single family residences on large lots. The Livermore ordinance is not made from this mold; it impartially bans all residential construction, expensive or inexpensive. Consequently plaintiff at bar has eschewed reliance upon any claim that the ordinance discriminates on a basis of race or wealth.

Plaintiff’s contention that the Livermore ordinance must be tested by a standard of strict scrutiny, and can be sustained only upon a showing of a compelling state interest, thus rests solely on plaintiff’s assertion that the ordinance abridges a constitutionally protected right to travel. As we shall explain, however, the indirect burden imposed on the right to travel by the ordinance does not warrant application of the plaintiff’s asserted standard of “compelling interest.”19

In asserting that legislation which burdens a right to travel requires strict scrutiny, and can be sustained only upon proof of compelling need, plaintiff relies on recent decisions of this court (In re King (1970) 3 Cal.3d 226 [90 Cal.Rptr. 15, 474 P.2d 983]) and the United States Supreme Court (Memorial Hospital v. Maricopa County (1974) 415 U.S. 250 [39 L.Ed.2d 306, 94 S.Ct. 1076]; Dunn v. Blumstein (1972) 405 U.S. 330 [31 L.Ed.2d 274, 92 S.Ct. 995]; Shapiro v. Thompson (1969) 394 U.S. 618 [22 L.Ed.2d 600, 89 S.Ct. 1322]). The legislation held invalid by those decisions, however, directly burdened the right to travel by distinguishing between nonresidents or newly arrived residents on the one hand and established residents on the other, and imposing penalties or disabilities on the former group.20

Both the United States Supreme Court and this court have refused to apply the strict constitutional test to legislation, such as the present *603ordinance, which does not penalize travel and resettlement but merely makes it more difficult for the outsider to establish his residence in the place of his choosing.21 (See Village of Belle Terre v. Boraas (1973) 416 U.S. 1, 7 [39 L.Ed.2d 797, 803, 94 S.Ct. 1536]; Ector v. City of Torrance (1973) 10 Cal.3d 129, 135 [109 Cal.Rptr. 849, 514 P.2d 433]; see also McCarthy v. Philadelphia Civil Serv. Comm’n (1976) 424 U.S. 645 [47 L.Ed.2d 366, 96 S.Ct. 1154]; Construction Ind. Ass’n, Sonoma Cty. v. City of Petaluma, supra, 522 F.2d 897, 906-907, fn. 13; Note, 50 N.Y.U.L.F. (1975) 1163, 1168.) The only contrary authority, the decision of the federal district court in Construction Ind. Ass’n, Sonoma Cty. v. City of Petaluma (N.D.Cal. 1974) 375 F.Supp. 574 holding that an ordinance limiting residential construction must be supported by a compelling state interest has now been reversed by the Court of Appeals for the Ninth Circuit. (Construction Ind. Ass’n, Sonoma Cty. v. City of Petaluma, supra, 522 F.2d 897, cert. den., 424 U.S. 934 [47 L.Ed.2d 342, 96 S.Ct. 1148].)

Most zoning and land use ordinances affect population growth and density. (See Construction Ind. Ass’n, Sonoma Cty v. City of Petaluma, supra, 522 F.2d 897, 906; Note, op. cit., supra, 26 Stan.L.Rev. 585, 606-607, fn. 91.) As commentators have observed, to insist that such zoning laws are invalid unless the interests supporting the exclusion are compelling in character, and cannot be achieved by an alternative method, would result in wholesale invalidation of land use controls and endanger the validity of city and regional planning. (See Note, op. cit., supra, 26 Hastings L.J. 849, 854.) “Were a court to . . . hold that an inferred right of any group to live wherever it chooses might not be abridged without some compelling state interest, the law of zoning would be literally turned upside down; presumptions of validity would become presumptions of invalidity and traditional police powers of a state would be severely circumscribed.” (Comment, Zoning, Communes and Equal Protection, 1973 Urban L.Ann. 319, 324.)

We conclude that the indirect burden upon the right to travel imposed by the Livermore ordinance does not call for strict judicial scrutiny. The validity of the challenged ordinance must be measured by the more *604liberal standards that have traditionally tested the validity of land use restrictions enacted under the municipal police power.22

This conclusion brings us to plaintiff’s final contention: that the Livermore ordinance exceeds the authority conferred upon the city under the police power. The constitutional measure by which we judge the validity of a land use ordinance that is assailed as exceeding municipal authority under the police power dates in California from the landmark decision in Miller v. Board of Public Works (1925) 195 Cal. 477 [234 P. 381, 38 A.L.R. 1479]. Upholding a Los Angeles ordinance which excluded commercial and apartment uses from certain residential zones, we declared that an ordinance restricting land use was valid if it had a “real or substantial relation to the public health, safety, mprals or general welfare.” (195 Cal. at p. 490.) A year later the United States Supreme Court, in the landmark case of Euclid v. Ambler Co. (1926) 272 U.S. 365 [71 L.Ed. 303, 47 S.Ct. 114, 54 A.L.R. 1016], adopted the same test, holding that béfore a zoning ordinance can be held unconstitutional, “it must be said . . . that [its] provisions are clearly arbitrary and unreasonable, having no substantial relation to the public health, safety, morals, or general welfare.” (272 U.S. at p. 395 [71 L.Ed. at p. 314].) Later California decisions confirmed that a land use restriction lies within the public power if it has a “reasonable relation to the public welfare.” (Lockard v. City of Los Angeles (1949) 33 Cal.2d 453, 461 [202 P.2d 38, 7 A.L.R.2d 990]; Hamer v. Town of Ross (1963) 59 Cal.2d 776, 783 [31 Cal.Rptr. 335, 382 P.2d 375]; see Town of Los Altos Hills v. Adobe Creek Properties, Inc., supra, 32 Cal.App.3d 488, 508-509 and cases there cited.)

In deciding whether a challenged ordinance reasonably relates to the public welfare, the courts recognize that such ordinances are presumed *605to be constitutional, and come before the court with every intendment in their favor. (Lockard v. City of Los Angeles, supra, 33 Cal.2d 453, 460.) “The courts may differ with the zoning authorities as to the ‘necessity or propriety of an enactment,’ but so long as it remains a ‘question upon which reasonable minds might differ,’ there will be no judicial interference with the municipality’s determination of policy.” (Clemons v. City of Los Angeles (1950) 36 Cal.2d 95, 98 [222 P.2d 439].) In short, as stated by the Supreme Court in Euclid v. Ambler Co., supra, “If the validity . . . be fairly debatable, the legislative judgment must be allowed to control.” (272 U.S. 365, 388 [71 L.Ed. 303, 311].)

Recent decisions of the United States Supreme Court and the Court of Appeals for the Ninth Circuit have applied this liberal standard and, deferring to legislative judgment, have upheld ordinances attacked as exclusionary. In Village of Belle Terre v. Boraas, supra, 416 U.S. 1 [39 L.Ed.2d 797], the court sustained an ordinance which banned all multiple family housing. The majority opinion by Justice Douglas found a rational basis for the ordinance in the community’s desire to preserve a pleasant environment; “[t]he police power,” he asserted, “is not confined to the elimination of filth, stench, and unhealthy places. It is ample to lay out zones where family values, youth values, and the blessings of quiet seclusion and clean air make the area a sanctuary for people.” (416 U.S. at p. 9 [39 L.Ed.2d at p. 804].) In dissent, Justice Marshall argued that the village’s exclusion of groups of three or more unrelated persons from living in a single residence violated protected rights of privacy and association. He agreed, however, that the village could properly enact ordinances to control population density and restrict uncontrolled growth so long as it did not abridge fundamental rights, and that in reviewing such ordinances the courts should defer to the legislative judgment. (See 416 U.S. at pp. 13, 19-20 [39 L.Ed.2d at pp. 806, 810-811].)

In Construction Ind. Ass’n, Sonoma Cty. v. City of Petaluma, supra, 522 F.2d 897, the Ninth Circuit Court of Appeals upheld a city ordinance fixing a housing development growth rate of 500 units per year. Relying largely on Belle Terre v. Boraas, supra, 416 U.S. 1 [39 L.Ed.2d 797], the court concluded that “the concept of public welfare is sufficiently broad *606to uphold Petaluma’s desire to preserve its small town character, its open space and low density of population, and to grow at an orderly and deliberate pace.” (522 F.2d at pp. 908-909.) The Supreme Court denied certiorari. (424 U.S. 934 [47 L.Ed.2d 342, 96 S.Ct. 1148].)

We conclude from these federal decisions that when an exclusionary ordinance is challenged under the federal due process clause, the standard of constitutional adjudication remains that set forth in Euclid v. Ambler Co., supra, 272 U.S. 365 [71 L.Ed. 303]: if it is fairly debatable that the ordinance is reasonably related to the public welfare, the ordinance is constitutional. A number of recent decisions from courts of other states, however, have declined to accord the traditional deference to legislative judgment in the review of exclusionaiy ordinances, and ruled that communities lacked authority to adopt such ordinances. Plaintiff urges that we apply the standards of review employed in those decisions in passing upon the instant ordinance.

The cases cited by plaintiff, however, cannot serve as a guide to resolution of the present controversy. Not only do those decisions rest, for the most part, upon principles of state law inapplicable in California, but, unlike the present case, all involve ordinances which impede the ability of low or moderate income persons to immigrate to a community but permit largely unimpeded entry by wealthier persons.23

*607We therefore reaffirm the established constitutional principle that a local land use ordinance falls within the authority of the police power if it is reasonably related to the public welfare. Most previous decisions applying this test, however, have involved ordinances without substantial effect beyond the municipal boundaries. The present ordinance, in contrast, significantly affects the interests of nonresidents who are not represented in the city legislative body and cannot vote on a city initiative. We therefore believe it desirable for the guidance of the trial court to clarify the application of the traditional police power test to an ordinance which significantly affects nonresidents of the municipality.

When we inquire whether an ordinance reasonably relates to the public welfare, inquiiy should begin by asking whose welfare must the ordinance serve. In past cases, when discussing ordinances without significant effect beyond the municipal boundaries, we have been content to assume that the ordinance need only reasonably relate to the welfare of the enacting municipality and its residents. But municipalities are not isolated, islands remote from the needs and problems of the area in which they are located; thus an ordinance, superficially reasonable from the limited viewpoint of the municipality, may be disclosed as unreasonable when viewed from a larger perspective.

These considerations impel us to the conclusion that the proper constitutional test is one which inquires whether the ordinance reasonably relates to the welfare of those whom it significantly affects. If its impact is limited to the city boundaries, the inquiiy may be limited accordingly; if, as alleged here, the ordinance may strongly influence the supply and distribution of housing for an entire metropolitan region, judicial inquiiy must consider the welfare of that region.24

As far back as Euclid v. Ambler Co., courts recognized “the possibility of cases where the general public interest would so far outweigh the „ interest of the municipality that the municipality would not be allowed to stand in the way.” (272 U.S. 365, 390 [71 L.Ed. 303, 311].) More recently, in Scott v. City of Indian Wells (1972) 6 Cal.3d 541 [99 Cal.Rptr. *608745, 492 P.2d 1137], we stated that “To hold . . . that defendant city may zone the land within its border without any concern for [nonresidents] would indeed ‘make a fetish out. of invisible municipal boundary lines and a mockery of the principles of zoning.’ ” (P. 548.) The New Jersey Supreme Court summed up the principle and explained its doctrinal basis: “[I]t is fundamental and not to be forgotten that the zoning power is a police power of the state and the local authority is acting only as a delegate of that power and is restricted in the same manner as is the state. So, when regulation does have a substantial external impact, the welfare of the state’s citizens beyond the borders of the particular municipality cannot be disregarded and must be recognized and served.” (So. Burlington Cty. N.A.A.C.P. v. Tp. of Mt. Laurel, supra, 336 A.2d 713, 726.)25

We explain the process by which a trial court may determine whether a challenged restriction reasonably relates to the regional welfare. The first step in that analysis is to forecast the probable effect and duration of the restriction. In the instant case the Liver-more ordinance posits a total ban on residential construction, but one which terminates as soon as public facilities reach specified standards. Thus to evaluate the impact of the restriction, the court must ascertain the extent to which public facilities currently fall short of the specified standards, must inquire whether the city or appropriate regional agencies have undertaken to construct needed improvements, and must determine when the improvements are likely to be completed.

The second step is to identify the competing interests affected by the restriction. We touch in this area deep social antagonisms. We allude to the conflict between the environmental protectionists and the egalitarian humanists; a collision between the forces that would save the benefits of nature and those that would preserve the opportunity of people in general to settle. Suburban residents who seek to overcome problems of inadequate schools and public facilities to secure “the blessing of quiet seclusion and clean air” and to “make the area a sanctuary for people” (Village of Belle Terre v. Boraas, supra, 416 U.S. 1, *6099 [39 L.Ed.2d 797, 804]) may assert a vital interest in limiting immigration to their community. Outsiders searching for a place to live in the face of a growing shortage of adequate housing, and hoping to share in the perceived benefits of suburban life, may present a countervailing interest opposing barriers to immigration.

Having identified and weighed the competing interests, the final step is to determine whether the ordinance, in light of its probable impact, represents a reasonable acommodation of the competing interests.26 We do not hold that a court in inquiring whether an ordinance reasonably relates to the regional welfare, cannot defer to the judgment of the municipality’s legislative body.27 But judicial deference is not judicial abdication. The ordinance must have a real and substantial relation to the public welfare. (Miller v. Board of Public Works, supra, 195 Cal. 477, 490.) There must be a reasonable basis in fact, not in fancy, to support the legislative determination. (Consolidated Rock Products Co. v. City of Los Angeles (1962) 57 Cal.2d 515, 522 [20 Cal.Rptr. 638, 370 P.2d 342].) Although in many cases it will be “fairly debatable” (Euclid v. Ambler Co., supra, 272 U.S. 365, 388 [71 L.Ed. 303, 311]) that the ordinance reasonably relates to the regional welfare, it cannot be assumed that a land use ordinance can never be invalidated as an enactment in excess of the police power.

The burden rests with the party challenging the constitutionality of an ordinance to present the evidence and documentation which the court will require in undertaking this constitutional analysis. Plaintiff in the present case has not yet attempted to shoulder that burden. *610Although plaintiff obtained a stipulation that as of the date of trial the ordinance’s goals had not been fulfilled, it presented no evidence to show the likely duration or effect of the ordinance’s restriction upon building permits. We must presume that the City of Livermore and appropriate regional agencies will attempt in good faith to provide that community with adequate schools, sewage disposal facilities, and a sufficient water supply; plaintiff, however, has not presented evidence to show whether the city and such agencies have undertaken to construct the needed improvements or when such improvements will be completed. Consequently we cannot determine the impact upon either Livermore or the surrounding region of the ordinance’s restriction on the issuance of building permits pending achievement of its goals.

With respect to the competing interests, plaintiff asserts the existence of an acute housing shortage in the San Francisco Bay Area, but presents no evidence to document that shortage or to relate it to the probable effect of the Livermore ordinance. Defendants maintain that Livermore has severe problems of air pollution and inadequate public facilities which make it reasonable to divert new housing, at least temporarily, to other communities but offer no evidence to support that claim. Without an evidentiaiy record to demonstrate the validity and significance of the asserted interests, we cannot determine whether the instant ordinance attempts a reasonable accommodation of those interests.

In short, we cannot determine on the pleadings and stipulations alone whether this ordinance reasonably relates to the general welfare of the region it affects. The ordinance carries the presumption of constitutionality; plaintiff cannot overcome that presumption on the limited record before us. Thus the judgment rendered on this limited record cannot be sustained on the ground that the initiative ordinance falls beyond the proper scope of the police power.

5. Conclusion.

For the reasons we have explained, the Livermore ordinance is neither invalid on the ground that it was enacted by initiative nor unconstitutional by reason of vagueness. The more difficult question whether the measure is one which reasonably relates to the welfare of the region affected by its exclusionary impact, and thus falls within the police power of the city, cannot be decided on the limited record here. That issue can *611only be resolved by a trial at which evidence is presented to document the probable impact of the ordinance upon the municipality and the surrounding region.

The judgment of the superior court is reversed, and the cause remanded for further proceedings consistent with the views expressed herein.

Wright, C. J., McComb, J., Sullivan, J., and Richardson, J., concurred.

CLARK, J.

I dissent.

The zoning provisions of our law applicable to general law cities and the initiative provisions are clearly in conflict as recognized in Hurst v. City of Burlingame (1929) 207 Cal. 134 [277 P. 308]. A long line of decisions by this court and the Courts of Appeal has followed Hurst. (E.g., Johnston v. City of Claremont (1958) 49 Cal.2d 826, 836-837 [323 P.2d 71]; Simpson v. Hite (1950) 36 Cal.2d 125, 134 [222 P.2d 225]; Taschner v. City Council (1973) 31 Cal.App.3d 48, 61 et seq. [107 Cal.Rptr. 214]; Laguna Beach Taxpayers’ Assn. v. City Council (1960) 187 Cal.App.2d 412, 415 [9 Cal.Rptr. 775]; see San Diego Bldg. Contractors Assn. v. City Council (1974) 13 Cal.3d 205, 215 [118 Cal.Rptr. 146, 529 P.2d 570, 72 A.L.R.3d 973].) Until today, it was held that because of the conflict general law cities’ zoning ordinances were not subject to enactment by initiative. The rationale was: the statute conferring upon the legislative body the power to enact zoning prescribes the enactment method thereby establishing the measure of the power to enact; where a state act specifies the steps to be followed by the local body in enacting legislation, the initiative could not be used unless the steps were taken, and the steps required for zoning ordinances could not be followed within the initiative process. (Id.) The reasoning is compelling and indeed conclusive; I would not overrule Hurst and the numerous cases following it.

When we look at constitutional and statutory provisions governing zoning, related matters, and initiative process, the conflict is apparent.

Zoning

As pointed out in Hurst, a general law city is limited in the exercise of its powers by the Constitution and the general laws. (207 Cal. at p. 138; *612see Bishop v. City of San Jose (1969) 1 Cal.3d 56, 61 [81 Cal.Rptr. 465, 460 P.2d 137].) The power of a general law city to zone is derived from article XI, section 11: “A county or city may make and enforce within its limits all local, police, sanitary, and other ordinances and regulations not in conflict with general laws.” (Italics added; Miller v. Board of Public Works (1925) 195 Cal. 477 [234 P. 381, 38 A.L.R. 1479], 483; People v. Johnson (1955) 129 Cal.App.2d 1, 5 [277 P.2d 45].)1

The Legislature has specifically authorized general law cities and counties to adopt zoning ordinances, enumerating many of the types of zoning regulations. (Gov. Code, §§ 65800, 65850.) Government Code section 65802 provides that the procedures for enactment of zoning laws are exclusive: “No provisions of this code, other than the provisions of this chapter, and no provisions of any other code or statute shall restrict of limit the procedures provided in this chapter by which the legislative body of any county or city enacts, amends, administers, or provides for the administration of any zoning law, ordinance, rule or regulation.”

The Legislature has expressly provided that a zoning ordinance changing property from one zone to another or imposing or removing any of the numerous regulations set forth in Government Code section 65850 shall be adopted in the manner specified in sections 65854 to 65857 inclusive. (Gov. Code, § 65853.)

The procedure established provides for notice and hearing by the planning commission, a written report and recommendation by the planning commission including specification of the relationship of the proposed ordinance to general and specific plans, public hearings by the city council or board of supervisors after notice, and a further report by the planning commission in the event of modification by the legislative body. (Gov. Code, §§ 65854-65857.) Interim ordinances may be adopted as urgency measures prohibiting uses in conflict with a contemplated zoning proposal but only by four-fifths vote and only for a short period' of time. (Gov. Code, § 65858.) Zoning ordinances are required to be consistent with the general plan. (Gov. Code, § 65860.) Extensive provisions regulate adoption and amendment of the general plan. (Gov. *613Code, §§ 65300-65552.) There is also provision for variances. (Gov. Code, § 65906.)

Although the zoning power is legislative, administrative duties in addition to the ones in the above code sections have been imported into the zoning process. Legislative bodies adopting zoning ordinances are not free to merely follow the interests of their constituents but must give consideration to the interests of residents of nearby communities. (Scott v. City of Indian Wells (1972) 6 Cal.3d 541, 546-549 [99 Cal.Rptr. 745, 492 P.2d 1137].) Recently, this court held that the California Environmental Quality Act (Pub. Resources Code, § 21050 et seq.) applied to zoning ordinances, that environmental impact reports must be prepared in cases of significant environmental impact, and that legislative bodies are required to. make a written finding of no significant impact before enacting zoning ordinances if the report is not prepared. (No Oil, Inc. v. City of Los Angeles (1974) 13 Cal.3d 68, 79 et seq. [118 Cal.Rptr. 34, 529 P.2d 66].)

Initiative

Article IV, section 25 of our Constitution provides: “Initiative and referendum powers may be exercised by the electors of each city or county under procedures that the Legislature shall provide.” Proponents of an initiative in a city must give notice thereof and then circulate petitions to voters. (Elec. Code, §§ 4000-4009.) If the requisite number of signatures are obtained, the ordinance is presented to the legislative body which may adopt it without change. (Elec. Code, §§ 4011, 4012.) If within 10 days it fails to adopt, the proposed ordinance must be submitted to the voters at a special or general election. (Id.) If the legislative body adopts the proposed ordinance without submission to the voters or if upon submission a majority of the voters approve, the proposed ordinance goes into effect, and the ordinance may not be repealed or amended except by vote of the People unless provision is otherwise made in the original ordinance. (Elec. Code, § 4015.)

Conflict

The zoning law and the initiative law conflict in a number of respects. Fundamentally, the zoning statutes contemplate that to achieve orderly and wise land use regulation any change in zoning ordinances is not to *614be made until the experts in the field have had an opportunity to evaluate the effects of the change after noticed hearing and report. Further, the zoning law contemplates that in evaluating zoning changes, the legislative body must refer modifications not covered by the initial report to the planning commission. Such reports as to the instant ordinance would show, for example, which lots are zoned solely for residential use and might indicate the potential liability, if any, of the city in inverse condemnation.2 The reports would probably indicate the anticipated effect of the ordinance on surrounding communities. Preparation of reports might also lead to clarification; for example, it is unclear whether the ordinance is limited to permits for new residences or extends to permits for additions to and modifications of existing residences. The environmental impact report might show potential increases in automobile congestion and air pollution which might result because adoption of the ordinance may require many people to commute to work in Livermore.

Because of the short time limitation in the initiative, the proposed initiative ordinance must be adopted without the notice, hearings, and reports the Legislature has required for zoning changes. The initiative law conflicts with the zoning law by permitting the voters or the city council to adopt the ordinance without compliance with the specified procedures designed to insure orderly land use planning.

There are additional conflicts and potential conflicts. There is no assurance that interests of nearby residents will be considered by the electorate, although such consideration is required. There is no procedure under the initiative law for determining compliance with the general plan as required by statute. Because the city council must either reject or accept the proposed ordinance without change, it does not have the opportunity to impose conditions and modifications in the initiative process as provided in the zoning statutes. There are potential conflicts between the initiative law’s requirement that amendment be by the voters and the zoning law’s provision for variances, and between the majority vote of the initiative and the zoning law’s specific requirements for interim zoning.

*615The conflict between the two statutes is clear. The zoning laws establish an administrative process which must be followed prior to the legislative act of adopting an ordinance. The initiative statutes leave no room to carry out the administrative function. Both the statutes governing zoning of general law cities and governing initiative in such cities find their authority in our Constitution. Thus, there is no basis for the majority’s thesis suggesting that the Constitution requires that initiative law take precedence over the zoning law insofar as there may be conflict. Rather, the familiar rule that the specific governs the general in cases of conflict is applicable, and as held in Hurst, the zoning statutes must be given effect. The reasoning of Hurst is as applicable today as it was when the case was decided in 1929, if not more so in view of new administrative procedures governing land use planning, and I would reaffirm Hurst.

It is ironic that today’s decision, reviewing a “no growth” ordinance, may provide a loophole for developers to avoid the numerous procedures established by the Legislature which in recent years have made real estate development so difficult. Seeking approval of planned unit developments, land developers with the aid of the building trade unions should have little difficulty in securing the requisite signatures for an initiative ordinance. Because of today’s holding that the initiative takes precedence over zoning laws, the legislative scheme of notice, hearings, agency consideration, reports, findings, and modifications can be bypassed, and the city council may immediately adopt the planned unit development or, if the council refuses, the voters may approve.3 However desirable the creation of the loophole and the elimination of so-called administrative red tape it is not for this court, but for the Legislature to determine whether the current housing crisis warrants bypassing the zoning laws.4

*616I would affirm the judgment.

MOSK, J.

I dissent.

Limitations on growth may be justified in resort communities, beach and lake and mountain sites, and other rural and recreational areas; such restrictions are generally designed to preserve nature’s environment for the benefit of all mankind. They fulfill our fiduciary obligation to posterity. As Thomas Jefferson wrote, the earth belongs to the living, but in usufruct.1

But there is a vast qualitative difference when a suburban community invokes an elitist concept to construct a mythical moat around its perimeter, not for the benefit of mankind but to exclude all but its fortunate current residents.

The procedural posture of the ordinance does not detain me; the majority is correct in overruling Hurst v. Burlingame (1929) 207 Cal. 134 [277 P. 308]. The Hurst doctrine has long outlived its usefulness; it should no longer hobble the initiative process. Where I part company with the majority is in its substantive holding that a total exclusion of new residents can be constitutionally accomplished under a city’s police power.

The majority, somewhat desultorily, deny that the ordinance imposes an absolute prohibition upon population growth or residential construction. It is true that the measure prohibits the issuance of building permits for single-family residential, multiple residential and trailer residential units until designated public services meet specified standards. But to see such restriction in practicality as something short of total prohibition is to employ ostrich vision.

First of all, the ordinance provides no timetable or dates by which the public services are to be made adequate. Thus the moratorium on permits is likely to continue for decades, or at least until attrition ultimately reduces the present population. Second, it is obvious that no inducement exists for present residents to expend their resources to render facilities adequate for the purpose of accommodating future *617residents. It would seem more rational, if improved services are really contemplated for any time in the foreseeable future, to admit the new residents and compel them to make their proportionate contribution to the cost of the educational,, sewage and water services. Thus it cannot seriously be argued that Livermore maintains anything other than total exclusion.

The trial court found, inter alia, that the ordinance prohibited the issuance of building permits for residential purposes until certain conditions are met, but the measure does not provide that any person or agency is required to expend or commence any efforts on behalf of the city to meet the requirements. Nor is the city itself obliged to act within any specified time to cure its own deficiencies. Thus, in these circumstances procrastination produces its own reward: continued exclusion of new residents.

The significant omissions, when noted in relation to the ordinance preamble, reveal that the underlying purpose of the measure is “to control residential building permits in the City of Livermore”—translation: to keep newcomers out of the city—and not to solve the purported inadequacies in municipal educational, sewage and water services. Livermore concedes no building permits are now being issued and it relates no current or prospective schedule designed to correct its defective municipal services.

A municipal policy of preventing acquisition and development of property by nonresidents clearly violates article I, sections 1 and 7, subdivisions (a) and (b), of the Constitution of California.

Exclusion of unwanted outsiders, while a more frequent phenomenon recently, is not entirely innovative. The State of California made an abortive effort toward exclusivity back in the 1930s as part of a scheme to stem the influx of poor migrants from the dust bowl states of the southwest. The additional burden these indigent new residents placed on California services and facilities was severely aggravated by the great depression of that period. In Edwards v. California (1941) 314 U.S. 160 [86 L.Ed. 119, 62 S.Ct. 164], the Supreme Court held, however, that the nature of the union established by the Constitution did not permit any one state to “isolate itself from the difficulties common to all of them by restraining the transportation of persons and property across its borders.” The sanction against immigration of indigents was invalidated.

*618If California could not protect itself from the growth problems of that era, may Livermore build a Chinese Wall,to insulate itself from growth problems today? And if Livermore may do so, why not every municipality in Alameda County and in all other counties in Northern California? With a patchwork of enclaves the inevitable result will be creation of an aristocracy housed in exclusive suburbs while modest wage earners will be confined to declining neighborhoods, crowded into sterile, monotonous, multifamily projects, or assigned to pockets of marginal housing on the urban fringe. The overriding objective should be to minimize rather than exacerbate social and economic disparities, to lower barriers rather than raise them, to emphasize heterogeneity rather than homogeneity, to increase choice rather than limit it.

I am aware, of course, of the decision in Village of Belle Terre v. Boraas (1974) 416 U.S. 1 [39 L.Ed.2d 797, 94 S.Ct. 1536], in which the Supreme Court, speaking through Justice Douglas, rejected challenges to an ordinance restricting land use to one-family dwellings, with a veiy narrow definition of “family,” excluding lodging houses, boarding houses, fraternity houses, or multiple-dwelling houses. The village sought to assure that it would never grow much larger than 700 persons living in 220 residences. Comparable, although some growth was permitted, was the ordinance approved in Construction Ind. Ass’n, Sonoma Cty. v. City of Petaluma (9th Cir. 1975) 522 F.2d 897. Also similar, although allowing phased growth, was Golden v. Planning Board of Town of Ramapo (1972) 30 N.Y.2d 359 [334 N.Y.S.2d 138, 285 N.E.2d 291].2

In Belle Terre, Justice Douglas declared, “The police power is not confined to elimination of filth, stench, and unhealthy places. ... It is ample to lay out zones where family values, youth values, and the blessings of quiet seclusion and clean air make the area a sanctuary for people .... A quiet place where yards are wide, people few, and motor vehicles restricted are legitimate guidelines in a land-use project addressed to family needs,”

This is a comforting environmentalist declaration with which few would disagree, although the result was to allow the village of Belle *619Terre to remain an affluent island. Nevertheless, “preservation of the character of the community” is a stirring slogan, at least where it is used for nothing more harmful than the exclusion of the six students who rented the large house in Belle Terre. Complications arise when ordinances are employed to exclude not merely student lodgers, but all outsiders. While the affluent may seek a congenial suburban atmosphere other than Belle Terre or Livermore, what are the alternatives for those in megalopolitan areas who cannot afford similar selectivity?

The right of all persons to acquire housing is not a mere esoteric principle; it has commanded recognition in a wide spectrum of aspects. In Shelley v. Kraemer (1948) 334 U.S. 1 [92 L.Ed. 1161, 68 S.Ct. 836, 3 A.L.R.2d 441], race restrictive covenants were declared to be constitutionally unenforceable. Chief Justice Vinson noted in his opinion that among the guarantees of the Fourteenth Amendment “are the rights to acquire, enjoy, own and dispose of property.” In Reitman v. Mulkey (1967) 387 U.S. 369 [18 L.Ed.2d 830, 87 S.Ct. 1627], the Supreme Court upheld our invalidation of a ballot proposition, declaring that “ ‘Neither the State nor any subdivision or agency thereof shall deny, limit of abridge, directly or indirectly, the right of any person, who is willing or desires to sell, lease or rent any part or all of his real property, to decline to sell, lease or rent such properly to such person or persons as he, in his absolute discretion, chooses.’ ” Justice Douglas, in a concurring opinion in Reitman, went even further to insist that “housing is clearly marked with the public interest.” (Id. at p. 385 [18 L.Ed.2d at p. 840].) Again in Jones v. Mayer Co. (1968) 392 U.S. 409, 418 [20 L.Ed.2d 1189, 1196, 88 S.Ct. 2186], a case involving racial discrimination in housing, Justice Stewart spoke of the right of all citizens “ ‘to inherit, purchase, lease, sell, hold, and convey real and personal property.’ ” (Also see Buchanan v. Warley (1917) 245 U.S. 60 [62 L.Ed. 149, 38 S.Ct. 16].)

One thing emerges with clarity from the foregoing and from numerous related cases: access to housing is regarded by the Supreme Court as a matter of serious social and constitutional concern. While this interest has generally been manifest in the context of racial discrimination, there is no valid reason for not invoking the principle when persons of all races and of all economic groups are involved. There are no invariable racial or economic characteristics of the goodly numbers of families which seek social mobility, the opportunities for the good life available in a suburban atmosphere, and access to types of housing, education and employment differing from those indigenous to crowded urban centers.

*620There is a plethora of commentary on efforts, in a variety of contexts, of local communities to discourage the influx of outsiders. In virtually every instance, however, the cities limited availability of housing; until now it has never been seriously contemplated that a commmunity would attempt total exclusion by refusing all building permits. (See, e.g., Williams & Doughty, Studies in Legal Realism: Mount Laurel, Belle Terre and Berman (1975) 29 Rutgers L.Rev. 73; Note, Phased Zoning: Regulation of the Tempo and Sequence of Land Development (1974) 26 Stan.L.Rev. 585; Note, The Right to Travel and Exclusionary Zoning (1974) 26 Hastings L.J. 849; Deutsch, Land Use Growth Controls: A Case Study of San Jose and Livermore, California (1974) 15 Santa Clara Law. 1; Schroeder, Public Regulation of Private Land Use, 1973 Law & Soc. Order 747; Large, This Land is Whose Land? Changing Concepts of Land as Property (1973) Wis.L.Rev. 1039; Gaffrey, Containment Policies for Urban Sprawl, Univ. of Kan. Publications, No. 27; McClaughry, The New Feudalism (1975) 5 EnvironmentalL.Rev. 675; Kohl, The Environmental Movement: What Might It Be? (1975) 15 Nat.Res.J. 327; Note, The Right to Travel: Another Constitutional Standard for Local Land Use Regulations? (1972) 39 U.Chi.L.Rev. 612; Note, The Responsibility of Local Zoning Authorities to Nonresident Indigents (1971) 23 Stan.L.Rev. 774; Note, Exclusionary Zoning and Equal Protection (1971) 84 Harv.L.Rev. 1645; Sager, Tight Little Islands: Exclusionary Zoning, Equal Protection, and the Indigent (1969) 21 Stan.L.Rev. 767.)

The trend in the more perceptive jurisdictions is to prevent municipalities from selfishly donning blinders to obscure the problems of their neighbors. The Supreme Court of New Jersey has taken the lead in frowning upon creation of local exclusive enclaves and in insisting upon consideration of regional housing needs. In Oakwood at Madison, Inc. v. Township of Madison (1971) 117 N.J.Sup. 11 [283 A.2d 353, 358], the court held, “In pursuing the valid zoning purpose of a balanced community, a municipality must not ignore housing needs, that is, its fair proportion of the obligation to meet the housing needs of its own population and of the region. Housing needs are encompassed within the general welfare. The general welfare does not stop at each municipal boundary.” (Italics added.)

Again in the oft-cited Mt. Laurel case (So. Burlington Cty. N.A.A.C.P. v. Tp. of Mt. Laurel (1975) 67 N.J. 151 [336 A.2d 713, 724]) the New Jersey Supreme Court required that municipalities afford the opportunity for housing, “at least to the extent of the municipality’s fair share of *621the present and prospective regional need therefor.” (Italics added.) (Also see Schere v. Township of Freehold (1972) 119 N.J.Sup. 433 [292 A.2d 35, 37].)

Pennsylvania is another state that has forthrightly spoken out against ordinances “designed to be exclusive and exclusionary.” In National Land and Investment Company v. Kohn (1966) 419 Pa. 504 [215 A.2d 597, 612], a case remarkably similar to the instant matter, the Easttown community refused to admit new residents “unless such admittance will not create any additional burdens upon governmental functions and services.” Justice Roberts, for the Supreme Court, replied: “The question posed is whether the township can stand in the way of the natural forces which send our growing population into hitherto undeveloped areas in search of a comfortable place to live. We have concluded not. A zoning ordinance whose primary purpose is to prevent the entrance of newcomers in order to avoid future burdens, economic and otherwise, upon the administration of public services and facilities cannot be held valid.”

In Appeal of Girsh (1970) 437 Pa. 237 [263 A.2d 395], the Pennsylvania Supreme Court again spoke from a broad perspective. The community involved there barred all apartment houses for the identical reasons advanced by Livermore here. Said the court with irrefutable logic: “Appellee argues that apartment uses would cause a significant population increase with a resulting strain on available municipal services and roads, and would clash with the existing residential neighborhood. But we explicitly rejected both these claims in National Land, supra: ‘Zoning is a tool in the hands of governmental bodies which enables them to more , elfectively meet the demands of evolving and growing communities. It must not and can not be used by those officials as an instrument by which they may shirk their responsibilities. Zoning is a means by which a governmental body can plan for the future—it may not be used as a means to deny the future. ... Zoning provisions may not be used ... to avoid the increased responsibilities and economic burdens which time and natural growth invariably bring.’ 419 Pa. at 527-528, 215 A.2d at 610....

“. . . Appellee here has simply made a decision that it is content with things as they are, and that the expense or change in character that would result from people moving in to find ‘a comfortable place to live’ *622are for someone else to worry about. That decision is unacceptable. Statistics indicate that people are attempting to move away from the urban core areas, relieving the grossly overcrowded conditions that exist in most of our major cities. ... It follows then that formerly ‘outlying,’ somewhat rural communities, are becoming logical areas for development and population growth—in a sense suburbs to the suburbs. With improvements in regional transportation systems, these areas also are now more accessible to the central city.

“In light of this, Nether Providence Township may not permissibly choose to only take as many people as can live in single-family housing, in effect freezing the population at near present levels. Obviously if every municipality took that view, population spread would be completely frustrated. Municipal services must be provided somewhere, and if Nether Providence is a logical place for development- to take place, it should not be heard to say that it will not bear its rightful part of the burden.” (Id. at pp. 398-399; fn. omitted.)

In Girsh the Pennsylvania court added: “Perhaps in an ideal world, planning and zoning would be done on a regional basis, so that a given community would have apartments, while an adjoining community would not. But as long as we allow zoning to be done community by community, it is intolerable to allow one municipality (or many municipalities) to close its doors at the expense of surrounding communities and the central city.” (Id. at p. 399, fn. 4.)

Ordinances comparable to those invalidated in New Jersey and Pennsylvania have also been held invalid in Michigan (Bristow v. City of Woodhaven (1971) 35 Mich.App. 205 [192 N.W.2d 322]), Maryland (Baltimore Planning Com’n v. Victor Development Co. (1971) 261 Md. 358 [275 A.2d 478]) and Connecticut (Beach v. Planning & Zoning Commission (1954) 141 Conn. 79 [103 A.2d 814]).

In sum, I realize the easiest course is for this court to defer to the political judgment of the townspeople of Livermore, on a they-know-what’s-best-for-them theory (Eastlake v. Forest City Enterprises, Inc. (1976) 426 U.S. 668 [49 L.Ed.2d 132, 96 S.Ct. 2358]; James v. Valtierra (1971) 402 U.S. 137 [28 L.Ed.2d 678, 91 S.Ct. 1331]). But conceptually, when a locality adopts a comprehensive, articulated program to prevent any populátion growth over the foreseeable future, it places its public *623policy intentions visibly on the table for judicial scrutiny and constitutional analysis.

Communities adopt growth limits from a variety of motives. There may be conservationists genuinely motivated to preserve general or specific envirónments. There may be others whose motivation is social exclusionism, racial exclusion, racial discrimination, income segregation, fiscal protection, or just fear of any future change; each of these purposes is well served by growth prevention.

Whatever the motivation, total exclusion of people from a community is both immoral and illegal. (Cal. Const., art. I, §§ 1, 7, subds. (a) & (b).) Courts have a duty to prevent such practices, while at the same time recognizing the validity of genuine conservationist efforts.

The problem is not insoluble, nor does it necessarily provoke extreme results. Indeed, the solution can be relatively simple if municipal agencies would consider the aspirations of society as a whole, rather than merely the effect upon their narrow constituency. (See, e.g., A.L.I. Model Land Development Code, art. 7.) Accommodation between environmental preservation and satisfaction of housing needs can be reached through rational guidelines for land-use decision-making. Ours, of course, is not the legislative function. But two legal inhibitions must be the benchmark of any such guidelines. First, any absolute prohibition on housing development is presumptively invalid. And second, local regulations, based on parochialism, that limit population densities in growing suburban areas may be found invalid unless the community is absorbing a reasonable share of the region’s population pressures.

Under the foregoing test,, the Livermpre ordinance is fatally flawed. I would affirm the judgment of the trial court.

The opinion of the Court was delivered by

Hall, J.

This case attacks the system of land use regulation by defendant Township of Mount Laurel on the ground that low and moderate income families are thereby unlawfully excluded from the municipality. The trial court so found, 119 N. J. Super. 164 (Law Div. 1972), and declared the township zoning ordinance totally invalid. Its judgment went on, in line with the requests for affirmative relief, to order the municipality to make studies of the housing needs of low and moderate income persons presently or formerly residing in the community in substandard housing, as well as those in such income classifications presently employed in the township and living elsewhere or reasonably expected to be employed therein in the future, and to present a plan of affirmative public action designed “to en*158able and encourage the satisfaction of the indicated needs.” Jurisdiction was retained for judicial consideration and approval of such a plan and for the entry of a final' order requiring its implementation.

The township appealed to the Appellate Division and those plaintiffs, not present or former residents, cross-appealed on the basis that the judgment should have directed that the prescribed plan take into account as well a fair share of the regional housing needs of low and moderate income families without limitation to those having past, present or prospective connection with the township. The appeals were certified on our own motion before argument in the Division. R. 2:12-1.1

The implications of the issue presented are indeed broad and far-reaching, extending much beyond these particular plaintiffs and the boundaries of this particular municipality.

There is not the slightest doubt that New Jersey has been, and continues to be, faced with a desperate need for housing, especially of decent living accommodations economically suitable for low and moderate income families.2 The situ*159ation was characterized as a “crisis” and fully explored and documented by Governor Cahill in two special messages to the Legislature — A Blueprint for Housing in New Jersey (1970) and New Horizons in Housing (1972).

Plaintiffs represent the minority group poor (black and Hispanic)3 seeking such quarters. But they are not the only category of persons barred from so many municipalities by reason of restrictive land use regulations. We have reference to young and elderly couples, single persons and large, growing families not in the poverty class, but who still cannot afford the only kinds of housing realistically permitted in most places — relatively high-priced, single-family detached dwellings on sizeable lots and, in some municipalities, expensive apartments. We will, therefore, consider the case from the wider viewpoint that the effect of Mount Laurel’s land use regulation has been to prevent various categories of persons from living in the township because of the limited extent of their income and resources. In this connection, we accept the representation of the municipality’s counsel at oral argument that the regulatory scheme was not adopted with any desire or intent to exclude prospective residents on the obviously illegal basis of race, origin or believed social incompatibility.

*160As already intimated, the issue here is not confined to Mount Laurel. The same question arises with respect to any number of other municipalities of sizeable land area outside the central cities and older built-up suburbs of our Earth and South Jersey metropolitan areas (and surrounding some of the smaller cities outside those areas as well) which, like Mount Laurel, have substantially shed rural characteristics and have undergone great population increase since World War II, or are now in the process of doing so, but still are not completely developed and remain in the path of inevitable future residential, commercial and industrial demand and growth. Most such municipalities, with but relatively insignificant variation in details, present generally comparable physical situations, courses of municipal policies, practices, enactments and results and human, governmental and legal problems arising therefrom. It is in the context of communities now of this type or which become so in the future, rather than with central cities or older built-up suburbs or areas still rural and likely to continue to be for some time yet, that we deal with the question raised.

Extensive oral and documentary evidence was introduced at the trial, largely informational, dealing with the development of Mount Laurel, including the nature and effect of municipal regulation, the details of the region of which it is a part and the recent history thereof, and some of the basics of housing, special reference being directed to that for low and moderate income families. The record has been supplemented by figures, maps, studies and literature furnished or referred to by counsel and the amici, so that the court has a clear picture of land use regulation and its effects in the developing municipalities of the state.

This evidence was not contradicted by the township, except in a few unimportant details. Its candid position is that, conceding its land use regulation was intended to result and has resulted in economic discrimination and ex-*161elusion of substantial segments of the area population, its policies and practices are in the best present and future fiscal interest of the municipality and its inhabitants and are legally permissible and justified. It further asserts that the trial court was without power to direct the affirmative relief it did.

I

The Facts

Mount Laurel is a flat, sprawling township, 22 square miles, or about 14,000 acres, in area, on the west central edge of Burlngton County. It is roughly triangular in shape, with its base, approximately eight miles long, extending in a northeasterly-southwesterly direction roughly parallel with and a few miles east of the Delaware River. Part of its southerly side abuts Cherry Hill in Camden County. That section of the township is about seven miles from the boundary line of the city of Camden and not more than 10 miles from the Benjamin Eranklin Bridge crossing the river to Philadelphia.

In 1950, the township had a population of 2817, only about 600 more people than it had in 1940. It was then, as it had been for decades, primarily a rural agricultural area with no sizeable settlements or commercial or industrial enterprises. The populace generally lived in individual houses scattered along country roads. There were several pockets of poverty, with deteriorating or. dilapidated housing (apparently 300 or so units of which remain today in equally poor condition). After 1950, as in so many other municipalities similarly situated, residential development and some commerce and industry began to come in. By 1960 the population had almost doubled to 5249 and by 1970 had more than doubled again to 11,221. These new residents were, of course, “outsiders” from the nearby central ■cities and older suburbs or from more distant places drawn *162here by reason of employment in the region. The township is now definitely a part of the outer ring of the South Jersey metropolitan area, which area we define as those portions of Camden, Burlington and Gloucester Counties within a semicircle having a radius of 20 miles or so from the heart of Camden city. And 65% of the township is still vacant land or in agricultural use.

The growth of the township has been spurred by the construction or improvement of main highways through or near it. The New Jersey Turnpike, and now route 1-295, a freeway paralleling the turnpike, traverse the municipality near its base, with the main Camden-Philadelphia turnpike interchange at the corner nearest Camden. State route 73 runs at right angles to the turnpike at the interchange and route 38 slices through the northeasterly section. Routes 70 and U. S. 130 are not far away. This highway network gives the township a most strategic location from the standpoint of transport of goods and people by truck and private ear. There is no other means of transportation.

The location and nature of development has been, as usual, controlled by the local zoning enactments. The general ordinance presently in force, which was declared invalid by the trial court, was adopted in 1964. We understand that earlier enactments provided, however, basically the same scheme but were less restrictive as to residential development. The growth pattern dictated by the ordinance is typical.

Under the present ordinance, 29.2% of all the land in the township, or 4,121 acres, is zoned for industry. This amounts to 2,800 more acres than were so zoned by the 1954 ordinance. The industrial districts comprise most of the land on both sides of the turnpike and routes 1-295, 73 and 38. Only industry meeting specified performance standards is permitted. The effect is to limit the use substantially to light manufacturing, research, distribution of goods, offices and the like. Some non-industrial uses, such as agriculture, *163farm dwellings, motels, a harness racetrack, and certain retail sales and service establishments, are permitted in this zone. At the time of trial no more than 100 acres, mostly in the southwesterly corner along route 73 adjacent to the turnpike and 1-295 interchanges, were actually occupied by industrial uses. They had been constructed in recent years, mostly in several industrial parks, and involved tax ratables of about 16 million dollars. The rest of the land so zoned has remained undeveloped. If it were fully utilized, the testimony was that about 43,500 industrial jobs would be created, but it appeared clear that, as happens in the case of so many municipalities, much more land has been so zoned than the reasonable potential for industrial movement or expansion warrants. At the same time, however, the land cannot be used for residential development under the general ordinance.

The amount of land zoned for retail business use under the general ordinance is relatively small — 169 acres, or 1.2% of the total. Some of it is near the turnpike interchange; most of the rest is allocated to a handful of neighborhood commercial districts. While the greater part of the land so zoned appears to be in use, there is no major shopping center or concentrated retail commercial area — “downtown” •— in the township.

The balance of the land area, almost 10,000 acres, has been developed until recently in the conventional form of major subdivisions. The general ordinance provides for four residential zones, designated R — 1, R-1D, R-2’ and R-3. All permit only single-family, detached dwellings, one house per lot — the usual form of grid development. Attached townhouses, apartments (except on farms for agricultural workers) and mobile homes are not allowed anywhere in the township under the general ordinance. This dwelling development, resulting in the previously mentioned quadrupling of the population, has been largely confined to the R-l and R-2 districts in two sections — the northeasterly and southwesterly corners adjacent to the turnpike and other major highways. The result has been quite intensive development of these sections, *164but at a low density. The dwellings are substantial; the average value in 1971 was $32,500 and is undoubtedly much higher today.

The general ordinance requirements, while not as restrictive as those in many similar municipalities, nonetheless realistically allow only homes within the financial reach of persons of at least middle income. The R-l zone requires a minimum lot area of 9,375 square feet, minimum lot width of 75 feet at the building line, and a minimum dwelling floor area of 1,100 square feet if a one-story building and 1,300 square feet if one and one-half stories or higher. Originally this zone comprised about 2,500 acres. Most of the subdivisions have been constructed within it so that only a few hundred acres remain (the testimony was at variance as to the exact amount). The R-2 zone, comprising a single district of 141 acres in the northeasterly corner, has been completely developed. While it only required a minimum floor area of 900 square feet for a one-story dwelling, the minimum lot size was 11,000 square feet; otherwise the requisites were the same as in the R-l zone.

The general ordinance places the remainder of the township, outside of the industrial and commercial zones and the R-1D district (to be mentioned shortly), in the R-3 zone. This zone comprises over 7,000 acres — slightly more than half of the total municipal area — practically all of which is located in the central part of the township extending southeasterly to the apex of the triangle. The testimony was that about 4,600 acres of it then remained available for housing development. Ordinance requirements are substantially higher, however, in that the minimum lot size is increased to about one-half acre (20,000 square feet). (We understand that sewer and water utilities have not generally been installed, but, of course, they can be.) Lot width at the building line must be 100 feet. Minimum dwelling floor area is as in the R-l zone. Presently this section is primarily in agricultural use; it contains as well most of the municipality’s substandard housing.

*165The R-1D district was created by ordinance amendment in 1968. The area is composed of a piece of what was formerly R-3 land in the western part of that zone. The district is a so-called “cluster” zone. See generally 2 Williams, American Planning Law. Land Use and the Police Power, §§ 47.01-47.05 (1974). That writer defines the concept as follows:

* * * Under the usual cluster-zoning provisions, both the size and the width of individual residential lots in a large (or medium-sized) development may be reduced, provided (usually) that the overall density of the entire tract remains constant — provided, that is, that an area equivalent to the total of the areas thus “saved” from each individual lot is pooled and retained as common open space. The most obvious advantages include a better use of many sites, and relief from the monotony of continuous development. § 47.01, pp. 212-213.

Here this concept is implemented by reduction of the minimum lot area from 20,000 square feet required in the R-3 zone to 10,000 square feet (12,000 square feet for corner lots) but with the proviso that one-family houses — the single permitted dwelling use — “shall not be erected in excess of an allowable development density of 2.25 dwelling units per gross acre.” The minimum lot width at the building line must be 80 feet and the minimum dwelling floor area is the same as in the R-3 zone. The amendment further provides that the developer must set aside and dedicate to the municipality a minimum of 15% and a maximum of 25% of the total acreage for such public uses as may be required by the Planning Board, including “but not limited to school sites, parks, playgrounds, recreation areas, public buildings, public utilities.” Some dwelling development has taken place in this district, the exact extent of which is not disclosed by the record. It is apparent that the dwellings are comparable in character and value to those in the other residential zones. The testimony was that 486 acres remained available in the district.4

*166A variation from conventional development has recently-occurred in some parts of Mount Laurel, as in a number of other similar municipalities, by use of the land use regulation device known as “planned unit development” (PUD). This scheme differs from the traditional in that the type, density and placement of land uses and buildings, instead of being detailed and confined to specified districts by local legislation in advance, is determined by contract, or “deal,” as to each development between the developer and the municipal administrative authority, under broad guidelines laid down by state enabling legislation and an implementing local ordinance. The stress is on regulation of density and permitted mixture of uses within the same area, including various kinds of living accommodations with or without commercial and industrial enterprises. The idea may be basically thought of as the creation of “new towns” in virgin territory, full-blown or in miniature, although most frequently the concept has been limited in practice, as in Mount Laurel, to residential developments of various sizes having some variety of housing and perhaps some retail establishments to serve the inhabitants. See generally, 2 Williams, supra, §§ 48.01 to 48.12; cf. Cheney v. Village 2 at New Hope, Inc., 429 Pa. 626, 241 A. 2d 81, 82-83 (1968).

New Jersey passed such enabling legislation in 1967 (L. 1967, c. 61, amended c. 286, N. J. S. A. 40:55-54, et seq.), which closely follows a model act found in 114 U. Pa. L. Rev. 140 (1965), and Mount Laurel adopted the implementing enactment as a supplement to its general zoning ordinance in December of that year. While the ordinance was repealed early in 1971, the township governing body in the interim had approved four PUD projects, which were specifically saved from extinction by the repealer.5

*167These projects, three in the southwesterly sector and one in the northeasterly sector, are very substantial and involve at least 10,000 sale and rental housing units of various types to be erected over a period of years. Their bounds were created by agreement rather than legislative specification on the zoning map, invading industrial, R-l, R-1D, R-3 and even flood plain zones. If completed as planned, they will in themselves ultimately quadruple the 1970 township population, but still leave a good part of the township undeveloped. (The record does not indicate how far development in each of the projects has progressed.) While multi-family housing in the form of rental garden, medium rise and high rise apartments and attached townhouses is for tihe first time provided for, as well as single-family detached dwellings for sale, it is not designed to accommodate and is beyond the financial reach of low and moderate income families, especially those with young children. The aim is quite the contrary; as with the single-family homes in the older conventional subdivisions, only persons of medium and upper income are sought as residents.

A few details will furnish sufficient documentation. Each of the resolutions of tentative approval of the projects con*168tains a similar fact finding to the effect that the development will attract a highly educated and trained population base to support the nearby industrial parks in the township as well as the business and commercial facilities. The approvals also sharply limit the number of apartments having more than one bedroom. Further, they require that the developer must provide in its leases that no school-age children shall be permitted to occupy any one-bedroom apartment and that no more than two such children shall reside in any two-bedroom unit. The developer is also required, prior to the issuance of the first building permit, to record a covenant, running with all land on which multi-family housing is to be constructed, providing that in the event more than .3 school children per multi-family unit shall attend the township school sjstem in any one year, the developer will pay the cost of tuition and other school expenses of all such excess numbers of children. In addition, low density, required amenities, such as central air conditioning, and specified developer contributions help to push rents and sales prices to high levels. These contributions include fire apparatus, ambulances, fire houses, and very large sums of money for educational facilities, a cultural center and the township library.6

Still another restrictive land use regulation was adopted by the township through a supplement to the general zoning ordinance enacted in September 1972 creating a new zone, R-4, Planned Adult Retirement Community (PARC). The supplementary enactment designated a sizeable area as the zone — perhaps 200 acres — carved out of the R-1D and R-3 districts in the southwesterly sector. The enactment recited a critical shortage of adequate housing in the township suitable "for the needs and desires of senior citizens and certain other adults over the age of 52.” The permission was essentially for single ownership development of the zone for mnlti-family *169housing (townhouses and apartments), thereafter to be either rented or sold as cooperatives or condominiums. The extensive development requirements detailed in the ordinance make it apparent that the scheme was not designed for, and would be beyond the means of, low and moderate income retirees. The highly restricted nature of the zone is found in the requirement that all permanent residents must be at least 52 years of age (except a spouse, immediate family member other than a child, live-in domestic, companion or nurse). Children are limited to a maximum of one, over age 18, residing with a parent and there may be no more than three permanent residents in any one dwelling unit.7

All this affirmative action for the benefit of certain segments of the population is in sharp contrast to the lack of action, and indeed hostility, with respect to affording any opportunity for decent housing for the township’s own poor living in substandard accommodations, found largely in the section known as Springville (R-3 zone). The 196,9 Master Plan Report recognized it and recommended positive action. The continuous official reaction has been rather a negative policy of waiting for dilapidated premises to be vacated and .then forbidding further occupancy. An earlier non-governmental effort to improve conditions had been effectively thwarted. In 1968 a private non-profit association sought to build subsidized, multi-family housing in the Springville section with funds to be granted by a higher level governmental agency. Advance municipal approval of the project was required. The Township Committee responded with a purportedly approving resolution, which found a need for “moderate” income housing in the area, but went on to specify that such housing must be constructed subject to all zoning, planning, building and other applicable ordinances and codes. This meant single-family detached dwellings on 20,000 square foot lots. (Fear was also *170expressed that such housing would attract low income families from outside the township.) Needless to say, such requirements killed realistic housing for this group of low and moderate income families.8

The record thoroughly substantiates the findings of the trial court that over the years Mount Laurel “has acted affirmatively to control development and to attract a selective type of growth” (119 N. J. Super. at 168) and that “through its zoning ordinances has exhibited economic discrimination in that the poor have been deprived of adequate housing and the opportunity to secure the construction of subsidized housing, and has used federal, state, county and local finances and resources9 solely for the betterment of middle and upper-income persons.” (119 N. J. Super. at 178).

There cannot be the slightest doubt that the reason for this course of conduct has been to keep down local taxes on property (Mount Laurel is not a high tax municipality) and that the policy was carried out without regard for non-fiscal considerations with respect to people, either within or without its boundaries. This conclusion is demonstrated not only by what was done and what happened, as we have related, but also by innumerable direct statements of municipal officials at public meetings over the years which are found *171in the exhibits. The trial court referred to a number of them. 119 N. J. Super. at 169-170. No official testified to the contrary.

This policy of land use regulation for a fiscal end derives from New Jersey’s tax structure, which has imposed on local real estate most of the cost of municipal and county government and of the primary and secondary education of the municipality’s children. The latter expense is much the largest, so, basically, the fewer the school children, the lower the tax rate. Sizeable industrial and commercial ratables are eagerly sought and homes and the lots on which they are situate are required to be large enough, through minimum lot sizes and minimum floor areas, to have substantial value in order to produce greater tax revenues to meet school costs. Large families who cannot afford to buy large houses and must live in cheaper rental accommodations are definitely not wanted, so we find drastic bedroom restrictions for, or complete prohibition of, multi-family or other feasible housing for those of lesser income.

This pattern of land use regulation has been adopted for the same purpose in developing municipality after developing municipality. Almost every one acts solely in its own selfish and parochial interest and in effect builds a wall around itself to keep out those people or entities not adding favorably to the tax base, despite the location of the municipality or the demand for varied kinds of housing. There has been no effective intermunieipal or area planning or land use regulation. All of this is amply demonstrated by the evidence in this ease as to Camden, Burlington and Gloucester counties. As to the similar situation generally in the state, see New Jersey Department of Community Affairs, Division of State and Regional Planning, Land Use Regulation, The Residential Land Supply (April 1972) (a study assembling and examining the nature and extent of municipal zoning practices in 16 counties as affecting residential land available for low and moderate income housing) *172and Williams and Norman, Exclusionary Land Use Controls: The Case of North-Eastern New Jersey, 22 Syracuse L. Rev. 475, 486-487 (1971). One incongruous result is the picture of developing municipalities rendering it impossible for lower paid employees of industries they have eagerly sought and welcomed with open arms (and, in Mount Laurel’s case, even some of its own lower paid municipal employees) to live in the community where they work.

The other end of the spectrum should also be mentioned because it shows the source of some of the demand for cheaper housing than the developing municipalities have permitted. Core cities were originally the location of most commerce and industry. Many of those facilities furnished employment for the unskilled and semi-skilled. These employees lived relatively near their work, so sections of cities always have housed the majority of people of low and moderate income, generally in old and deteriorating housing. Despite the municipally confined tax structure, commercial and industrial ratables generally used to supply enough revenue to provide and maintain municipal services equal or superior to those furnished in most suburban and rural areas.

The situation has become exactly the opposite since the end of World War II. Much industry and retail business, and even the professions, have left the cities. Camden is a typical example. The testimonial and documentary evidence in this case as to what has happened to that city is depressing indeed. Eor various reasons, it lost thousands of jobs between 1950 and 1970, including more than half of its manufacturing jobs (a reduction' from 43,267 to 20,671, while all jobs in the entire area labor market increased from 94,507 to 197,037). A large segment of retail business faded away with the erection of large suburban shopping centers. The economically better situated city residents helped fill up the miles.of sprawling new housing developments, not fully served by public transit. In a society which *173came' to depend more and more on expensive individual motor vehicle transportation for all purposes, low income employees very frequently could not afford to reach outlying places of suitable employment and they certainly could not afford the permissible housing near such locations. These people have great difficulty in obtaining work and have been forced to remain in housing which is overcrowded, and has become more and more substandard and less and less tax productive. There has been a consequent critical erosion of the city tax base and inability to provide the amount and quality of those governmental services — education, health, police, fire, housing and the like — so necessary to the very existence of safe and decent city life. This category of city dwellers desperately needs much better housing and living conditions than is available to them now, both in a rehabilitated city and in outlying municipalities. They make up, along with the other classes of persons earlier mentioned who also cannot afford the only generally permitted housing in the developing municipalities, the acknowledged great demand for low and moderate income housing.

II

The Legal Issue

The legal question before us, as earlier indicated, is whether a developing municipality like Mount Laurel may validly, by a system of land use regulation, make it physically and economically impossible to provide low and moderate income housing in the municipality for the various categories of persons who need and want it and thereby, as Mount Laurel has, exclude such people from living within its confines because of the limited extent of their income and resources. Necessarily implicated are the broader questions of the right of such municipalities to limit the kinds of available housing and of any obligation to make possible a variety and choice of types of living accommodations.

*174 We conclude that every such municipality must, by its land use regulations, presumptively make realistically possible an appropriate variety and choice of housing. More specifically, presumptively it cannot foreclose the opportunity of the classes of people mentioned for low and moderate income housing and in its regulations must affirmatively afford that opportunity, at least to the extent of the municipality’s fair share of the present and prospective regional need therefor. These obligations must be met unless the particular municipality can sustain the heavy burden of demonstrating peculiar circumstances which dictate that it should not be required so to do.10

We reach this conclusion under state law and so do not find it necessary to consider federal constitutional grounds urged by plaintiffs. We begin with some fundamental principles as applied to the scene before us.

Land use regulation is encompassed within the state’s police power. Our constitutions have expressly so provided since an amendment in 1927. That amendment, now Art. IY, sec. YI, par. 2 of the 1947 Constitution, authorized legislative delegation of the power to municipalities (other than counties), but reserved the legislative right to repeal or alter the delegation (which we take it means repeal or alteration in whole or in part). The legislative delegation of the zoning power followed in 1928, by adoption of the standard zoning enabling act, now found, with subsequent amendments, in N. J. S. A. 40:55-30 to 51.

It is elementary theory that all police power enactments, no matter at what level of government, must conform to the basic state constitutional requirements of substantive due process and equal protection of the laws. These *175are inherent in Art. I, par. 1 of onr Constitution,11 the requirements of which may be more demanding than those of the federal Constitution. Robinson v. Cahill, 62 N. J. 473, 482, 490-492 (1973); Washington National Insurance Co. v. Board of Review, 1 N. J. 545, 553-554 (1949). It is required that, affirmatively, a zoning regulation, like any police power enactment, must promote public health, safety, morals or the general welfare. (The last term seems broad enough to encompass the others). Conversely, a zoning enactment which is contrary to the general welfare is invalid. See generally, e. g., Roselle v. Wright, 21 N. J. 400, 409—410 (1956); Katobimar Realty Co. v. Webster, 20 N. J. 114, 122-123 (1955); Schmidt v. Board of Adjustment of Newwark, 9 N. J. 405, 413-419 (1952); Collins v. Board of Adjustment of Margate City, 3 N. J. 200, 206 (1949). Indeed these considerations are specifically set forth in the zoning enabling act as among the various purposes of zoning for which regulations must be designed. N. J. S. A. 40:55-32. Their inclusion therein really adds little; the same requirement would exist even if they were omitted. If a zoning regulation violates the enabling act in this respect, it is also theoretically invalid under the state constitution. We say “theoretically” because, as a matter of policy, we do not treat the validity of most land use ordinance provisions as involving matters of constitutional dimension; that classification is confined to major questions of fundamental import. Cf. Tidewater Oil Co. v. Mayor and Council of the Borough of Carteret, 44 N. J. 338, 343 (1965). We consider the basic importance of housing and local regulations restricting its availability to substantial segments of the population to fall within the latter category.

*176The demarcation between the valid and the invalid in the field of land nse regulation is difficult to determine, not always .clear and subject to change. This was recognized almost fifty years ago in the basic case of Village of Euclid v. Ambler Realty Co., 272 U. S. 365, 47 S. Ct. 114, 71 L. Ed. 303 (1926) :

The ordinance now under review and all similar laws and regulations must find their justification in some aspect of the police power, asserted for the public welfare. The line which in this field separates the legitimate from the illegitimate assumption of power is not capable of precise delimitation. It varies with circumstances and conditions. (272 U. S. at 387, 47 S. Ct. at 118, 71 L. Ed. at 310).

This court has also said as much and has plainly warned, even in cases decided some years ago sanctioning a broad measure of restrictive municipal decisions, of the inevitability of change in judicial approach and view as mandated by change in the world around us. Lionshead Lake, Inc. v. Township of Wayne, 10 N. J. 165, 172-173 (1952), appeal dismissed 344 U. S. 919, 73 S. Ct. 386, 97 L. Ed. 708 (1953) (approving requirement of minimum floor area for dwellings, the same in all residential districts); Fischer v. Township of Bedminster, 11 N. J. 194, 205 (1952) (sanctioning minimum lot area of five acres in a then rural municipality); Pierro v. Baxendale, 20 N. J. 17, 29 (1955) (holding valid an ordinance permitting boarding and rooming houses, but not hotels and motels, in residential districts); Vickers v. Township Committee of Gloucester Township, 37 N. J. 232, 250 (1962), cert. den. 371 U. S. 233, 83 S. Ct. 326, 9 L. Ed. 2d 495 (1963) (sustaining ordinance provisions prohibiting mobile home parks throughout the township). The warning is perhaps best put in Pierro:

We are aware of the extensive academic discussion following the decisions in the Lionshead and Bedminster cases, supra, and the suggestion that the very broad principles which they embody may intensify dangers of economic segregation which even the more traditional modes of zoning entail * * *. In the light of existing population and land conditions within our State these [municipal zoning] *177powers may fairly be exercised without in anywise endangering the needs or reasonable expectations of any segments of our people. If and when conditions change, alterations in zoning restrictions and pertinent legislative and judicial attitudes need not be long delayed. (20 N. J. at 29).

The warning implicates the matter of whose general welfare must be served or not violated in the field of land use regulation. Erequently the decisions in this state, including those just cited, have spoken only in terms of the interest of the enacting municipality, so that it has been thought, at least in some quarters, that such was the only welfare requiring consideration. It is, of course, true that many eases have dealt only with regulations having little, if any, outside impact where the local decision is ordinarily entitled to prevail. However, it is fundamental and not to be forgotten that the zoning power is a police power of the state and the local authority is acting only as a delegate of that power and is restricted in the same manner as is the state. So, when regulation does have a substantial external impact, the welfare of the state’s citizens beyond the borders of the particular municipality cannot be disregarded and must be recognized and served.

This essential was distinctly pointed out in Euclid, where Mr. Justice Sutherland specifically referred to * * the possibility of cases where the general public interest would so far outweigh the interest of the municipality that the municipality would not be allowed to stand in the way.” (272 U. S. at 390, 47 S. Ct. at 119, 71 L. Ed. at 311). Chief Justice Vanderbilt said essentially the same thing, in a different factual context, in the early leading case of Duffcon Concrete Products, Inc. v. Borough of Cresskill, 1 N. J. 509 (1949), when he spoke of the necessity of regional considerations in zoning and added this:

* * * The effective development of a region should not and cannot be made to depend upon the adventitious location of municipal boundaries, often prescribed decades or even centuries ago, and based in many instances on considerations of geography, of commerce, or *178of politics that are no longer significant with respect to zoning. The direction of growth of residential areas on the one hand and of industrial concentration on the other refuses to be governed by such artificial lines. Changes in methods of transportation as well as in living conditions have served only to accentuate the unreality in dealing with zoning problems on the basis of the territorial limits of a municipality. (1 N. J. at 513).

See, to the same general effect, Borough of Cresskill v. Borough of Dumont, 15 N. J. 238, 247-249 (1954).

In recent years this court has once again stressed this non-local approach to the meaning of “general welfare” in cases involving zoning as to facilities of broad public benefit as distinct from purely parochial interest. See Roman Catholic Diocese of Newark v. Ho-Ho-Kus Borough, 42 N. J. 556, 566 (1964), id., 47 N. J. 211 (1966). In this case we pointed out local action with respect to private educational projects largely benefitting those residing outside the borough must be exercised “with due concern for values which transcend municipal lines.” (47 N. J. at 218). Likewise in Kunzler v. Hoffman, 48 N. J. 277 (1966), a case unsuccessfully attacking a use variance granted a private hospital to serve the emotionally disturbed in a wide area of the state, we rejected the contention that local zoning authorities are limited to a consideration of only those benefits to the general welfare which would be received by residents of the municipality, pointing out that “general welfare” in the context there involved “comprehends the benefits not merely within municipal boundaries but also those to the regions of the State relevant to the public interest to be served.” 48 N. J. at 288.

This brings us to the relation of housing to the concept of general welfare just discussed and the result in terms^of land use regulation which that relationship mandates. Thére cannot be the slightest doubt that shelter, along with food, are the most basic human needs. See Robinson v. Cahill, supra (62 N. J. at 483). “The question of whether a citizenry has adequate and sufficient housing is certainly one of the prime considerations in assessing the general health and welfare of that body.” New Jersey Mortgage Finance Agency v. *179McCrane, 56 N. J. 414, 420 (1970). Cf. DeSimone v. Greater Englewood Housing Corp. No. 1, 56 N. J. 428, 442 (1970). The same thought is implicit in the legislative findings of an extreme, long-time need in this state for decent low and moderate income housing, set forth in the numerous statutes providing for various agencies and methods at both state and local levels designed to aid in alleviation of the need. See, e. g., Mortgage Einance Agency Law, N. J. S. A. 17:1B-5 (L. 1970, c. 38); Department of Community Affairs Demonstration Grant Law, N. J. S. A. 52:27D-61 (L. 1967, c. 82); Local Housing Authorities Law, N. J. S. A. 55:14A-2 (L. 1938, c. 19); Housing Co-operation Law, N. J. S. A. 55:14B-2 (D. 1938, c. 20); Redevelopment Companies Law, N. J. S. A. 55:14D-2 (L. 1944, c. 169); State Housing Law, N. J. S. A. 55:14H—2 (L. 1949, c. 303); Senior Citizens Nonprofit Rental Housing Tax Law, N. J. S. A. 55:141-2 (L. 1965, c. 92); Housing Finance Agency Law, N. J. S. A. 55:14J-2 (L. 1967, c. 81); Limited-Dividend Nonprofit Housing Corporations or Associations Law, N. J. S. A. 55:16-2 (as amended D. 1967, c. 112).

It is plain beyond dispute that proper provision for adequate housing of all categories of people is certainly an absolute essential in promotion of the general welfare required in all local land use regulation. Further the universal and constant need for such housing is so important and of such broad public interest that the general welfare which developing municipalities like Mount Laurel must consider extends beyond their boundaries and cannot be parochially confined to the claimed good of the particular municipality. It has to follow that, broadly speaking, the presumptive obligation arises for each such mpinieipality affirmatively to plan and provide, by its land use regulations, the reasonable opportunity for an appropriate variety and choice of housing, including, of course, low and moderate cost housing, to meet the needs, desires and resources of all categories of people who may desire to live within its boundaries. Negatively, it *180may not adopt regulations or policies which thwart or preclude that opportunity.

It is also entirely clear, as we pointed out earlier, that most developing municipalities, including Mount Laurel, have not met their affirmative or negative obligations, primarily for local fiscal reasons. Governor Cahill summed it up in his 1970 special legislative message, A Blueprint for Housing in New J ersey, supra, at 10-11:

We have reached a point in the State where the zoning criteria in many municipalities is two-fold; dwelling units of all kinds must be curtailed; industrial development must be encouraged. This is a far cry from the original concept of municipal zoning and planning * * *. The fundamental objective of (the) constitutional amendment and the implementing Municipal Zoning Enabling Act was local control of zoning and planning for the purpose of effecting the public good * • *. The original concept of local planning and zoning never contemplated prohibition in lieu of regulation nor the welfare of the few in place of the general welfare.

The exclusionary details are fully set forth in Land Use Regulation, The Residential Land Supply, previously referred to.

In sum, we are satisfied beyond any doubt that, by reason of the basic importance of appropriate housing and the longstanding pressing need for it, especially in the low and moderate cost category, and of the exclusionary zoning practices of so many municipalities, conditions have changed, and consistent with the warning in Pierro, supra, judicial attitudes must be altered from that espoused in that and other cases cited earlier, to require, as we have just said, a broader view of the general welfare and the presumptive obligation on the part of developing municipalities at least to afford the opportunity by land use regulations for appropriate housing for all.

We have spoken of this obligation of such municipali-. ties as “presumptive.” The term has two aspects, procedural and substantive. Procedurally, we think the basic importance of appropriate housing for all dictates that, when it is shown that a developing municipality in its land use regulations has *181not made realistically possible a variety and choice of housing, including adequate provision to afford the opportunity for low and moderate income housing or has expressly prescribed requirements or restrictions which preclude or substantially hinder it, a facial showing of violation of substantive due process or equal protection under the state constitution has been made out and the burden, and it is a heavy one, shifts to the municipality to establish a valid basis for its action or non-action. Robinson v. Cahill, supra, 62 N. J. at 491-492, and cases cited therein. The substantive aspect of “presumptive” relates to the specifics, on the one hand, of what municipal land use regulation provisions, or the absence thereof, will evidence invalidity and shift the burden of proof and, on the other hand, of what bases and considerations will carry the municipality’s burden and sustain what it has done or failed to do. Both kinds of specifics may well vary between municipalities according to peculiar circumstances.

We turn to application of these principles in appraisal of Mount Laurel’s zoning ordinance, useful as well, we think, as guidelines for future application in other municipalities.

The township’s general zoning ordinance (including the cluster zone provision) permits, as we have said, only one type of housing — single-family detached dwellings. This means that all other types — multi-family including garden apartments and other kinds housing more than one family, town (row) houses, mobile home parks — are prohibited.12 *182Concededly, low and moderate income housing has been intentionally excluded. While a large percentage of the population living outside of cities prefers a one-family house on its own sizeable lot, a substantial proportion do not for various reasons. Moreover, single-family dwellings are the most expensive type of quarter’s and a great number of families cannot afford them.13 Certainly they are not pecuniarily feasible for low and moderate income families, most young people and many elderly and retired persons, except for some of moderate income by the use of low cost construction on small lots.

As previously indicated, Mount Laurel has allowed some multi-family (housing by agreement in planned unit developments, but only for the relatively affluent and of no benefit to low and moderate income families. And even here, the contractual agreements between municipality and developer sharply limit the number of apartments having more than one bedroom.14 While the township’s PUD ordinance has been repealed, we mention the subject of bedroom restriction because, assuming the overall validity of the PUD technique (see footnote (5), supra), the measure could be *183reenacted and the subject is of importance generally. The design of such limitations is obviously to restrict the number of families in the municipality having school age children and thereby keep down local education costs.15 Such restrictions are so clearly contrary to the general welfare as not to require further discussion. Cf. Molino v. Mayor and Council of Borough of Glassboro, 116 N. J. Super. 195 (Law Div. 1971).

Mount Laurel’s zoning ordinance is also so restrictive in its minimum lot area, lot frontage and building size requirements, earlier detailed, as to preclude single-family housing for even moderate income families. Eequired lot area of at least 9,375 square feet in one remaining regular residential zone and 20,000 square feet (almost half an acre) in the other, with required frontage of 75 and 100 feet, respectively, cannot he called small lots and amounts to low density zoning, very definitely increasing the cost of purchasing and improving land and so affecting the cost of housing.16 As to building size, the township’s general requirements of a minimum dwelling floor area of 1,100 square feet for all one-story houses and 1,300 square feet for all of one and one-half stories or higher is without regard to required minimum lot size or frontage or the number of occupants (see Sente v. Mayor *184and Municipal Council of City of Clifton, 66 N. J. 204, 208-209 (1974)). In most aspects these requirements are greater even than those approved in Lionshead Lake, Inc. v. Township of Wayne, supra, 10 N. J. 165, almost 24 years ago and before population decentralization, outer suburban development and exclusionary zoning had attained today’s condition. See also Williams and Wacks, Segregation of Residential Areas Along Economic Lines: Lionshead Lake Revisited, 1969 Wis. L. Rev. 827.17 Again it is evident these requirements increase the size and so the cost of housing. The conclusion is irresistible that Mount Laurel permits only such middle and upper income housing as it believes will have sufficient taxable value to come close to paying its own governmental ivay.

Akin to large lot, single-family zoning restricting the population is the zoning of very large amounts of land for industrial and related uses. Mount Laurel has set aside almost 30% of its area, over 4,100 acres, for that purpose; the only residential use allowed is for farm dwellings. In almost a decade only about 100 acres have been developed industrially. Despite the township’s strategic location for motor transportation purposes, as intimated earlier, it seems plain that the likelihood of anywhere near the whole of the zoned area being used for the intended purpose in the foreseeable future is remote indeed and that an unreasonable amount of land has thereby been removed from possible residential development, again seemingly for local fiscal reasons.18

*185Without further elaboration at this point, our opinion is that Mount Laurel’s zoning ordinance is presumptively contrary to the general welfare and outside the intended scope of the zoning power in the particulars mentioned. A facial showing of invalidity is thus established, shifting to the municipality the burden of establishing valid superseding reasons for its action and non-action.19 We now examine the reasons it advances.

The township’s principal reason in support of its zoning plan and ordinance housing provisions, advanced especially strongly at oral argument, is the fiscal one previously adverted to, i. e., that by reason of New Jersey’s tax structure which substantially finances municipal governmental and educational costs from taxes on local real property, every municipality may, by the exercise of the zoning power, allow only such uses and to such extent as will be beneficial to the local tax rate. In other words, the position is that any municipality may zone extensively to seek and encourage the “good” tax ratables of industry and commerce, and limit the permissible types of housing to those having the fewest school children or to those providing sufficient value to attain or approach paying their own way taxwise.

We have previously held that a developing municipality may properly zone for and seek industrial ratables to create a better economic balance for the community vis-a-vis educational and governmental costs engendered by residential development, provided that such was “* * * done reasonably as part of and in furtherance of a legitimate comprehensive plan for the zoning of the entire municipality.” Gruber v. Mayor and Township Committee of Raritan Township, 39 N. J. 1, 9—11 (1962). We ad*186here to that view'today. But we were not there concerned with, and did not pass upon, the validity of municipal exclusion by zoning of types of housing and kinds of people for the same local financial end. We have no hesitancy in now saying, and do so emphatically, that, considering the basic importance of the opportunity for appropriate housing for all classes of our citizenry, no municipality may exclude or limit categories of housing for that reason or purpose. While we fully recognize the increasingly heavy burden of local taxes for municipal governmental and school costs on homeowners, relief from the consequences of this tax system will have to be furnished by other branches of government. It cannot legitimately be accomplished by restricting types of housing through the zoning process in developing municipalities.

The? propriety of zoning ordinance limitations on housing for ecological or environmental reasons seems also to be suggested by Mount Laurel in support of the one-half acre minimum lot size in that very considerable portion of the township still available for residential development. It is said that the area is without sewer or water utilities and that the soil is such that this plot size is required for safe individual lot sewage disposal and water supply. The short answer is that, this being flat land and readily amenable to such utility installations, the township could require them as improvements by developers or install them under the special assessment or other appropriate statutory procedure. The present environmental situation of the area is, therefore, no sufficient excuse in itself for limiting housing therein to single-family dwellings on large lots. Cf. National Land and Investment Co. v. Kohn, 419 Pa. 504, 215 A. 2d 597 (1965). This is not to say that land use regulations should not take due account of ecological or environmental factors or problems. Quite the contrary. Their importance, at last being recognized, should always be considered. Generally only a relatively small portion of a de*187veloping municipality will be involved, for, to have a valid effect, the danger and impact must be substantial and very real (the construction of every building or the improvement of every plot has some environmental impact) — not simply a makeweight to .support exclusionary housing measures or preclude growth — and the regulation adopted must be only that reasonably necessary for public protection of a vital interest. Otherwise difficult additional problems relating to a “taking” of a property owner’s land may arise. See AMG Associates v. Township of Springfield, 65 N. J. 101, 112, n. (4) (1974).

By way of summary, what we have said comes down to this. As a developing municipality, Mount Laurel must, by its land use regulations, make realistically possible the opportunity for an appropriate variety and choice of housing for all categories of people who may desire to live there, of course including those of low and moderate income. It must permit multi-family housing, without bedroom or similar restrictions, as well as small dwellings on very small lots, low cost housing of other types and, in general, high density zoning, without artificial and unjustifiable minimum requirements as to lot size, building size and the like, to meet the full panoply of these needs. Certainly when a municipality zones for industry and commerce for local tax benefit purposes, it without question must zone to permit adequate housing within the means of the employees involved in such uses. (If planned unit developments are authorized, one would assume that each must include a reasonable amount of low and moderate income housing in its residential “mix,” unless opportunity for such housing has already been realistically provided for elsewhere in the municipality.) The amount of land removed from residential use by allocation to industrial and commercial purposes must be reasonably related to the present and future potential for such purposes. In other words, *188such municipalities must zone primarily for the living welfare of people and not for the benefit of the local tax rate.20

We have earlier stated that a developing municipality’s obligation to afford the opportunity for decent and adequate low and moderate income housing extends at least to “* * * the municipality’s fair share of the present and prospective regional need therefor.”21 Some comment on *189that conclusion is in order at this point. Frequently it might be sounder to have more of such housing, like some specialized land uses, in one municipality in a region than in another, because of greater availability of suitable land, location of employment, accessibility of public transportation or some other significant reason. But, under present New Jersey legislation, zoning must be on an individual municipal basis, rather than regionally.22 So long as that situation persists under the present tax structure, or in the absence of some kind of binding agreement among all the municipalities of a region, we feel that every municipality therein must bear its fair share of the regional burden. (In this respect our holding is broader than that of the trial court, which was limited to Mount Laurel-related low and moderate income housing needs.)

The composition of the applicable “region” will necessarily vary from situation to situation and probably no hard and fast rule will serve to furnish the answer in every case. Con*190finement to or within a certain county appears not to be realistic, but restriction within the boundaries of the state seems practical and advisable. (This is not to say that a developing municipality can ignore a demand for housing within its boundaries on the part of people who commute to work in another state.) Here we have already defined the region at present as “those portions of Camden, Burlington and Gloucester Counties within a semicircle having a radius of 20 miles or so from the heart of Camden City.” The concept of “fair share” is coming into more general use and, through the expertise of the municipal planning adviser, the county planning boards and the state planning agency, a reasonable figure for Mount Laurel can be determined, which can then be translated to the allocation of sufficient land therefor on the zoning map. See generally, New Jersey Trends, eh. 27, Listolcin, Fair Share Housing Distribution: An Idea Whose Time Has Gome?, p. 353.23 We may add that we think that, in arriving at such a determination, the type of information and estimates, which the trial judge (119 N. J. Super. at 178) directed the township to compile and furnish to him, concerning the housing needs of persons of low and moderate income now or formerly residing in the township in substandard dwellings and those presently employed or reasonably expected to be employed therein, will be pertinent.

There is no reason why developing municipalities like Mount Laurel, required by this opinion to afford the opportunity for all types of housing to meet the needs of various categories of people, may not become and remain attractive, viable communities providing good living and adequate services for all their residents in the kind of atmosphere which a democracy and free institutions demand. They can have industrial sections, commercial sections and sections for every kind of housing from low cost and multi-family to lots *191of more than an acre with very expensive homes. Proper planning and governmental cooperation can prevent over-intensive and too sudden development, insure against future suburban sprawl and slums and assure the preservation of open space and local beauty. We do not intend that developing municipalities shall be overwhelmed by voracious land speculators and developers if they use the powers which they have intelligently and in the broad public interest. Under our holdings today, they can be better communities for all than they previously have been.

Ill

The Remedy

As outlined at the outset of this opinion, the trial court invalidated the zoning ordinance in toto and ordered the township to make certain studies and investigations and to present to the court a plan of affirmative public action designed “to enable and encourage the satisfaction of the indicated needs” for township related low and moderate income housing. Jurisdiction was retained for judicial consideration and approval of such a plan and for the entry of a final order requiring its implementation.

We are of the view that the trial court’s judgment should be modified in certain respects. We see no reason why the entire zoning ordinance should be nullified. Therefore we declare it to be invalid only to the extent and in the particulars set forth in this opinion. The township is granted 90 days from the date hereof, or such additional time as the trial court may find it reasonable and necessary to allow, to adopt amendments to correct the deficiencies herein specified. It is the local function and responsibility, in the first instance at least, rather than the court’s, to decide on the details of the same within the guidelines we have laid down. If plaintiffs desire to attack such amendments, they may do so by supplemental complaint filed in this cause within 30 days of the final adoption of the amendments.

*192We are not at all sure what the trial judge had in mind as ultimate action with reference to the approval of a plan for affirmative public action concerning the satisfaction of indicated housing needs and the entry of a final order requiring implementation thereof. Courts do not build housing nor do municipalities. That function is performed by private builders, various kinds of associations, or, for public housing, by special agencies created for that purpose at various levels of government. The municipal function is initially to provide the opportunity through appropriate land use regulations and we have spelled out what Mount Laurel must do in that regard. It is not appropriate at this time, particularly in view of the advanced view of zoning law as applied to housing laid down by this opinion, to deal with the matter of the further extent of judicial power in the field or to exercise any such power. See, hmvever, Pascack Association v. Mayor and Council of Township of Washington, 131 N. J. Super. 195 (Law Div. 1974), and cases therein cited, for a discussion of this question. The municipality should first have full opportunity to itself act without judicial supervision. We trust it will do so in the spirit we have suggested, both by appropriate zoning ordinance amendments and whatever additional action encouraging the fulfillment of its fair share of the regional need for low and moderate income housing may be indicated as necessary and advisable. (We have in mind that there is at least a moral obligation in a municipality to establish a local housing agency pursuant to state law to provide housing for its resident poor now living in dilapidated, unhealthy quarters.) The portion of the trial court’s judgment ordering the preparation and submission of the aforesaid study, report and plan to it for further action is therefore vacated as at least premature. Should Mount Laurel not perform as we expect, further judicial action may be sought by supplemental pleading in this cause.

The judgment of the Law Division is modified as set forth herein. No costs.

*193Mountain, J.

(concurring). I agree with the conclusions reached in the Court’s opinion and essentially with the opinion itself. In one important respect, however, I disagree. The Court rests its decision upon a ground of State constitutional law. I reach the same result by concluding that the term, “general welfare,” appearing in N. J. S. A. 40:55-32, can and should properly be interpreted with the same amplitude attributed to that phrase in the opinion of the Court, as well as otherwise in the manner there set forth. I therefore would rest the conclusions we here announce upon an interpretation of the statute, and not upon the State constitution.

Accordingly, since I read the statute — without resort to the Constitution — to justify, if not compel, our decision, I find it unnecessary to express any view as to the merits of the constitutional argument set forth in the Court’s opinion.

Pashman, J.

(concurring). With this decision, the Court begins to cope with the dark side of municipal land use regulation — the use of the zoning power to advance the parochial interests of the municipality at the expense of the surrounding region and to establish and perpetuate social and economic segregation.

The problem is not a new one. Early opponents of zoning advanced the possibility of such abuse as an argument against allowing municipalities the power to zone. See, e. g., Ambler Realty Co. v. Euclid, 297 F. 307, 316 (N. D. Ohio 1924), rev’d 272 U. S. 365, 47 S. Ct. 114, 71 L. Ed. 303 (1926). Later, even those sympathetic to the goals and methods of zoning began to express concern. See, e. g., Haar, “Zoning for Minimum Standards: The Wayne Township Case,” 66 Harv. L. Rev. 1051 (1953). In that spirit, Justice Jacobs wrote for this Court in Pierro v. Baxendale, 20 N. J. 17, 29 (1955):

We are aware of the extensive academic discussion following the decisions in the Lionshead and Bedminster cases, supra, and the suggestion that the very broad principles which they embody may intensify dangers of economic segregation which even the more tradi*194tional modes of zoning entail. * * * In the light of existing population and land conditions within our State these powers may fairly be exercised without in anywise endangering the needs or reasonable expectations of any segments of our people. If and when conditions change, alterations in zoning restrictions and pertinent legislative and judicial attitudes need not be long delayed. ''

The growth of the new suburbs, first as affluent residential communities and, more recently, as sites for commercial and industrial development, leaving persons with low or even moderate incomes housed inadequately in the cities and the older, inner suburbs, far from new sources of employment, magnified the importance of the problem, moving it from the realm of speculation to that of physical and social reality. Justice Hall was among the first to recognize the new significance of the problem in his now classic dissent to Vickers v. Gloucester Tp., 37 N. J. 232, 252 (1962), appeal dismissed 371 U. S. 233, 83 S. Ct. 326, 9 L. Ed. 2d 495 (1963). The facts of this ease, as well as the information compiled by various governmental agencies, of which the Court may take notice, e. g., Nat’l Comm’n on Urban Problems, Building the American City, H. R. Doc. No. 34, 91st Cong. 1st Sess. 211 (1968); N. J. Dept. of Community Affairs, Land Use Regulation: The Residential Land Supply (1972),1 demonstrate that judicial action in this area is long overdue.

Therefore, I join in the thoughtful and eloquent majority opinion of Justice Hall. I differ from the majority only in that I would have the Court go farther and faster in its implementation of the principles announced today. The fact that abuses of the municipal zoning power are now widespread and derive from attitudes and premises deeply ingrained in the suburban planning and zoning processes requires that the Court not restrict itself to the facts of this particular case but, rather, lay down broad guidelines for judicial review of municipal zoning decisions which implicate *195these abuses. Cf. Busik v. Levine, 63 N. J. 351, 363-64 (1973), appeal dismissed 414 U. S. 1106, 94 S. Ct. 831, 38 L. Ed. 2d 733 (1973).

I

The misuse of the municipal zoning power at issue in this case, generieally described as “exclusionary zoning,” see, e. g., BrooTcs, Exclusionary Zoning 8 (Am. Soc’y of Planning Officials 1970), involves two distinct but interrelated practices: (1) the use of the zoning power by municipalities to take advantage of the benefits of regional development without having to bear the burdens of such development; and (2) the use of the zoning power by municipalities to maintain themselves as enclaves of affluence or of social homogeneity.

Both of these practices are improper and to be strongly condemned. They are violative of the requirement, found both in the Constitution of 1947, Art. I, § 1 and the zoning enabling statute itself, N. J. S. A. 40:55-32, that municipal zoning ordinances further the general welfare. Cf. Cresskill v. Dumont, 15 N. J. 238, 247-49 (1954); Duffcon Concrete Products, Inc. v. Cresskill, 1 N. J. 509 (1949). They aré inconsistent with the fundamental premise of the New Jersey zoning legislation that zoning is concerned with the physical condition of the municipality not its social condition. In a deeper sense, they are repugnant to the ideals of the pluralistic democracy which America has become.

The motivation for exclusionary zoning practices are deeply embedded in the nature of suburban development. In part, these practices are motivated by fear of the fiscal consequences of opening the community to all social and economic classes. Residents of the municipality anticipate that higher density development will require the construction of additional roads, sewers, and water systems, the provision of additional municipal services, and the increase of school expenditures, all of which must be financed through local property taxes. Often, although not universally, this is a reasonable concern, *196see generally Sternlieb, Residential Development, Urban Growth and Municipal Costs (1973); N. J. Cty. & Mun. Gov. Study Comm’n, Housing & Suburbs: Fiscal & Social Impact of Multifamily Development (1974), and, as long as these costs are primarily financed through local property taxes, will continue to impel suburban communities to use the zoning laws to encourage commercial development and discourage settlement of less affluent families. But cf. Robinson v. Cahill, 62 N. J. 473 (1973), cert. denied 414 U. S. 976, 94 S. Ct. 292, 38 L. Ed. 2d 219 (1973). Insofar as this fiscal situation prevails, suburban communities will find the temptation of exclusionary zoning alluring.

In addition, exclusionary zoning practices are also often motivated by fear of and prejudices against other social, economic, and racial groups.2 Nat’l Comm. Against Discrimination in Housing, Jobs and Housing: Final Summary Report on the Housing Component, 25-29 (1972). Thus, in a recent survey of suburban municipal leaders, 42.6% identified social and racial conflict as being the chief impact of low and moderate cost subsidized housing on the municipality, while only 21.3% identified fiscal problems as the chief impact. Cty. & Mun. Gov. Study Comm’n, Housing & Suburbs: Fiscal & Social Impact of Multifamily Development, supra at 86. A large proportion felt that even State assumption of the additional municipal costs of a balanced housing policy would not make a great impact on the general unacceptability of low or moderate income housing. Id. at 89. ETor are these attitudes, however disappointing we may find them at this late date, wholly surprising. Many people who settle in suburban areas do so with the specific intention of living in affluent, socially homogeneous communities and of escaping what they perceive to be the problems of the cities. See generally, Clawson, Suburban Land Conversion in the *197United States, 45 (1971). They do not wish their insular communities to be disturbed by the introduction of diverse social, racial, and economic groups. The experience of the nation over the past 20 years must serve as a caution that, however much we might wish it, we cannot expect rapid, voluntary reversal of such attitudes.

Exclusionary zoning may assume a wide variety of forms. Ultimately, the existence of such practices must be measured by exclusionary intent and actual or potential exclusionary effect. Cf. Hawkins v. Shaw, 437 F. 2d 1286 (5 Cir. 1971); Hobson v. Hansen, 269 F. Supp. 401 (D. D. C. 1967), aff'd sub nom. Smuck v. Hobson, 132 U. S. App. D. C. 372, 408 F. 2d 175 (D. C. Cir. 1969). Some zoning devices, however, which are inherently exclusionary in effect or which lend themselves especially readily to abuse have come into wide-spread use and are a revealing gauge of the extent of exclusionary zoning in New Jersey:

1) Minimum house size requirements

As of 1970, 92% of the land in the Department of Community Affairs study area3 zoned for single family housing was covered by some minimum house size requirement. More than 65% was zoned for houses with 1,000 square feet or more of floor space, and 38.9% for houses of 1,200 square feet or more. By contract, the controversial case of Lionshead *198Lake v. Wayne Tp., 10 N. J. 165 (1953), appeal dismissed 344 U. S. 919, 73 S. Ct. 386, 97 L. Ed. 708 (1953), upheld a miuimum of 768 square feet in all districts. There is wide variation from county to county and within the various counties.4 In the so-called “outer-ring” counties in northern New Jersey — Morris, Somerset, Middlesex, and Monmouth5 — houses of less than 1,000 square feet may be built on only about 10% of the land zoned for single family dwellings. On 77% of the land zoned for single family dwellings,- houses must have 1,200 square feet or more of floor space. In the South Jersey outer-ring counties, Burlington, Camden, and Gloucester, the figures are 31.9% and 43.5% respectively.

The effect on the cost of housing of such requirements is obvious. If one assumes construction costs of $20 per square foot of floor space,6 a 1,000 square foot minimum imposes a corresponding minimum figure of $20,000 upon the portion of the cost of a new house attributable to construction. A recent *199study of housing costs indicates that floor space is the single most important factor contributing to differences in prices for new housing, even more important than the socio-economic status of the municipality. Sagalyn & Sternlieb, supra at 48. 2) Minimum lot size and minimum frontage requirements

On two-thirds of the land in the Department of Community Affairs study area zoned in 1970 for single family dwellings, houses could not be built on lots of less than an acre. Upon only 5.1% could houses be built on 10,000 square feet or less. Approximately 10% of such land in the outer-ring counties in South Jersey was zoned for 10,000 square foot lots or less; 45.9% was zoned for an acre or more. In the North Jersey outer-ring counties only 1.2% of the land zoned for single family dwellings was available for use as lots of 10,000 square feet or less; 77% was zoned for one acre or larger lots. Here, too, there are wide variations among counties. In Camden, 24.5% of the land was zoned for lots of 10,000 square feet or less, and less than 34% for lots of an acre or more. In Somerset County, only .2% of the land was zoned for lots of 10,000 square feet or less; 85.3% was zoned for lots of an acre or more, and 24.6% was zoned for three acres or more. By way of comparison, the American Public Health Association, a vigorous advocate of high minimum standards, recommends 6,000 square feet as a suitable minimum lot size based upon health considerations. Am. Public Health Ass’n, Planning the Neighborhood, 37 (1948).

Minimum frontage requirements frequently, although not invariably, are found together with minimum lot size requirements. The Residential Land Supply at 21-24. Only 13.5% of the land zoned in 1970 for single family housing in the Department of Community Affairs study area was zoned for 100 foot minimum frontage or less. In that area, 54.3% was zoned for 150 feet or more. This device was widely used in the northern outer-ring counties, where only 5% of the land is zoned for less than 100 foot frontage and 68.4% is zoned for more than 150 feet, but somewhat less widely used in the southern outer-ring counties, where 22.7% of *200the land was zoned for less than 100 foot frontage and 42.5% was zoned for more than 150 feet.

Analysis of the exclusionary impact of the widespread use of minimum lot size and minimum frontage requirements is a more complex task than that of analyzing minimum building requirements. See generally, Building the American City, supra at 213-15; Williams & Norman, supra at 493-97; Sagalyn & Sternlieb, supra at 6-16, 66-67. There is a significant correlation between lot size and price of housing in areas without sewage service and between frontage and price in areas with sewage service. Sagalyn & Sternlieb, supra at 54-56. At the very least, it can be said with certainty that extensive mapping for large lots or large lot widths drives up the cost of smaller lots and thereby significantly raises the overall price of housing. Williams & Norman, supra at 496-97.

3) Prohibition of multifamily housing

Realistically, much of the housing needs of persons with low or moderate incomes will have to be met through various forms of multifamily housing. Williams & Norman, supra at 481. Hence, restrictions upon the construction of such housing have a highly exclusionary effect. In the Department of Community Affairs study area, construction of multifamily housing was permitted on only 6.2% of the land zoned for residential uses. If six aberrant rural municipalities are disregarded, the percentage falls to 1.1%.7 In the South Jersey outer-ring counties, 2.7% of such land is zoned for multifamily housing; in the northern outer-ring counties, only of 1% is so zoned. There is no land zoned for multi*201family housing in Somerset County and only .006% is so zoned in Monmouth County.8

4) Bedroom restrictions

The effect of zoning against multifamily dwellings is magnified by restrictions upon the number of bedrooms which may be included in each dwelling unit. In the Department of Community Affairs study area9 59% of the already limited area of land zoned for multifamily dwellings is restricted to one-bedroom or efficiency apartments. On only 20.4% of this land is construction of apartments of three or more bedrooms permitted. In addition, in many areas where some construction of larger apartments is permitted, they are limited to a small percentage of any individual development. The Residential Land Supply, supra at 11. In the North Jersey outer-ring counties, 78% of the land zoned for multifamily housing is burdened with bedroom restrictions. In the South Jersey outer-ring counties, 83% of the land zoned for multifamily housing is so restricted. The situation is particularly acute in Burlington County, where 95.8% of the land zoned for multifamily housing has bedroom restrictions.

The Department of Community Affairs concluded from these figures that:

*202* * * [I]n general, the multi-family zoned land is geared to accommodate the housing needs of single people, married couples without children, and retired people, and not geared to the housing needs of the large part of the population living as families with children. [The Residential Land Supply, supra at 12; footnote omitted],

5) Prohibition of mobile homes

Mobile homes offer an alternate, less expensive form of housing. They have long since ceased to be mere “house trailers” but have become an important form of mass produced semi-permanent housing. Indeed, for many persons they may be the only form of new housing available. However, only .1% of the land zoned for residential use in the Department of Community Affairs study area was zoned for use by mobile homes. In the South Jersey outer-ring counties, .3% of the residential land was so zoned, the bulk of it being in Gloucester County, which had twice as much land zoned for mobile homes as the rest of the study area combined. None was zoned for this purpose in Camden County. No land was zoned for mobile homes in the northern outer-ring counties.10

6) Ov&rzoning for nonresidential uses

Zoning a great proportion of the developable land in a municipality noncumulatively for nonresidential uses may have the effect of forcing the price of land zoned for residential purposes up beyond the reach of persons with low or moderate incomes. Neither statewide nor countywide figures provide unambiguous evidence of the use of such practice at present in New Jersey. Land Use Regulation, supra at 6-8; Sagalyn & Sternlieb, supra at 96. At the municipal level, the use of such practices is more evident in some areas. Thus, in Mt. Laurel itself, 29.2% of the land in the township, totaling 4,121 acres, is zoned for industrial uses, although *203only 100 acres within the township has actually been developed for such use in the past 10 years, and there is no reasonable prospect of industrial rises expanding to such proportions.

If anything, these figures underestimate the extent of exclusionary zoning in this State. A wide variety of other techniques may be used to achieve an exclusionary effect. In addition, a municipality need not use all of these techniques to achieve exclusionary ends. Municipalities which have large lot-size and frontage requirements may not have building-size requirements and vice versa. Thus, only 18% of the land in the Department of Community Affairs study area zoned for single family residences permitted houses with less than 1,200 square feet of floor space to be constructed on a acre or less site with 100 foot or less frontage.

Forceful judicial intervention is necessitated not only by the already widespread use of exclusionary zoning practices and by the fact that the motivations for such are deeply ingrained in the suburban zoning and planning process,' but also by certain extrinsic factors of which the Court may take notice.

First, the United States suffers from an acute national housing shortage. It has been estimated that over 10 million dwelling units would be needed to provide each family in the country with adequate housing. Building the American City, supra at 75. In New Jersey, it has been estimated that there is an immediate need for over 400,000 dwelling units. Dep’t of Community Affairs, The Housing Crisis in New Jersey, 1970 (1970).11 New Jersey, already the second most densely populated state in the country, is experiencing continuing population growth — it is estimated that by 1985 the total population will have increased from its 1970 figure of 7,200,000 to about 10,000,000. Special Message to the Legislature by Governor Cahill, A Blueprint for Housing in New *204Jersey, Dec. 7, 1970, at 1. Housing, particularly in urban areas, is deteriorating. The percentage of substandard units throughout the State increased from 14.8% in 1960 to 17.4% in 1969. In Hudson County, the increase was from 22.3% to 31.3%. Housing Crisis in New Jersey, supra at 14. Some of these units dropped out of the housing market altogether. It has been estimated that simply to keep up with population growth and to replace units which drop out of the housing market, 100,000 new units would have to be constructed in the State each year. Special Message to the Legislature by Governor Cahill, A Blueprint for Housing in New Jersey, Dec. 7, 1970, at 1. In fact, the construction of new housing in the State peaked in 1964, when permits were issued for the construction of 68,078 units, and has declined steadily since then. In 1970, permits were issued for construction of only 39,897 units. Sagalyn & Sternlieb, Zoning & Housing Costs, 98 (1972).

The brunt of this shortage is, of course, borne by persons with low or moderate incomes. As of 1970, it was estimated that not only were half of all low income families in the State obliged to live in inadequate housing, but so were approximately 125,000 families with moderate incomes. Housing Crisis in New Jersey, supra at iv. The median cost of a new single family detached house was $30,000 in the northeastern region of the country in 1969. Sagalyn & Sternlieb, supra at 20. Prices since then have risen precipitously. A study made in 1971 found median new house costs in suburban counties to range from $33,263 in Burlington to $62,500 in Somerset and $67,000 in Bergen. Id. at 22. Such 'housing was effectively beyond the reach of families with incomes of less than $15,000 per year. Housing Crisis in New Jersey, supra at 42. As of the time of that study, the median family income in New Jersey was $11,407 per year. Analyses by both the federal and state governments, Building the American City, supra at 93; Housing Crisis in New Jersey, supra at 40-43, indicate that the majority of families can afford to neither rent nor buy new housing at current prices. Other *205authorities estimate that such housing may be beyond- the financial capacity of as much as ¿4 of all the families in the State, Sagalyn & Sternlieb, supra at 64, and as much as 90% of those families in 'which the head of the household is below the age of 35. Nat'l Comm. Against Discrimination in Housing, Housing and Jobs: Final Summary Report on the Sousing Component, supra at 22. In theory, low and moderate income families should benefit even from construction of new housing which they themselves cannot afford because such housing creates vacancies which “filter down.” In reality, however, most of these vacancies are absorbed by the enormous lag between population growth and new housing construction. Sagalyn & Sternlieb, supra at 42. The housing which does “filter down” to persons with low or moderate incomes is often badly dilapidated and in deteriorating neighborhoods. Building the American City, supra at 11; Clawson, Suburban Land Conversion in the United States, 330 (1970).

The existence of this housing shortage has been amply recognized by all branches of government in this State. See, e. g., Inganamort v. Borough of Fort Lee, 62 N. J. 521 (1973); N. J. Mortgage Finance Agency v. McCrane, 56 N. J. 414 (1970); Marini v. Ireland, 56 N. J. 130 (1970) ; Mortgage Finance Agency Law, N. J. S. A. 17:1B-5 (L. 1970, c. 38); Department of Community Affairs Demonstration Grant Law, N. J. S. A. 52:27D-61 (L. 1967, c. 82); Special Message to the Legislature by Governor Cahill, A Blueprint for Sousing in New Jersey, Dec. 7, 1970; Special Message to the Legislature by Governor Cahill, New Sorizons in Sousing, Mar. 27, 1972.

Second, the growing movement of commerce and industry to the suburbs is imposing a heavy burden upon employees who are unable to obtain housing in these suburban areas. The trend, which began after World War II and has continued unabated, arises from a variety of causes — need for additional land for expansion, automated methods of handling goods which make single-floor layout of manufacturing plants economically desirable, increased access pro*206vided by superhighways, desire for aesthetic surroundings, lower suburban property taxes, etc. See generally, Clawson, Suburban Land Conversion in the United States, 40 (1971). Retail establishments have also relocated in the suburbs, taking advantage of the shift in the affluent population, the access provided by suburban highways, and the more attractive surroundings. Id. at 40-41. The result has been a shift of blue-collar jobs from the cities to the suburbs. Id. at 40. Thus in the New York metropolitan region,12 75% of the 990,000 new jobs created between 1959 and 1967 were located outside of New York City. Jobs at manufacturing production sites outside New York City increased during that period by 138,440, while such jobs within New York City diminished by 47,110. Of the 100,600 new jobs created in retailing between 1959 and 1965, 95% were located outside of New York City. The new jobs created within New York City in recent years have been confined almost exclusively to services, finance, insurance, communications, utilities, government and manufacturing headquarters offices, all of which are fields with high percentages of white-collar employment. It appears that these trends will continue into the foreseeable future. It has been estimated that between 1970 and 1985 New York City will lose another 137,700 factory jobs, and the suburbs gain 122,700. Nat'l Comm. Against Discrimination in Housing, Jobs and Housing, 6-9 (1970). Job movement in the Philadelphia metropolitan region displays an essentially identical pattern. Nat’l Comm. Against Discrimination in Housing, Impact of Housing Patterns on Job Opportunities, 21-26 (1968). This is, of course, the natural and foreseeable consequence of “fiscal zoning” that encourages the development within a municipality of commercial establishments, which are net tax-providers, and discourages the development of housing for per*207sons who would work in such establishments, on the grounds that they are net revenue-absorbers.

This trend is one that imposes unfair burdens on the worker who is locked out of suburban residential areas. For blue-collar workers, commutation from the cities to suburban job locations is both time-consuming and prohibitively expensive. There is often no access at all by public mass transit and even when such transportation is available in theory it is frequently impractical in fact. Nat’l Comm. Against Discrimination in Housing, The Impact of Housing Patterns on Job Opportunities, supra at 27-30; Nat’l Comm. Against Discrimination in Housing, Jobs and Housing, supra at 23-26. See generally, Babcock and Bosselman, Exclusionary Zoning: Land Use Requlation in the 1970s, 114-15 (1973).

Third, even as we write, development proceeds apace. Once an area is developed, it becomes much more difficult to alter its social and economic character. There is a hazard that prolonged judicial inaction will permit exclusionary practices to continue to operate and will allow presently developing communities to acquire permanent exclusionary characteristics. The concern is not that New Jersey will soon be without developable land, but that large areas now in the process of development will have already acquired irrevocably exclusionary characteristics before the courts effectively intervene. Thus, the Delaware Yalley Regional Planning Commission has estimated that the amount of developed land in the Philadelphia metropolitan area (including Burlington, Camden, Gloucester, and Mercer Counties) will increase by 38% between 1960 and 1985, Clawson, supra at 294, and the Regional Plan Association has estimated that intensive land use in the New York metropolitan area (which includes most of northern New Jersey) will double in the same period, Clawson, supra at 279.

Finally, we must take notice of the fact that the cost of building new housing has increased steadily over the past 10 years, and shows all signs of continuing to increase in *208the future. Between 1963 and 1969, the median sales price of new single-family housing in the northeastern part of the United States rose from $20,000 to $30,500. Sagalyn & Sternlieb, supra at 20. The costs of building rental housing had increased comparably. See generally, Clawson, supra at 82-83. As the costs of housing slip farther beyond the reach of persons of low and moderate incomes, the practical valué of zoning reform diminishes and becomes increasingly contingent on the establishment of new State and federal housing subsidy programs.

Today’s decision by its terms expressly concerns exclusionary zoning practices in municipalities which are developing but which “still are not completely developed and remain in the path of inevitable future residential, commercial and industrial demand and growth.” Ante at 160. As to these communities, the Court holds:

* * * [E]very such municipality must, by its land use regulations, presumptively make realistically possible an appropriate variety and choice of housing. More specifically, presumptively it cannot foreclose the opportunity of the classes of people mentioned for low and moderate income housing and in its regulations must affirmatively afford that opportunity, at least to the extent of the municipality’s fair share of the present and prospective regional need therefor. These obligations must be met unless the particular municipality can sustain the heavy burden of demonstrating peculiar circumstances which dictate that it should not be required so to do. I Ante at 174; footnote omitted].

The majority has chosen not to explore in this case either the extent of the affirmative obligations upon developing municipalities or the role of the courts in enforcing those obligations. It has also chosen not to consider the degree to which the principles applicable to developing municipalities are also applicable to rural ones and to largely developed ones. The facts set out above seem to me to demonstrate that exclusionary zoning is a problem of such magnitude and depth as to require that the Court extend these principles to all municipalities in the State, recognizing, of *209course, that they may have different implications for municipal conduct when applied in different areas, and that the Court establish a policy of active judicial enforcement, not only of the negative obligations imposed upon municipalities by this decision but also of the affirmative obligations.

II

I consider first the extent of the affirmative obligation to plan and provide for housing opportunities for persons with low and moderate incomes that municipalities assume when they choose to avail themselves of land use controls permitted by statute. Although this discussion will concern itself initially with developing municipalities, many of the same considerations also apply mutatis mutandi to developed municipalities and rural areas, as will subsequently become clear.

A municipality need not exercise at all the powers permitted it by the zoning and planning statutes, N. J. S. A. 40:55-30 et seq. and N. J. S. A. 40:55-1.1 et seq.13 Once, however, it chooses to enter the field of land use regulation it assumes a duty — one of constitutional dimensions, deriving from N. J. Const. (1947), Art. I, § 1 — to act affirmatively to provide its fair share of the low and moderate income housing necessary to meet the regional housing needs. Cf. Southern Alameda Spanish Speaking Organization v. Union City, 424 F. 2d 291, 295-6 (9 Cir. 1970); Williams, American Planning Law: Land Use and the Police Power §§ 66.15, 66.16 (1974).

The substantive content of this affirmative obligation will necessarily vary from municipality to municipality, depend*210ing upon, among other things, the intensity of the regional housing needs, the extent of previous exclusionary practices by the municipality, and the degree to which the municipality is benefiting, directly or indirectly, from; regional economic development. A factor of special importance is the sufficiency of local housing opportunities for persons who might fill jobs created by new commercial and industrial development in the locality. Cf. Building the American City, supra at 243; ALI, Model Land Development Code, § 7-405 (Ten. Draft No. 3, 1971); Babcock & Bosselman, Exclusionary Zoning: Land Use Regulation and Housing in the 1970s, 114-15 (1973).

Every developing municipality has at least a duty to consider regional housing needs in all its planning activities, both formal and informal, including its formulation of the comprehensive plan underlying its zoning ordinance, N. J. S. A. 40:55-21, its adoption of a master plan, N. J. S. A. 40:55-1.10 and its consideration of applications for zoning variances, N. J. S. A. 40:55—39, and for approval of subdivision plats, N. J. S. A. 40:55-1.14.14 In addition, since effective planning for regional needs is virtually impossible without some degree of intergovernmental cooperation, all developing municipalities also have an affirmative obligation to cooperate, where appropriate, in regional planning efforts, to cooperate, for example, with regional planning boards established pursuant to N. J. S. A. 40:27-9 and in area review procedures established under the Intergovernmental Cooperation Act, 42 U. S. C. 4231 and implemented by U. S. Office of Management and Budget Circular A-95 (July 24, 1969) and N. J. A. C. 5:42-1.1 et seq. See generally Babcock & Bosselman, Exclusionary Zoning: Land Use Regulation and Homing in the 1970s, 135-47 (1973).

*211There is little hope that the private housing construction industry will be able to satisfy the State’s housing needs in the foreseeable future, even if all exclusionary barriers are Temoved. Building the American City, supra at 93. To meet these needs, State or federal assistance will be required. This fact has been recognized by both the State Legislature and Congress in a lengthy series of statutes providing governmental subsidies for private construction and ownership of low and moderate income housing. See, e. g., Housing and Community Development Act of 1974, 88 Stat. 633 (codified at various places in 12, 42 U. S. C.); National Housing Act of 1959, § 303, as amended, 12 U. S. C. 1701q; National Housing Act, §§ 335, 336, as amended, 12 U. S. C. 1715z, 1715z-1 et seq.; Mortgage Einance Agency Law, N. J. S. A. 17:1B-4 et seq.; Housing Einance Agency Law, N. J. S. A. 55:14J-1 et seq.; Department of Community Affairs Demonstration Grant Act, N. J. S. A. 52:27D-59 ei seq. To a greater or lesser degree, all of the programs require active municipal cooperation. Eailure to actively cooperate in the implementation of such programs as effectively thwarts the meeting of regional needs for low and moderate income housing as does outright exclusion. See, e. g., Farmworkers of Florida Housing Projects, Inc. v. Delray Beach, 493 F. 2d 799 (5 Cir. 1974); Kennedy Park Homes Ass’n v. Lackawanna, 318 F. Supp. 669 (W. D. N. Y. 1970), aff’d 436 F. 2d 108 (2 Cir. 1970), cert. den. 401 U. S. 1010, 91 S. Ct. 1256, 28 L. Ed. 2d 546 (1971). Developing municipalities have a duty to make all reasonable efforts to encourage and facilitate private efforts to take advantage of these programs.

Finally, there may be circumstances in which the municipality has an affirmative duty to provide housing for persons with low and moderate ineomes through public construction, ownership, or management. See, e. g., Community Development and Housing Act of 1974, Title II, 42 U. S. C. 1401 et seq.; Local Housing Authority Law, N. J. S. A. *21255:14A-1; cf. Mahaley v. Cuyahoga Metropolitan Housing Authority, 355 F. Supp. 1257 (N. D. Ohio 1973) rev’d 500 F. 2d 1087 (6 Cir. 1974), cert. den. 419 U. S. 1108, 95 S. Ct. 781, 42 L. Ed. 2d 805 (1975).

There are certain important limitations on the scope of these affirmative obligations. While municipalities must plan and provide for regional housing needs, no municipality need assume responsibilty for more than its fair share of these needs. The purpose of land use regulation is to create pleasant, well-balanced communities, not to recreate slums in new locations. It is beyond dispute, that when the racial and socioeconomic composition of the population of a community shifts beyond a certain point, the white and affluent begin to abandon the community. While the attitudes underlying this “tipping” effect must not be catered to, the phenomenon must be recognized as a reality. See, e. g., Graves v. Romney, 502 F. 2d 1062 (8 Cir. 1974), cert den. - U. S. -, 95 S. Ct. 1354, 43 L. Ed. 2d 440 (1975); Otero v. New York City Housing Authority, 484 F. 2d 1122 (2 Cir. 1973). Municipalities have a legitimate interest in placing an upper limit on the extent of uses which, if permitted to expand without limit, might reasonably be feared to operate to the general detriment. Tidewater Oil Co. v. Carteret, 84 N. J. Super. 525 (App. Div. 1965), aff’d 44 N. J. 338 (1965). The limitation of the municipality’s affirmative duty to one of providing for its fair share of reasonable needs responds to this interest. Cf. Mass. Gen. Laws Ann., c. 40B, §§ 20-23 (a statute authorizing the state to override local zoning restrictions for low and moderate income housing projects, but limiting the municipality’s obligations to fixed annual and total maxima). A number of regions have, in response to the problem of exclusionary zoning, voluntarily sought to put such fair share housing plans into effect. See Babcock & Bosselman, supra at 109-13.

Nor need a municipality altogether give up control of the pace and sequence of development. A municipality has *213a legitimate interest in insuring that residential development proceeds in an orderly and planned fashion, that the burdens upon municipal services do not increase faster than the practical ability of the municipality to expand the capacity of those services, and that exceptional environmental and historical features are not simply concreted over. See, e. g., Golden v. Ramapo Planning Board, 30 N. Y. 2d 359, 334 N. Y. S. 2d 138, 285 N. E. 2d 291 (1972), appeal dismissed 409 U. S. 1003, 93 S. Ct. 436, 34 L. Ed. 2d 294 (1972); Construction Industry Ass’n of Sonoma County v. Petaluma, 375 F. Supp. 574 (N. D. Cal. 1974); Mass. Gen. Laws Ann., c. 40B, §§ 20, 23.15 On the other hand, such regulations must be reasonable, substantially related to the purpose which they seek to achieve, and must adopt the least exclusionary means practical. “Zoning is a means by which a governmental body can plan for the future — it may not be used as a means to deny the future.” National Land and Investment Co. v. Kohn, 419 Pa. 504, 528, 215 A. 2d 597, 610 (Pa. Sup. Ct. 1965). By way of illustration, large lot zoning is commonly rationalized as a device for preventing premature development. Such zoning, it is claimed, merely creates holding zones. In practice, however, it appears that land zoned for large lots, even where intended as an interim holding zone, tends to become frozen in a pattern of low density development. Williams & Norman, supra at 495. Such zoning is not a reasonable device for regulating the pace and sequence of development. Its effects on development, if any, are merely exclusionary.

Finally, the affirmative duty to plan and provide for regional needs does not require the municipality to make any specific piece of property available for low or moderate income housing, absent a showing that there are inadequate alternative sites realistically available for that type of de*214velopment. A municipality must zone in accordance with a comprehensive plan. N. J. S. A. 40:55-32. Once it has adopted a comprehensive plan which properly provides for the community’s fair share of the regional housing needs, it is entitled to be able to enforce that plan through its zoning ordinances. To permit a developer to com,e in at a later date and demand, as a matter of right, that a piece of property not presently zoned to permit development of low or moderate cost housing be so zoned, is to undermine the entire premise of land use regulations. Williams, supra at § 66.15; see Confederation de la Raza Unida v. Morgan Hill, 324 F. Supp. 895 (N. D. Cal. 1971). The one exception to this principle is the situation in which the developer can show that, as a matter of practical fact, sufficient land is not available for development in the areas zoned for low or moderate income housing. See, e. g., Kennedy Park Homes Association v. Lackawanna, 318 F. Supp. 669 (W. D. N. Y. 1970), aff’d 436 F. 2d 108 (2 Cir. 1970), cert. den. 401 U. S. 1010, 91 S. Ct. 1256, 28 L. Ed. 2d 546 (1971) (construction of multi-family housing in area zoned for it would perpetuate a segregated housing pattern and add to existing problem of overcrowding); Pascack Ass’n v. Washington Tp., 131 N. J. Super. 195 (Law Div. 1974) (area zoned for multi-family housing was already largely occupied by other, non-residential uses, and was burdened with other zoning requirements that made construction of low or moderate income housing impractical).

The affirmative obligations of developing municipalities so far discussed are legally binding and judicially enforceable. It is a truism that courts have no inherent expertise in matters of land use planning. They are not equipped to sit as higher planning boards and substitute their judgment for municipal bodies lawfully established for the purpose of making planning and zoning decisions. Bow & Arrow Manor v. West Orange, 63 N. J. 335, 343 (1973); Kozesnik v. Montgomery Tp., 24 N. J. 154, 167 (1957). The decision as to *215how the municipality should go about performing the affirmative duties set out above is one initially to be made by the officials of the municipality itself. Nevertheless, if the municipality has failed to take affirmative steps to make realistically possible a variety and choice of housing so as to meet its fair share of the regional housing needs, its actions are presumptively illegal and the burden shifts to the municipality to justify them. The mere fact that local land use control issues are involved does not preclude the court from making such determinations, nor, if a court finds that the municipality has failed to meet its obligation, from exercising the full panoply of equitable powers to remedy the situation. Norwalk CORE v. Norwalk Redevelopment Agency, 395 F. 2d 920 (2 Cir. 1968); Hawkins v. Shaw, 437 F. 2d 1286 (5 Cir. 1971); Pascack Ass’n v. Washington Tp., 131 N. J. Super. 195 (Law Div. 1974).

Judicial enforcement of municipal obligations, both negative and affirmative, to plan and provide for a fair share of regional housing needs, even if only directed to one municipality, necessarily has grave implications for the entire region. In dealing with such cases courts must act both deliberately and imaginatively. In administering such relief the trial court ought to proceed in four steps:

(1) identify the relevant region;16

(2) determine the present and future housing needs of the region;

(3) allocate these needs among the various municipalities in the region;17 and

*216(4) shape a suitable remedial order.

Cf. Williams, American Planning Law: Land Use and the Police Power § '66.38 (1974). Needless to say, all of these steps involve difficult factual determinations based upon expert testimony and statistical evidence. It may well be appropriate for the court to appoint independent experts or consultants for its assistance, see Pascack Ass’n v. Washington Tp., 131 N. J. Super. 195 (Law Div. 1974); cf. Handleman v. Marwen Stores Corp., 53 N. J. 404 (1969); Polulich v. J. G. Schmidt Tool Die & Stamping Co., 46 N. J. Super. 135 (Cty. Ct. 1957); Manual for Complex Litigation, Pt. 1 §§ 1.42, 1.46, 2.60, 3.40 (1973), or to invite participation by the Department of Community Affairs as amicus curiae.

Since conflicting decisions within a given region would be highly undesirable, all municipalities in the region should be joined as parties at the earliest practical point in the proceedings, if not at the instance of one of the parties, then on the motion of the court. R. 4:28-1, 4:30.

*217The trial court must be flexible aud imaginative in molding remedies to fit the facts of each case, balancing the need to vindicate the rights of persons who have been or will be deprived of the opportunity for decent housing if no relief is granted against the principle of local decision-making in land use planning matters. Pascack Ass’n v. Washington Tp., supra; see e. g., Kennedy Park Homes Ass’n v. Lackawanna, 318 F. Supp. 669 (W. D. N. Y. 1970) aff’d 436 F. 2d 108 (2 Cir. 1970) cert. den. 401 U. S. 1010, 91 S. Ct. 1256, 28 L. Ed. 2d 546 (1971); Mahaley v. Cuyahoga Metropolitan Housing Authority, 355 F. Supp. 1257 (N. D. Ohio 1973) rev'd on other grounds, 500 F. 2d 1087 (6 Cir. 1974), cert. denied 419 U. S. 1108, 95 S. Ct. 781, 42 L. Ed. 2d 805 (1975); United Farmworkers of Florida Housing Projects, Inc. v. Delray Beach, 493 F. 2d 799 (5 Cir. 1974); Lakewood Homes, Inc. v. Lima Bd. of Adjustment, 23 Ohio Misc. 211, 52 Ohio Op. 2d 213, 258 N. E. 2d 470 (Ohio Ct. C. P. 1970); mod. 25 Ohio App. 2d 125, 267 N. E. 2d 595 (Ohio Ct. App. 1971).

Ill

It can hardly be denied that there are some suburban municipalities which have already developed in an exclusionary mold. These communities, which have benefited from regional development, have, by their land use controls, contributed to the regional housing shortages. Cf. United States v. Black Jack, 372 F. Supp. 319 (E. D. Mo. 1974). It would be both highly inequitable to absolve them of any responsibility for solving those problems and inconsistent with the legal analysis developed by the Court today. Although the majority does not reach this issue in the present case, I would hold that developed suburban municipalities which have availed themselves of the land use controls permitted by statute and which have not provided sufficient opportunities for development of low and moderate income housing to meet their fair share of regional needs, have both *218a negative obligation not to use zoning and subdivision controls to obstruct the construction of such housing and an affirmative duty to plan and provide for such housing, insofar as these obligations can be carried out without grossly disturbing existing neighborhoods. It is, of course, neither practical nor wise to demand that such communities completely rezone established neighborhoods; to do so would in all likelihood contribute to neighborhood instability and permit certain property owners and developers to obtain windfalls rather than actually effecting construction of low or moderate income housing.

Occasions, however, arise in every community when land becomes available for development or redevelopment. It is on these occasions that these obligations come into play most strongly. Thus the existence of an unmet regional need for low and moderate income housing in appropriate cases must be given great weight in considering applications for variances under N. J. S. A. 40:55-39(d) to permit the construction of such housing. De Simone v. Greater Englewood Housing Corp. No. 1, 56 N. J. 428 (1970); Brunelti v. Madison Tp., 130 N. J. Super. 164 (Law Div. 1974).

The discussion above of judicial enforcement applies equally to developed suburban communities, save only that in formulating relief the trial judge must be alert to take into consideration the delicacy and difficulty of altering the character of already developed areas.

IV

Substantial portions of New Jersey are neither experiencing a surge of development nor situated in the imminently foreseeable path of development. These include much of Cape May, Cumberland, and Salem Counties, portions of Atlantic, Ocean, Sussex and Warren Counties, and some rural areas in other parts of the State. In these municipalities, it is not meaningful to speak of failure to meet regional housing needs, not because there are no persons who are inadequately *219housed,18 but because it is not yet meaningful to speak of “regional” needs nor is it clear that land use controls play a significant role in the housing shortage at the present time. Nevertheless, the time may well come when the frontiers of suburbia will reach these areas. Municipalities may not act to deter the future development of a diversified housing stock by establishing land use controls which are inherently exclusionary and which bear no substantial relationship to any legitimate zoning purpose.

Without purporting to exhaust the list of zoning devices which are presumptively objectionable, I would note that minimum house size requirements which bear no substantial relationship to health needs 19 and requirements as to the minimum or maximum number of bedrooms which a dwelling unit may contain, cf. Molino v. Glassboro, 116 N. J. Super. 195 (Law Div. 1971), are presumptively invalid. Zoning for excessively large lots and large frontages presents more difficult analytic problems, cf. Steel Hill Development, Inc. v. Sanbornton, 469 F. 2d 956 (1 Cir. 1972), but excessive mapping for such lots is, absent extraordinary environmental factors, also presumptively invalid. Cf. Williams & Norman, supra at 496-97.

These obligations, too, are judicially enforceable, albeit without need for the more elaborate procedures appropriate for litigation concerning developing and developed areas which are discussed above.

*220Y

The problems we begin to face today are of awesome magnitude and importance, both for New Jersey and for the nation as a whole. It will not do to approach them gingerly; they call out for forceful and decisive judicial action.

The flow of low and moderate income persons is toward urban areas, hut the cities have neither the space nor the resources to house these people. The question is whether the suburbs will act to accommodate this growth in an orderly way or will simply and blindly resist. Two well-entrenched zoning objectives, low density land use and favorable fiscal balance, though sometimes at odds with each other, have on the whole cooperated to create a milieu of discriminatory zoning which threatens to make the next 30 years of suburban growth a disaster.

The shape of the possible disaster can now be foreseen. The inevitable alternative to assumption by suburban communities of an obligation to provide for their fair share of regional housing needs is an increase in the size of slums with all their attendant miseries. The consequences of such economic, social, and racial segregation are too familiar to need recital here. See Nat'l Advisory Comm’n on Civil Disorders, Report (1968). Justice must be blind to both race and income.

It is not the business of this Court or any member of it to instruct the municipalities of the State of New Jersey on the good life. Fevertiheless, I cannot help but note that many suburban communities have accepted at face value the traditional canard whispered by the “blockbuster”: “When low income families move into your neighborhood, it will cease being a decent place to live.” But as there is no difference between the love of low income mothers and fathers and those of high income for their children, so there is no difference between the desire for a decent community felt by one group and that felt by the other. Many low income families have learned from necessity the desirability of community involve*221ment and improvement. At least as well as persons with higher incomes, they have learned that one cannot simply leave the fate of the community in the hands of the government, that things do not run themselves, but simply run down.

Equally important, many suburban communities have failed to learn the lesson of cultural pluralism. A homogeneous community, one exhibiting almost total similarities of taste, habit, custom and behavior is culturally dead, aside from being downright boring. New and different life styles, habits and customs are the lifeblood of America. They are its strength, its growth force. Just as diversity strengthens and enriches the country as a whole, so will it strengthen and enrich a suburban community. Like animal species that over-specialize and breed out diversity and so perish in the course of evolution, communities, too, need racial, cultural, social and economic diversity to cope with our rapidly changing times.

Einally, many suburban communities have failed to recognize to whom the environment actually belongs. By environment, I mean not just land or housing, but air and water, flowers and green trees. There is a real sense in which clean air belongs to everyone, a sense in which green trees and flowers are everyone’s right to see and smell. The right to enjoy these is connected to a citizen’s right to life, to pursue his own happiness as he sees fit provided his pursuit does not infringe another’s rights.

The people of New Jersey should welcome the result reached by the Court in this case, not merely because it is required by our laws, but, more fundamentally, because the result is right and true to the highest American ideals.

Mountain and P ashman, JJ., concurring in the result.

For modification — Chief Justice Hughes and Justices Jacobs, Hall, Mountain, Sullivan, Pashman and Clifford — 7.

Opposed — None.

Tauro, C.J.

These are two suits in equity brought by (1) the board of appeals of the town of Hanover (Hanover board) and (2) the board of appeals of the town of Concord (Concord board) each reserved and reported by a Superior Court judge without decision. We have before us the records of the proceedings before both boards and the Housing Appeals Committee (committee) as certified by the committee. The boards denied applications for comprehensive permits to build low and moderate income housing filed pursuant to the provisions of G. L. c. 40B, § § 20 and 21, and the applicants appealed to the committee pursuant to G. L. c. 40B, § 22. The committee rendered two decisions, in each case reversing the board and ordering the issuance of the permit. The suits, brought under G. L. c. 40B, § 22, and G. L. c. 30A, § 14, seek review of the committee’s decisions. The cases were argued together. The bills present similar questions concerning the constitutional validity and the substantive and procedural effects of G. L. c. 40B, § § 20-23, *343inserted by St. 1969, c. 774, § 1 (c. 774). The cases are, therefore, decided together.

The Hanover Proceedings.

In April, 1970, Country Village Corporation filed an application with the Hanover board for a comprehensive permit to construct eighty-eight units of low and moderate income housing for the elderly on approximately ten acres of land. After holding public hearings on the application, as required by G. L. c. 40B, § 21, and G. L. c. 40A, § 17, the board refused to issue the permit. The board then filed with the clerk of the town of Hanover its decision denying the permit and the reasons for the denial.1 In December, 1970, the applicant appealed to the committee. G. L. c. 40B, § 22. The committee, after holding public hearings on the appeal pursuant to G. L. c. 40B, §§22 and 23, vacated the decision of the board in July, 1971, and ordered the board to issue a comprehensive permit for the project, subject to specified conditions.2 The Hanover board then filed this bill for review in the Superior Court.

*344 The Concord Proceedings.

In January, 1971, the Concord Home Owning Corporation filed with the Concord board an application for a comprehensive permit to construct sixty garden apartment units of low and moderate income housing on approximately five and one-half acres of land. The board then held public hearings on the application as required by G. L. c. 40B, § 21, and G. L. c. 40A, § 17, and refused to isue the permit. The board filed with the clerk of the town of Concord its decision denying the permit and the reasons for the denial.3 In April, 1971, the applicant appealed to the committee. G. L. c. 40B, § 22. After holding the required public hearing on the appeal, the committee vacated the decision of the board and ordered the board to issue the applicant a comprehensive permit for the project, subject to specified conditions.4 The board then filed this bill for review in the Superior Court.

Issues Presented.

The bills present three issues for resolution concerning the powers and procedures of the boards of appeals and the committee under G. L. c. 774. We must determine:

(a) whether c. 774 confers power upon both the committee and the boards to override zoning regulations which hamper the construction of low and moderate housing;

(b) whether such power to' override zoning regulations, if it exists, is constitutional; and

*345(c) whether such power to override zoning regulations, if it exists, was properly exercised by the committee in the instant cases.

Chapter 7745 permits a qualified applicant6 interested in building low or moderate income housing to file with a board of appeals an application for a comprehensive permit instead of filing separate applications with each local agency or official having jurisdiction over various aspects of the proposed project. The statute allows the board of appeals to grant a single comprehensive permit for construction after considering the recommendations of the local agencies or officials. General Laws c. 40B, § 21, establishes a specific time period within which the board of appeals must make its decision. If the board makes no decision within this period, “the application shall be deemed to have been allowed and the comprehensive permit or approval shall forthwith issue.” § 21. If the board issues a comprehensive permit, any person aggrieved by its decision may appeal to the District or Superior Court as provided in § 21 of G. L. c. 40A.

Whenever the board of appeals denies an application or grants it with conditions which make the building or operation of the proposed housing project “uneconomic,” 7 § 22 grants the applicant a right of appeal to the committee. The committee is required to conduct a full hearing with a stenographic record within twenty days of the receipt of the applicant’s appeal and it must *346render a written decision “stating its findings of fact, its conclusions and the reasons therefor within thirty days after the termination of the hearing, unless such time shall have been extended by mutual agreement between the committee and the applicant.” Section 23 limits the committee’s review to determining whether the board of appeals’ decision to deny the application was “reasonable and consistent with local needs” 8 as defined by § 20. If the board grants the application subject to conditions, § 23 limits the committee’s review to determining whether the conditions make the construction or operation of the housing uneconomic and whether the conditions are consistent with local needs. If the board’s decision to deny the permit or to impose uneconomic conditions on its approval is found to be consistent with local needs, the committee must affirm the board’s decision. “If the committee finds, in the case of a denial, that the decision of the board of appeals was unreasonable and not consistent with local needs, it shall vacate such decision and shall direct the board to issue a comprehensive permit or approval to the applicant.” § 23. If the committee finds that the conditions imposed by the board in approving the application make the building or operation of the housing “uneconomic” and that the board’s decision is not consistent with local needs, the committee “shall order such board to modify or remove any such condition or requirement so as to make the proposal no longer uneconomic and to issue any necessary permit or approval.” The committee’s decision is subject to judicial review in accordance with c. 30A’s provisions. G. L. c. 40B, § 22.

1. Does c. 774 Confer upon both the Committee and the Boards the Power to Override Local Zoning Regulations to the Extent that such Regulations Conflict with the Implementation of c. 774?

The boards argue that the Legislature did not intend *347to grant the power to override local zoning by-laws or ordinances to any authority when it enacted c. 774. Thus, their respective decisions to deny comprehensive permits in these cases were reasonable and consistent with local needs because neither they nor the committee had authority to issue comprehensive permits for a use of land not permitted under local zoning by-laws. The boards contend that the Legislature’s purpose in enacting c. 774 was merely to provide a streamlined procedure for processing applications for the necessary local approvals of construction of low or moderate income housing. Where previously an applicant was forced to negotiate with various local agencies or officials before gaining their approval, c. 774’s new time limits expedited the process by allowing the applicant to apply for and obtain a comprehensive permit from a single agency, the board of appeals. The committee argues, to the contrary, that the text, history, and context of c. 774 indicate that the Legislature intended to confer to both the board and the committee the power to override any local requirements and regulations, including zoning bylaws, which prevented the construction of low and moderate income housing when such housing is deemed “consistent with local needs.”

To resolve this controversy over c. 774’s essential purpose, we must examine its detailed legislative history to determine the nature of the problem that the statute was designed to remedy. The boards’ interpretation of the statute rests on their argument that the Legislature was chiefly concerned with speeding up the processing of applications for the construction of low and moderate income housing. However, the legislative history of c. 774 indicates that the Legislature was more concerned with the cities’ and towns’ possible use of their zoning powers to exclude low and moderate income groups.

The legislative history of c. 774 begins with a 1967 Senate Order, No. 933, which directed the Legislative Research Council (Council) to “undertake a study and investigation relative to the feasibility and implications *348of restricting the zoning power to cities and county governments with particular emphasis on the possibility that the smaller communities are utilizing the zoning power in an unjust manner with respect to minority groups.”

The Council’s report focused on the economic discrimination which resulted from restrictive local zoning practices. “The Research Bureau was unable to uncover any recent comprehensive studies concerning possible ‘anti-minority’ uses of local zoning in Massachusetts, especially in an ethnic or religious minority sense. . . . Most complaints received by the Research Bureau were directed, instead, against alleged ‘economic discrimination’ in local zoning which reportedly impairs the effort of low and modest income people of all racial and religious origins to find homes within their financial means. Hence, this Council report emphasizes that aspect of zoning discrimination in Massachusetts. . . .” Report of the Legislative Research Council relative to Restricting the Zoning Power to City and County Governments (the Report), 1968 Senate No. 1133, p. 28.

Since the Council received information concerning the zoning practices of 113 of the Commonwealth’s 351 municipalities, its report represents a well researched statement of the unavailability of low and moderate income housing in suburban Massachusetts communities and the reasons for the shortage. The Report examined the effects of local restrictive zoning practices (i.e. minimum lot size requirements, green space zoning, minimum frontage and setback requirements, minimum floor area requirements, maximum building areas of lots, building height limitations, inspection and permit fees), and indicated that “[t]he interplay of these municipal regulations determines, in substantial degree, the extent to which additional modest income housing is possible in relation to the local supply of ‘buildable’ land.” P. 91.

What follows is a summary of the Report’s conclusions:

(1) Large lot requirements (minimum lot size) have a substantial negative effect on the availability of land *349in the suburbs which could be used for low and moderate income housing. P. 102. The Report listed twenty-one municipalities, including Hanover, that restricted 50% or more of their territory to large lot zoning. P. 98.

(2) Building height limitations were also found to have a significant negative impact on low and moderate income housing. The Report noted that forty-six municipalities had height limitations and thirty-one, including Hanover, forbade apartment buildings altogether. The Report concluded, “Beyond these height restrictions . . . lies the greater public policy issue of providing sufficient sites in metropolitan areas for adequate multifamily and apartment housing at modest rentals for that portion of the area population which cannot afford to own or to rent a single-family home .... To the extent that inner suburban communities prohibit multi-family and apartment housing, or attach height or other restrictions which make such housing feasible only on a ‘luxury’ basis, the modest income housing problems of the entire metropolitan area are aggravated.” Id. at 118.

After determining that seven of the eight local restrictive zoning practices studied had a negative impact on the construction of low and moderate income housing, the Report noted that the housing shortage problem had reached crisis proportions. In 1947, the entire State population could be housed in buildings on land as then zoned. Pp. 32-33. But by 1968, low income persons could not find places to live in the suburbs at prices which they could afford to pay. Id. at 87-88. One study estimated that by 1968, 20,000 additional housing units would be needed to accommodate low income families (id. at 121; Massachusetts Special Commisison on Low-Income Housing, Final Report, 1965 House No. 4040, at 82-83), and that 125,000 new housing units would be needed to meet population increases alone by 1975. Report at 88.

The Report concluded with the dire prediction that if existing exclusionary zoning practices by municipalities were left unregulated, the supply of vacant land would *350be eliminated by the 1990’s (id. at 102) because the communities were not inclined to act on their own to alleviate the problem (id. at 123) and the courts were unwilling to intervene as long as the discrimination involved in the exclusionary zoning practices was economic.9 The Council recommended a plan which would leave with cities and towns the general power to direct their own development but would permit, in appropriate cases, the circumvention of exclusionary zoning by-laws where their enforcement would frustrate the State’s need for more low and moderate income housing.

The Council’s Report prompted the submission of the following bills in the 1969 session of the Legislature: Senate Bills Nos. 1137 and 1141, and House Bills Nos. 2924, 3175 and 3603. All of these bills called for the imposition of State control over the suburbs’ exercise of exclusionary zoning practices in order to insure the availability of low and moderate income housing throughout the State. In different ways and to different degrees, these bills provided for the overriding of local zoning ordinances and by-laws which frustrated the construction of low and moderate income housing in the suburbs.10

These five bills were referred to the Committee on Urban Affairs. In June of 1969 that committee reported *351out House Bill No. 5429 which, in modified form, became c. 774. The committee also sent a report to explain the bill’s purpose and the problems it was designed to resolve. “The committee on urban affairs has found that there is an acute shortage of decent, safe, low and moderate cost housing throughout the commonwealth. . . . Unless land in less densely populated areas is available there will not be enough housing in any town or city for even the children of the present residents, municipal employees .... [Njecessary land in such areas is often unavailable because of restrictive zoning controls or similar local regulations. Moreover, where land is available, the process of obtaining local approval is so protracted as to discourage all but the most determined and well-financed builders . . ..” Although the bill allowed local governments to implement its provisions initially, the Committee on Urban Affairs made it clear that cities and towns would not be allowed “to unreasonably obstruct the construction of a limited amount of adequate low cost housing.”

The Concord board claims that the Council’s Report, the bills submitted in response to it, and the final draft of House Bill No. 5429 which was reported out of the Urban Affairs Committee along with its report are “misleading indication[s] of the intent of the General Court.” Although all of the prior legislative history of House Bill No. 5429 reflects a consistent attempt on the Legislature’s part to resolve the housing shortage problem by establishing an effective means of circumventing the suburbs’ exclusionary zoning provisions, the Concord board claims that this intended approach was eliminated when House Bill No. 5429 was redrafted by the Committee on Ways and Means.

However, a careful examination of the changes made by the Committee on Ways and Means reveals that the Concord board’s argument rests on a relatively insignificant rephrasing of language which changed the form but not the substance of the bill. When the bill was reported out by the Committee on Urban Affairs, the sec-*352tian concerning the committee’s review of a local board of appeals’ decision that imposed “uneconomic” conditions on the permit read, “the commission or the committee shall, unless the action of the board of appeals is found to be consistent with local needs, order such board to issue any necessary permit or approval or to modify or remove any requirement, including but not limited to zoning or building code requirements” (emphasis supplied). When the bill was reported out of the Committee on Ways and Means, 1969 House Bill No. 5381, the “including but not limited to” clause had been eliminated so the clause read “it [the committee] shall order such board to modify or remove any such condition or requirement . . . and to issue any necessary permit or approval . . ..” The omission of the superfluous “including but not limited to” phrase which delineated only part but not all of the power granted by the clause which it modified did not negate the Legislature’s primary intent to circumvent the suburbs’ exclusionary zoning practices. The insignificance of this redrafting is further emphasized by the proviso following that section which establishes minimum construction safety standards for housing built under the statute. This provision would have been unnecessary if in all instances local regulations were to apply.

Examination of the subsequent legislative history of c. 774 as redrafted by the Committee on Ways and Means (renumbered House Bill No. 5581) discloses neither a change of purpose nor a change of method on the part of the Legislature.11 The redrafted bill had the same title as the Urban Affairs Committee bill, “An Act providing for the construction of low or moderate income housing in cities and towns and for relief from local restrictions hampering such construction.” Since the title to a statute may be considered in its construction, Silverman v. *353 Wedge, 339 Mass. 244, 245, the identical titles suggest that no major change of substance was intended by the redrafting of the Committee on Ways and Means.12 Moreover, records of the legislative debate reveal that both proponents and opponents13 of the redrafted bill realized that the bill would grant the power to override local zoning regulations which hampered the construction of low and moderate income housing. The House Journal indicates that opposition to the bill was based on the bill’s incursion into the municipalities’ power to exercise a veto over low and moderate income housing. In both the House and the Senate, opponents of the bill attempted unsuccessfuly to amend the bill in a way which would have stripped the legislation of its primary effect by excluding the bill’s applicability to any city or town which refused to accept it. See 1969 House Journal, p. 2444, 1969 Senate Journal, p. 1912. Clearly, such an amendment would not have been necessary if local zoning powers were to remain inviolate under the new law because city councils and town meetings already had the power to prevent any nonconforming use.

In view of the descriptions of the redrafted bill’s effect by the legislators themselves, and considering the failure of these amendments, both in the House and Senate, which would have given local communities a veto over application of the statute to their municipality, we conclude that the Legislature’s intent in passing c. 774 was *354to provide relief from exclusionary zoning practices which prevented the construction of badly needed low and moderate income housing.

Our determination of the Legislature’s intent enables us to interpret the text of the statute which admittedly is not without its ambiguities. The statute must be construed in a manner that effectuates its intent. With this in mind, we turn to the ambiguities in the statute’s text noted by the boards.

The boards claim that the statute’s failure to refer specifically to local zoning ordinances or by-laws indicates that no power to override them was in fact conferred. However, c. 40B, § 20, in delineating when a board’s decision is consistent with local needs, refers to “requirements and regulations.” Though this reference is somewhat ambiguous, the Legislature’s clear purpose in passing this statute requires our construction of this term to include local zoning by-laws and ordinances. Although the word “zoning” is not specifically used, the statute’s legislative history and avowed purpose to facilitate the construction of low and moderate income housing in areas which have exclusionary zoning practices compel our decision to construe the statute so that zoning ordinances or by-laws are treated like any other local requirements which hamper the construction of low and moderate income housing. All these local “requirements and regulations” will be applicable if they are “consistent with local needs”; if they are not, they must be modified or ignored.

Our decision that the committee has the authority to override local requirements and regulations that are inconsistent with local needs by implication necessitates a construction of the statute which confers this .same authority upon the boards. As the Concord board noted, “if anyone has the power to override zoning ordinances or by-laws under G. L. c. 40B, §§ 20-23, then both the Committee and all local boards of appeals have that power. If the result were otherwise, it is obvious that in any case where a proposed project is not permitted by a zoning by-law, the applicant must follow the futile proce*355dure of applying for relief before a forum which has no power to grant the relief being sought.” Since we must avoid a construction of statutory language which produces irrational results (Johnson v. Commissioner of Pub. Safety, 355 Mass. 94, 99), that portion of § 21 which states that the “board of appeals . . . shall have the same power to issue permits or approvals as any local board or official who would otherwise act with respect to such application . . .” (emphasis supplied) cannot be read to exclude the additional power to override local requirements and regulations granted by § 20 and § 23. Thus, we construe § 21 to confer upon boards “the same power ... as any local board or official” in addition to that power otherwise conferred by the provisions of this act.

We have often stated that a construction of a statute which would completely negate legislative intent should be avoided. Assessors of Newton v. Pickwick Ltd. Inc. 351 Mass. 621, 625. The boards’ interpretation of this statute would have exactly this effect. Streamlining local permit procedures could not possibly serve the statute’s clear purpose of promoting the construction of low and moderate income housing if the cities and towns retained the unlimited power to enforce restrictive zoning ordinances or by-laws which prevented the construction of such housing. Therefore, we hold that c. 774 confers on boards of appeals and the Housing Appeals Committee the power to override local “requirements and regulations,” including zoning ordinances or by-laws, which are not “consistent with local needs.”

2. Does c. 77U Violate the Home Rule Amendment?

We must now determine whether the grant of such power is constitutional. There is no question that the Legislature could have conferred the power to override local zoning regulations prior to the adoption in 1966 of the Home Rule Amendment (art. 89). Prior to its amendment by art. 89, art. 2 of the Amendments gave to the Legislature the “full power and authority to erect and constitute municipal or city governments, in any *356corporate town or towns in this commonwealth, and to grant to the inhabitants . . . such powers, privileges, and immunities, not repugnant to the constitution as the general court shall deem necessary or expedient for the regulation and government thereof . . . provided . . . that all by-laws made by such municipal or city government . . . [are] subject, at all times to be annulled by the general court.” This article made the cities and towns creatures of the State, Moore v. Election Commrs. of Cambridge, 309 Mass. 303, 316-317, subject to the will of the Legislature. Commonwealth v. Hudson, 315 Mass. 335, 345.

Our interpretation of art. 2 coincided with the general principle (popularly known as Dillon’s Rule) that municipal corporations possess “and can exercise the following powers, and no others: First, those granted in express words; second, those necessarily or fairly implied in or incident to the. powers expressly granted; third, those essential to the accomplishment of the declared objects and purposes of the corporation .... Any fair, reasonable, substantial doubt concerning the existence of power is resolved by the courts against the corpo- ■ ration, and the power is denied.” Dillon, Municipal Corporations (5th ed.) § 237, pp. 449-450.

In light of this formulation of State and local relations, all municipal rights and powers were derived from their creator, the State. Since the municipalities had no inherent or independent rights, a doctrine soon developed in our case law that the State had absolute authority to remove any right or power it had originaly conferred upon a municipality. What the State gave, it could also take away because “the towns of the Commonwealth possess no inherent right to self-government.” Paddock v. Brookline, 347 Mass. 230, 238.

Thus, this court has upheld repeatedly the Legislature’s “supreme” power in zoning matters. Bennett v. Board of Appeal of Cambridge, 268 Mass. 419, 422. Attorney Gen. v. Dover, 327 Mass. 601, 604-605. The power to regulate in the first instance included within it the *357more limited power to pass State laws which carved out exceptions to local zoning ordinances or by-laws. Therefore, in Opinion of the Justices, 341 Mass. 760, 788-789, we advised that proposed legislation which would authorize the Boston Redevelopment Authority to permit a redevelopment project that deviated from local zoning law would be constitutional. We said, “Article 60 of the Amendments provides that the Legislature ‘shall have power to limit buildings according to their use or construction to specified districts of cities and towns.’ ‘The power of the General Court over the subject of zoning’ has been said to be ‘supreme’ .... [Citations omitted.] This power, together with the broad legislative power to revise the governmental structures of cities and towns and the powers and duties of municipal bodies and boards, amply warrants delegating under appropriate standards to the Authority, subject to the approval of the mayor, the power to grant zoning deviations.” If these decisions are still in effect, they would provide ample authority for upholding the Legislature’s right to enact laws which carve out exceptions to local zoning by-laws and ordinances.

However, the amici curiae contend that the adoption of the Home Rule Amendment prevents the Legislature from interfering with the municipalities’ exercise of their zoning powers and renders all of our pre-1966 zoning decisions which rely on the State’s supreme zoning power inapplicable to the present controversy.

It is true that art. 89’s adoption has “effected substantial changes in the legislative powers of the General Court and the cities and towns.” Opinion of the Justices, 356 Mass. 775, 787. The significance of our pre-home rule decisions which rested solely on the premise that no municipality has any vested right in its form of government has been considerably diminished by art. 89’s implicit repudiation of the Dillon Rule’s characterization of city-State relations. See Belin v. Secretary of the Commonwealth, 362 Mass. 530, 535.

Section 6 of art. 89 establishes a broad general grant *358of home rule powers to cities and towns. “Any city or town may, by the adoption, amendment, or repeal of local ordinances or by-laws, exercise any power or function which the general court has power to confer upon it, which is not inconsistent with the constitution or laws enacted by the general court in conformity with powers reserved to the general court by section eight, and which is not denied, either expressly or by clear implication, to the city or town by its charter.” Section 7 of art. 89 then specifies certain exceptions to § 6’s broad grant of home rule powers. Taken together, these two sections repudiate the conception that all powers lie in the State except those expressly delegated to cities and towns. The Home Rule Amendment grants cities and towns independent municipal powers which they did not previously inherently possess. Municipalities are now free to exercise any power or function, excepting those denied to them by their own charters or reserved to the State by § 7, which the Legislature has the power to confer on them, as long as the exercise of these powers is not inconsistent with the Constitution or laws enacted by the Legislature in accordance with § 8. See § 6 of art. 89 and § 13 of the Home Rule Procedures Act.14

Therefore, in order to resolve the question of whether the Legislature can enact a statute which overrides local zoning ordinances or by-laws, we must first consider the source of the municipalities’ authority to adopt local zoning ordinances or by-laws in light of the material changes which were made in art. 2 by the adoption of the Home Rule Amendment. We must decide first whether the power to zone is one of the municipalities’ independent powers conferred to them by § 6 or whether it constitutes a power expressly reserved to the Legislature by § 7.

Section 7 (5) provides that “[n]othing in this article shall be deemed to grant to any city or town the power . . . (5) to enact private or civil law governing civil relationships except as an incident to an exercise of an *359independent municipal power.” We construed the phrase “independent municipal power” in Marshal House, Inc. v. Rent Review & Grievance Bd. of Brookline, 357 Mass. 709, 718. We said “[t]he quoted vague language points, in our opinion, to viewing separately the various component powers making up the broad police power, with the consequence that a municipal civil law regulating a civil relationship is permissible . . . only as an incident to the exercise of some independent, individual component of the municipal police power.” Unlike the rent control ordinance which we struck down in the Marshal House case because of its direct effect upon civil relationships, any effect zoning ordinances or by-laws may have on civil relationships is incidental to the exercise of the town’s independent police powers to control its land usages in an orderly, efficient, and safe manner to promote the public welfare. At the very least, the term “independent municipal power” is intended “as a validation of the inevitable impact of public regulation on private rights, as, for example, a zoning ordinance affects the use which .a landowner may make of his property.” Sand-alow, The Limits of Municipal Power Under Home Rule: A Fióle for the Courts, 48 Minn. L. Rev. 643, 676. Therefore, we hold that the zoning power is one of a city’s or town’s independent municipal powers included in art. 89, § 6’s broad grant of powers to adopt ordinances or by-laws for the protection of the public health, safety, and general welfare.

However, our holding that local zoning ordinances or by-laws flow from the municipalities’ independent municipal powers does not mean that c. 774 unconstitutionally encroached upon home rule powers. “Plainly the Legislature, under art. 89, may forbid the enactment or control the form of such by-laws in some manner, without regard to whether such a by-law . . . comes within § 7 (5).” The Marshal House case, supra, at 717. Section 6’s broad grant of home rule powers contains a provision which limits the exercise of these powers to acts “not inconsistent with the constitution or laws enacted by the *360general court in conformity with powers reserved to the general court by section eight.” Section 8 provides that “[t]he general court shall have the power to act in relation to cities and towns, but only by general laws which apply alike to all cities, ... or to all cities and towns, or to a class of not fewer than two, and by special laws enacted . . ..”

Thus, municipalities can pass zoning ordinances or bylaws as an exercise of their independent police powers but these powers cannot be exercised in a manner which frustrates the purpose or implementation of a general or special law enacted by the Legislature in accordance with § 8’s provisions. The adoption of the Home Rule Amendment has not altered the Legislature’s supreme power in zoning matters as long as the Legislature acts in accordance with § 8. Chapter 774 is a proper exercise of the powers reserved to the Legislature by § 8 because it is a general law which applies to two or more municipalities. It is in addition to the Legislature’s continuing exercise of such reserve power in its numerous amendments to G. L. c. 40A, which is the basic enabling statute for zoning by-laws and ordinances. Therefore, the statute’s grant of power to boards of appeals and to the committee to override local zoning ordinances or by-laws which would otherwise frustrate the statute’s objective of providing for the critical regional need for low and moderate income housing does not violate the Home Rule Amendment.

3. Does the Boards’ or Committee’s Exercise of their Power to Override Local Zoning Regulations Deemed Inconsistent with Local Needs Constitute Spot Zoning?

The amici curiae argue that when the committee exercises its power to override local zoning by-laws deemed inconsistent with local needs, it engages in illegal spot zoning because its act singles out a particular parcel of property in a community for treatment different from that given to similar surrounding land indistinguishable from it in character.

*361However, we have frequently held that a spot zoning violation involves more than a mere finding that a parcel of property is singled out for less restrictive treatment than that of surrounding land of a similar character. If we accepted the amici curiae’s test for spot zoning, the State could never exercise its undoubted power to override local zoning ordinances or by-laws for legitimate public purposes because there would always be a disparity of treatment between the area where the local ordinance or by-law is ignored and the rest of that zoning area where it is enforced. Moreover, zoning variances and amendments would also be subject to serious challenge on spot zoning grounds if the test were merely one of dissimilar treatment of similar parcels of property.

We stated the definition of spot zoning in Lamarre v. Commissioner of Pub. Works of Fall River, 324 Mass. 542, a case concerning a zoning amendment which converted a parcel of land that had previously been zoned as a single and general (not more than three family housing) residence district into a multi-family residence district. The local housing authority had requested the amendment and it had been approved by the local planning board and the city council because of a shortage of rental housing in Fall River. The facts of the Lamarre case are closely analogous to those in the instant case. In both cases, a particular parcel of property was singled out for treatment different from that given to surrounding property of a similar character. The only difference is that instead of the amendment of the local zoning regulation by municipal authorities as in the Lamarre case, c. 774 enables the board of appeals and the committee to avoid the effect of the applicable zoning restrictions when they are inconsistent with local needs, a difference which has no bearing on whether either decision constitutes spot zoning.

We held in the Lamarre case at 545-546, that “it [the zoning amendment] was not an instance of ‘spot’ zoning. Tt does not appear that there was “a singling out of one lot for different treatment from that accorded to similar *362surrounding land indistinguishable from it in character, all for the economic benefit of the owner of that lot.” ’ Marblehead v. Rosenthal, 316 Mass. 124, 126. Whitte-more v. Building Inspector of Falmouth, 313 Mass. 248, 249. See Smith v. Board of Appeals of Salem, 313 Mass. 622” (emphasis supplied). Our definition of spot zoning is in accord with the general view held by most State courts that spot zoning problems arise where a zoning change is designed solely for the economic benefit of the owner of the property receiving special treatment and is not in accordance with a well considered plan for the public welfare.15 See Anderson, Am. Law of Zoning, § 5.04 and cases cited therein; Jones v. Zoning Bd. of Adjustment of Long Beach, 32 N. J. Super. 397.

Thus, the central question posed by this spot zoning challenge is whether the difference of treatment accorded by c. 774 serves the public welfare or merely affords an economic benefit to the owner of the land receiving special treatment. We decided in the Lamarre case that zoning changes affording special treatment to encourage the construction of multi-family residences in cities with housing shortages promote the public welfare. P. 546. We followed this decision in Henze v. Building Inspector of Lawrence, 359 Mass. 754, where we held that the need for low and moderate income housing justified a zoning reclassification of a parcel of land from a one and two-family residential district to a multi-family residential district.16 We think these decisions are dispositive of *363the spot zoning challenge. Chapter 774 reflects the Legislature’s judgment that the special treatment accorded to a site proposed for the construction of low and moderate income housing and necessitated by exclusionary zoning practices serves the general welfare by promoting the construction of badly needed housing units in the suburbs. The statute’s “consistent with local needs” standard and its provisions for judicial review of the board’s and committee’s decisions insure that special treatment will be allowed only when it serves the public interest. Thus, we hold that the exercise of c. 774’s power to override local zoning by-laws and ordinances deemed inconsistent with local needs does not constitute spot zoning.

4. Is c. 774 Unconstitutionally Vague?

The boards contend that c. 774 improperly delegates legislative authority to an administrative agency and is void for vagueness because its provisions fail to provide standards sufficient to guide administrative action and to limit the exercise of untrammeled discretion. See Smith v. Board of Appeals of Fall River, 319 Mass. 341, 344. These constitutional claims of “void for vagueness” and unlawful delegation of legislative authority are closely related. The principal question posed by both claims is whether the statute is so vague "that men of common intelligence must necessarily guess at its meaning and differ as to its application.” Commonwealth v. Carpenter, 325 Mass. 519, 521. O’Connell v. Brockton Bd. of Appeals, 344 Mass. 208, 212. Such vagueness would permit “untrammeled [administrative] discretion” (MacGibbon v. Board of Appeals of Duxbury, 356 Mass. 635, 638) and arbitrary and capricious decisions in violation of the due process clause of the Fourteenth Amendment to the Constitution of the United States and of art. *36410 of the Massachusetts Declaration of Rights. See Connolly v. General Constr. Co. 269 U. S. 385, 391; Ashton v. Kentucky, 384 U. S. 195, 200; Opinion of the Justices, 358 Mass. 827, 829; Opinion of the Justices, 341 Mass. 760, 776.

The Concord board claims that “[t]he constitutional defect in G. L. c. 40B, § § 20-23, is not that the standards provided local boards are not specific enough, but that the statute provides no standards at all for use by such boards in passing on applications for comprehensive permits under G. L. c. 40B, § 21.” The boards’ claim of lack of standards to guide their decisions rests on their assumption that the “consistent with local needs” and “uneconomic” standards referred to in § 23 and defined in §20 apply only to the committee’s review of the boards’ decisions. We cannot agree.

We must construe the statute, “if fairly possible, so as to avoid not only the conclusion that it is unconstitutional but also grave doubts upon that score. United States v. Delaware & Hudson Co., 213 U. S. 366, 408.” United States v. Jin Fuey Moy, 241 U. S. 394, 401, cited in Worcester County Natl. Bank v. Commissioner of Banks, 340 Mass. 695, 701. Our construction of the statute must be made “upon the whole statute, [so] that no clause, sentence or word shall prove superfluous, void or insignificant, if, by any other construction they may all be made useful and pertinent.” Commonwealth v. McCaughey, 9 Gray 296, 297. Beloin v. Bullett, 310 Mass. 206, 210.

Our examination of c. 774 leads us to conclude that the standards to be applied by boards of appeals in deciding whether to issue comprehensive permits are the same as those to be applied by the committee in reviewing the boards’ decisions, namely, whether the grant of the permit is “reasonable and consistent with local needs” and whether any conditions imposed on the permit are “uneconomic,” It is irrelevant that these standards are set forth in sections (§ 23 and § 20) different from the sec-*365tian dealing expressly with the hearing before the board of appeals (§21) because §§ 20-23 must be construed as a whole. See Massachusetts Medical Serv. v. Commissioner of Ins. 344 Mass. 335, 338-339; Commonwealth v. McCaughey, supra, at p. 297. If the board’s decision does not comply with these standards, its decision will be reversed by the committee on appeal, so by necessary implication its own decision must be based on the same standards. Any other construction of the statute would render the hearing before the board a useless procedure.

Finally, there is no merit to the claim that these standards are impermissibly vague. Section 20 contains detailed definitions of the factors to be considered by both the board of appeals and the committee in determining whether a proposal is “consistent with local needs” or whether conditions which might be imposed would render the construction or operation of the project “uneconomic.” The “consistent with local needs” standard requires both the board and the committee to balance the regional need for low and moderate income housing against any objection to the details of the proposed plan. This standard requires the local board to consider the “regional need for low and moderate income housing . . . with the number of low income persons in the city or town affected” and with the local need to “protect the health or safety of the occupants of the proposed housing or of the residents of the city or town, to promote better site and building design in relation to the surroundings, or to preserve open spaces.” § 20. In weighing these local concerns, the board must apply municipal requirements and regulations “as equally as possible to both subsidized and unsubsidized housing” applications. § 20.

Moreover, § 20 gives the board of appeals specific guidelines to aid its determination of whether the regional need for low and moderate income housing requires the board to override local “requirements and regulations,” including restrictive zoning ordinances or *366by-laws, which would prevent the board from approving a proposal that is otherwise “reasonable17 and consistent with local needs.” Section 20 provides that the board need not override local “requirements and regulations” where (1) “low or moderate income housing exists which is in excess of ten per cent of the housing units reported in the latest decennial census” or (2) such housing exists “on sites comprising one and one half per cent or more of the total land area zoned for residential, commercial or industrial use” (excluding public land) or (3) “the application before the board would result in the commencement of construction of such housing on sites comprising more than three tenths of one percent of such land area or ten acres, whichever is larger, in any one calendar year.” § 20.

The value of these three specific alternative definitions of when local “requirements and regulations shall be considered consistent with local needs” is that they define precisely the municipality’s minimum housing obligations “under the statute and permit it to do some intelligent, long-range planning about how and where the necessary housing should be built.” Rodgers,18 Snob Zoning in Massachusetts, 1970 Ann. Surv. Mass. Law, pp. 487, 490. These precise guidelines set forth in § 20 place a ceiling on the extent to which a local board must override local requirements and regulations, including exclusionary zoning laws, where the board decides that the application is reasonable and consistent with local needs.

Our construction of c. 774 does not mean that the board must automatically grant comprehensive permits in all cases where the community has not met its minimum *367housing obligation as it is specifically defined in § 20. The statute merely prevents the board from relying on local requirements or regulations, including applicable zoning by-laws and ordinances which prevent the use of the site for low and moderate income housing, as the reason for the board’s denial of the permit or its grant with uneconomic conditions.

In cases where the locality has not met its minimum housing obligations, the board must rest its decision on whether the required need for low and moderate income housing outweighs the valid planning objections to the details of the proposal such as health, site design, and open spaces (see § 20). If the regional need for such housing outweighs these objections, the board must override any restrictive local requirements and regulations which prevent the construction of the housing and grant the comprehensive permit. However, the municipality’s failure to meet its minimum housing obligations, as defined in § 20, will provide compelling evidence that the regional need for housing does in fact outweigh the objections to the proposal. See 7 Harv. J. on Legislation, 246.

However, once the municipality has satisfied its minimum housing obligation, the statute deems local “requirements and regulations,” including its restrictive zoning ordinances or by-laws, as “consistent with local needs” and thereby enforceable by the board if it wants to apply them. In this situation, only the board retains the power to override these requirements and regulations in order to grant a comprehensive permit. This result reflects the Legislature’s desire to preserve local autonomy once the community has satisfied its minimum obligation.19

*368We have upheld the constitutionality of statutes delegating broader authority to an administrative agency and creating less definite standards than those provided by § 20.20 See Burnham v. Board of Appeals of Gloucester, 333 Mass. 114; MacGibbon v. Board of Appeals of Duxbury, 356 Mass. 635. These zoning cases support our conclusion that since the standards contained in c. 774 are more than adequate, the statute is neither void for vagueness nor an unlawful delegation of legislative authority.

5. Does c. 77U Require a De Novo Hearing before the Committee?

Having determined the scope and the constitutionality of c. 774, we now discuss the questions raised by the boards concerning the procedural implementation of the statute. The Concord and the Hanover boards both question the scope of the committee’s power to review the boards’ decisions. The Concord board argues that the *369committee’s review of its decision permits only “a reexamination of . . . [the] proceeding for the purpose of preventing a result which may not be based on the exercise of an unbiased and reasonable judgment.” The Hanover board argues that the committee is empowered to decide only whether, on the evidence presented to the board, its decision was reasonable. Both boards argue that the statute does not provide for a de nova hearing before the committee.

The boards point out that the statute’s language provides that when an application for a comprehensive permit is denied by the board, the applicant may “appeal” to the committee for “review” of the decision. From the use of these words in the statute the boards argue that a de nova hearing before the committee was not intended. We cannot agree. The boards’ argument would be more persuasive if these were the only words in the statute relating to the type of hearing to be held before the committee. The statute, however, indicates that the committee must conduct a de nova review of the application.

When a board of appeals has denied an application for a permit, c. 774 requires the committee to determine whether the board’s decision was “reasonable and consistent with local needs.” § 23. The statute provides that the committee “shall render a written decision . . . stating its findings of fact, its conclusions and the reasons therefor . . ..” § 22. We believe that such language provides a clear indication that the Legislature intended that there should be a new evidentiary hearing before the committee at which the parties present evidence relevant to the issues to be decided. Rodgers, Snob Zoning in Massachusetts, 1970 Ann. Surv. Mass. Law, 487, 491. Chapter 774 also requires a stenographic record to be kept of the committee’s hearing but makes no provision for such a record to be kept of the hearing before the board. § 22. This is further indication that the Legislature intended de nova review by the committee.

When making its decision, the committee has before it *370a copy of the board’s decision and a statement of the reasons upon which the decision was based. § 22. Chapter 774 does not indicate whether the committee is to give any weight to the board’s decision or stated reasons. In view of the statute’s requirement that the committee must find the facts, draw conclusions and state its own reasons for its decision, we believe that, although the board’s decision and reasons (including any voluntary findings of fact made by the board) are competent evidence to enable the committee to ascertain the conclusions the board reached, the committee must make its own decision on the basis of a new evidentiary hearing.

The legal issues properly before the committee are circumscribed by c. 774. When the board has denied an application for a comprehensive permit, the committee is required to determine whether the board’s decision was “reasonable and consistent with local needs.” § 23. The phrase “consistent with local needs” is defined by § 20 in terms of factors that must be weighed in reaching a decision on the issuance of a permit. But the fact that the issues before the committee are clearly defined does not mean that the committee should not hear evidence related to those issues.

In reaching its decision the committee has before it the board’s decision and its reasons for the denial of the application. It is quite possible that the board may not have included vital considerations in its list of stated reasons for its denial of the application either because they were not before the board or because they were considered irrelevant. If the committee is to fulfil its duty to determine whether the board’s decision is in fact “reasonable and consistent with local needs” then the committee must be free to consider any evidence relevant to those issues.

The argument that a de nova hearing before the committee would permit the applicant to make a purely pro forma presentation before the board is not persuasive in light of our holding that the board has the power to grant applications for comprehensive permits, despite *371contrary zoning laws. An applicant would be unlikely to choose an approach that would require two presentations, with added costs and delays, where one presentation could suffice. Furthermore, it is not likely that an applicant would risk alienating the board or the committee by such obvious tactics.

Although the committee stated that its hearings were not de nova, the record is clear that the committee did in fact base its decisions on new evidentiary hearings. The fact that the committee did not label its hearings de nova is not significant because c. 774 clearly requires de nova hearings before the committee.

6. Are the Alternative Methods of Review Afforded an Applicant Who Has Been Denied a Permit and an Aggrieved Person Appealing from the Grant of a Permit Violative of the Equal Protection of the Laws Guaranteed by the Federal and Massachusetts Constitutions?

The boards argue that the statute’s provisions for alternative methods of review, depending upon whether the application has been granted or denied, violates the equal protection of the laws guaranteed by the Federal and the State Constitutions.21 The thrust of these arguments was premised upon the view that the hearing before the committee was restricted to an inquiry into whether the board’s decision was based upon “substantial evidence” (cf. G. L. c. 30A, §§ 1[6] and 14[8][e]) whereas the hearing before the court afforded an aggrieved person appealing from the grant of a permit is essentially a de nova review. G. L. c. 40B, § 21. G. L. c. 40A, § 21. See Green v. Board of Appeal of Norwood, 358 Mass. 253, 256. Since we hold that the statute requires the committee to hear any evidence relevant to the issue whether the denial of the application was reasonable and consistent with local needs, there are no substantial differences between the alternative methods of review.

*3727. If the-Committee Decides that the Boards’ Decisions Are Incorrect, Must the Committee Remand the Cases to the Boards?

The boards argue that if the committee decides that a board’s decision is incorrect, then the committee must remand the case to the board so that the board may have an opportunity to correct its errors. Such a procedure would, however, be contrary to the plain language of the statute. Section 23 is explicit in its directives to the committee: “If the committee finds, in the case of a denial, that the decision of the board of appeals was unreasonable and not consistent with local needs, it shall vacate such decision and shall direct the board to issue a comprehensive permit or approval to the applicant.” Any notion of a remand procedure (other than ordering the board to carry out the committee’s decision) is precluded by the clear language of the statute. The absence of a remand procedure comports with the time limit fixed for each stage of the hearing process which together indicate the Legislature’s intent to speed up the permit procedure and to facilitate the building of low and moderate income housing.

8. Does the Committee Have the Power to Order the Issuance of Permits Subject to' the Conditions as Specified?

The Hanover board argues that the committee exceeded its authority in ordering the issuance of a permit with conditions that (a) stated that the drainage and sewage disposal systems must be approved by the appropriate State authorities, (b) empowered the board, at its option, to require the applicant to make a full disclosure concerning its leasing arrangements, and (c) set no time limits either for the fulfilment of the four specified conditions22 or for the commencement or com*373pletion of the project. The board maintains that such a conditional permit is impermissible because it constitutes an advisory opinion and only partially commits the committee to the issuance of the permit.

It is clear that the committee may order the issuance of conditional permits. Chapter 774 gives a board the power to issue permits or approvals “including but not limited to the power to attach to said permit or approval conditions and requirements with respect to height, site plan, size or shape, or building materials as are consistent with the terms of this section.” G. L. c. 40B, § 21. When a board has denied an application, the committee is empowered to vacate the board’s decision *374if it is unreasonable and not consistent with local needs and to direct the board to issue a comprehensive permit or approval to the applicant. G. L. c. 40B, § 23. Since the committee is issuing an order for the board to act pursuant to the board’s own power, and since the board is specifically authorized to issue conditional permits or approvals, it follows that the committee has the power to order the board to issue conditional permits or approvals.

To support its contention that the committee may not order the issuance of a permit subject to the conditions mentioned, the Hanover board has misplaced reliance upon Weld v. Board of Appeals of Gloucester, 345 Mass. 376, 378. In the Weld case, we held that the decision of the board of appeals granting a special permit to build a hotel subject to the condition that the “water situation must be arranged to the satisfaction of all concerned” was in excess of its authority to issue conditional per*375mits, as defined by the enabling statute and the city zoning ordinance passed pursuant thereto. The objection to such a condition was that it invoked undefined standards and necessarily implied that the board would have to make a further determination to redefine the standard required.

In that case we stated that “[w]e assume that the board may condition the right to operate under a permit presently issued upon the completion of proposed work in accordance with identified plans or other certain standards” (emphasis supplied). We did not find the condition that “[t]he sewage disposal system shall be completed in accordance with plans previously filed, subject to the approval of the local and state boards of health,” impermissible because the standards specified were sufficiently definite.

The two conditions objected to in the instant case set forth clear and certain standards by referring to “appropriate state authorities” with respect to the approval of the drainage and sewage systems, and “the Department of Community Affairs” with respect to the sufficiency of the disclosure of the leasing arrangement. These conditions are set forth in precise terms, with definite standards, and in no way constitute an advisory opinion.

The time within which the four conditions must be fulfilled is either stated directly (e.g., the first condition must be fulfilled “[bjefore beginning construction”) or is implicitly an on-going process coinciding with the progress of the construction (e.g., the fourth condition provides that “local officials shall carry out compliance inspections in the usual manner”). The completion of the project necessarily depends upon numerous factors, including the outcome of this litigation. In the circumstances of this case it is within the committee’s discretion to refuse to set final dates for the completion of the project.

9. Are the Committee’s Decisions Supported by “Substantial Evidence’’ ?

In reviewing the committee’s decisions, the court, *376whether the Superior Court in the usual case or this court to which the matter was reported, must decide whether there was “substantial evidence” to support those decisions, that is, “such evidence as a reasonable mind might accept as adequate to support a conclusion.” G. L. c. 30A, §§ 1 (6) and 14 (8) (e). Boston Edison Co. v. Selectmen of Concord, 355 Mass. 79, 92. Labor Relations Commn. v. University Hosp. Inc. 359 Mass. 516, 521. Jaffe, Judicial Control of Administrative Action, 596. The reviewing court must examine the committee’s decisions “upon consideration of the entire record.” G. L. c. 30A, § 14 (8). This statutory directive has been construed to require taking into account evidence that detracts from the weight of the supportive evidence. Cohen v. Board of Registration in Pharmacy, 350 Mass. 246, 253, citing by analogy Universal Camera Corp. v. National Labor Relations Bd. 340 U. S. 474, 488. The boards have argued that the committee’s decisions are not supported by substantial evidence. However, we disagree.

The Hanover Proceeding before the Committee.

In deciding that the Hanover applicant should be granted a conditional permit the committee found that the board’s, denial of the application was unreasonable and not consistent with local needs. The committee made subsidiary findings of fact relating to the requirements of the statute and to the board’s enunciated reasons for its decisions, all of which amply support the committee’s decision.

The applicant sought a permit to build eighty-eight units of low and moderate income housing for elderly persons. The need for such housing was not disputed by thé board. The committee had before it an exhibit prepared by the Metropolitan Area Planning Council referred to as “Maximum Housing and Annual Land Area Guidelines under Chapter 774 for Hanover, Massachusetts.” This report indicated that there is a current need for low and moderate income housing in Hanover and that there is no such housing in the town. A town meet*377ing report showed that over 100 elderly Hanover residents desire and are eligible for such housing, and a 1969 supplement to the town plan projects a steady increase in elderly persons residing in Hanover. This, we believe, presented substantial evidence to support the committee’s conclusion that the town of Hanover had not met its minimum housing obligations as defined by G. L. c. 40B, § 20.

The committee found that the applicant (Country Village Corporation) had sufficient property interest in the site to qualify as an applicant under c. 774.23 The evidence before the committee revealed that the site is owned by George Oulton, president and treasurer of the applicant corporation, and his wife Anne R. Oulton, a director of the corporation. The Oultons had leased the land to the applicant under a fifty-five year term, to take effect on April 10, 1970, with renewal options for a further twenty years. Notice of this lease was filed with the register of deeds for Plymouth County on June 9, 1970.

The statute does not explicitly state the requisite property interest necessary to qualify as an applicant for a comprehensive permit. Section 21, however, specifies that “[a]ny public agency or limited dividend or nonprofit organization proposing to build low or moderate income housing may submit to the board of appeals . . . a single application to build such housing . . ..” Low or moderate income housing is defined as “any housing subsidized by the federal or state government under any program to assist the construction of low or moderate income housing as defined in the applicable federal or state statute . . ..” G. L. c. 40B, §20.

In the instant case the Hanover applicant applied for funding to the Massachusetts Housing Finance Agency (established pursuant to St. 1966, c. 708), an appropriate State, funding authority under c. 774. The committee had before it a letter from the Massachusetts Housing *378Finance Agency stating that “if the local approvals are given, or approval is secured through Chapter 774 of the Acts of 1969, a loan probably would be issued.”

We believe that the statute’s use of standards set by State or Federal funding agencies to define “low or moderate income housing” indicates that the Legislature intended to define the requisite property interest for a permit in terms of the selected financing agency’s property interest requirements. Therefore, we believe that c. 774 does not require the applicant for a comprehensive permit to establish before the board or committee a present title in the proposed site.24 However, such an interpretation does not prevent the board or the committee from requiring the applicant to disclose fully any property interest presently held in the site or plans to acquire any such interest.25 The committee and the board have the power to order the issuance of a permit subject to the applicant’s receiving financing from appropriate State or Federal programs. The applicant must, of course, satisfy any property interest requirements of the selected financing program before the funding issues and before the conditional permit becomes operative.

The committee was, therefore, warranted in finding that the Hanover applicant had sufficient property interest in the site.

Chapter 774 provides that an application for a comprehensive permit may be submitted only by a “public agency or limited dividend or nonprofit organization.” The committee found that the Hanover applicant was a limited dividend organization within the meaning of the *379statute. The rules and regulations for the business of the committee define a “limited dividend organization” as “any applicant which (a) proposes to sponsor housing under Chapter 40B, and (b) is not a public agency and (c) is eligible to receive a subsidy from a State or Federal Agency after a comprehensive permit has been issued.” (Rules and Regulations of Housing Appeals Committee, 1 [f].) The Hanover applicant has set forth in its articles of organization the fact that the corporation is organized for the specific purpose of constructing low and moderate income homes and that the corporation may not make any distributions to stockholders, in any year, in excess of the amounts permitted for eligible funding applicants under the Massachusetts Housing Finance Agency’s program. St. 1966, c. 708, § 5 (d). This provision was inserted in order to insure eligibility for State (and Federal) funding programs.

Testimony from the board chairman before the committee indicated that the board was prepared to leave the question of eligibility as a limited dividend organization to the appropriate State agencies. We believe that the board’s interpretation of c. 774 (which is also enunciated in the regulations for the business of the committee) is in accordance with the intent of the statute. The eligibility of the applicant in this respect turns on the applicant’s ability to qualify for the appropriate funding. To so qualify the applicant must propose to build “low or moderate income housing.” Since the phrase “low or moderate income housing” is defined in c. 774 in terms of standards set by the selected subsidizing agency, we believe that the question of standards for eligibility as a limited dividend organization is properly left to the appropriate State or Federal funding agency. Such an interpretation does not prevent the board or the committee from requiring full disclosure of the organization’s legal status and further requiring compliance with pertinent statutory and regulatory provisions.

Since the committee had before it a letter from the Massachusetts Housing Finance Agency stating that it *380considered the applicant to be a limited dividend organization eligible for agency financing and since the applicant is proposing to build housing under c. 40B and is not a public agency, the committee was justified in finding that the applicant was a limited dividend corporation within the meaning of c. 774,

The Hanover board argues that before the committee can order the board to grant a comprehensive permit or approval, it must find “that there was a reasonable likelihood that the purported project would be subsidized by a federal or state agency.” Section 21 of G. L. c. 40B states that the applicant shall be “proposing to build low or moderate income housing.” The regulations for the business of the committee make it clear that such an applicant must be eligible to receive the appropriate funding. The letter submitted to the committee from the Massachusetts Housing Finance Agency stated in effect that the applicant was eligible for agency financing and that “if the local approvals are given, or approval is secured through Chapter 774 of the Acts of 1969, a loan probably would be issued.” In view of this evidence the committee was warranted in concluding that the Hanover applicant was “proposing to build low or moderate income housing” within the meaning of c. 774.

Chapter 774 provides that one of the considerations in determining whether an applicant’s proposal is “consistent with local needs” is “the need to protect the health or safety of the occupants of the proposed housing or of the residents of the city or town, to promote better site and building design in relation to the surroundings . . ..” G. L. c. 40B, § 20. The committee found that the applicant’s plans were sufficient to satisfy these requirements. This finding was amply supported by the testimony. There was expert evidence that the increase in daily traffic generated by the project would be minimal and always far below the capacity of the surrounding streets. The plans submitted showed a parking space provided for every housing unit even though studies in similar projects in Boston suburbs showed that a ratio of one *381parking space per 2.4 housing units would have provided adequate space for all cars likely to be owned by the project’s inhabitants. The evidence before the committee was sufficient to warrant the conclusion that the proposed parking facilities in the applicant’s project met the requirements of c. 774.

The committee’s conclusion that the driveway plans were adequate was based upon plans showing a private way, twenty feet in width, running from Oakland Avenue to the project and terminating in a turn-around parking area, with a separate set of foot paths connecting the units and leading out of the project. There was evidence that the average street in Hanover is between eighteen and thirty-two feet wide and there was testimony by Hanover’s police chief that Oakland Avenue, which also has a paved width of at least twenty feet, was adequate for emergency vehicles. The applicant’s landscape plans were also shown to be adequate. There was testimony that many trees would remain on the site and that the topography of the site requires only a minimum of grading.

The committee’s conclusion that the traffic, parking and landscape plans of the project met the requirements of c. 774 was thus supported by substantial evidence.

The committee found that the board’s demand that the applicant should submit a revised set of plans for drainage and sewage disposal was unreasonable in view of the fact that new plans would be extremely expensive and that there was considerable doubt whether a permit would be issued by the board even if new plans were forthcoming. Since the board could have issued a permit subject to the condition of tendering a suitable disposal plan 26 and since these plans had to comply with State standards, whatever their particular design, the committee’s decision that the board had unreasonably rejected the applicant’s original plans was warranted.

*382 The Concord Proceeding before the Committee.

In deciding that the Concord applicant should be granted a conditional permit the committee found that the Concord board’s decision to deny the application was unreasonable and not consistent with local needs. The committee made subsidiary findings of fact which relate to the requirements of the statute and the board’s enunciated reasons for its decision, all of which amply support the committee’s decision.

The parties to the proceeding before the committee stipulated that there is both a regional need for low and moderate income housing and a local need for such housing within the town of Concord. Such housing within the town comprises less than ten per .cent of the housing units according to the 1960 census' and uses less than one and one-half per cent of the land in the town zoned for residential, commercial or industrial uses. Under the present zoning by-law only twenty-nine acres or one quarter of one per cent of the land in Concord zoned for residential, commercial or industrial uses is available for low and moderate income housing. These stipulations provide adequate support for the committee’s conclusion that the town of Concord had not met its minimum housing obligations as defined by c. 774.

Both parties to the proceedings before the committee presented well qualified experts to testify to the efficacy of the applicant’s plans to deal with the water problem. It was not disputed that the project site and the surrounding area are beset with difficult water problems. The committee took a view of the site in August, 1971, and observed the soggy and water laden ground in the lower parts of the land. The applicant’s expert testified that the proposed measures to control the surface flooding and lower ground water levels would provide an adequate solution to the water problem.27 In fact there was *383evidence that the proposed plans would not only provide an adequate system for the residents of the project but might also improve the water problems of the surrounding area. We therefore conclude that there was substantial evidence before the committee that the applicant’s proposed plans were adequate.

We have reviewed the committee’s decisions on the Hanover and the Concord applications “upon consideration of the entire record” and conclude that both decisions were supported by substantial evidence.

10. In conclusion, c. 774 represents the Legislature’s attempt to satisfy the regional need for housing without stripping municipalities of their power to zone. By creating a “consistent with local needs” criterion which expands the scope of relevant local needs considered by the local boards to include the regional need for low and moderate income housing, the Legislature has given the boards the power to override the local exclusionary zoning practices in order to encourage the construction of such housing in the suburbs. By fixing a ceiling on the extent to which a board must override local zoning regulations, the Legislature has clearly delineated that point where local interests must yield to the general public need for housing. This ceiling establishes the minimum share of responsibility that each community must shoulder in order to alleviate the housing crisis that confronts the Commonwealth.

As we noted in Simon v. Needham, 311 Mass. 560, 565-566, the municipality’s power to control its character and land usages by its zoning regulations is not unlimited. “A zoning by-law cannot be adopted for the purpose of setting up a barrier against the influx of thrifty and respectable citizens who desire to live there and who *384are able and willing to erect homes upon lots upon which fair and reasonable restrictions have been imposed nor for the purpose of protecting the large estates that are already located in the district. The strictly local interests of the town must yield if it appears that they are plainly in conflict with the general interests of the public at large, and in such instances the interest of The municipality would not be allowed to stand in the way.’ Euclid v. Ambler Realty Co. 272 U. S. 365, 390. See Eubank v. Richmond, 226 U. S. 137, 144; Edwards v. California, 314 U. S. 160; Salisbury Land & Improvement Co. v. Commonwealth, 215 Mass. 371; Bradley v. Zoning Adjustment Board of Boston, 255 Mass. 160, 162; .. . State v. Stahlman, 81 W. Va. 335.”

The legislative reports which prompted c. 774’s passage demonstrated how local restrictive zoning regulations have set up, in fact if not intentionally, a barrier against the introduction of low and moderate income housing in the suburbs. Moreover, this barrier exists at a time when our housing needs for the low and moderate income groups cannot be met by the “inner cities.” This housing crisis demands a legislative and judicial approach28 *385that requires “the strictly local interests of the town” to yield to the regional need for the construction of low and moderate incoming housing. Chapter 774 represents the Legislature’s use of its own zoning powers to respond to this problem.

The Legislature’s zoning power may be used “where the interests of the public require such action and where the means employed are reasonably necessary for the accomplishment of the purpose.” Simon v. Needham, 311 Mass. 560, 562. Within these broad limits, the General Court is the sole judge as to how and when the power is to be exercised as long as it acts in accordance with the powers reserved to it by § 8 of the Home Rule Amendment. Our responsibility in examining the Legislature’s exercise of its zoning power is limited to the determination of whether the legislation adopts a reasonable means to serve a legitimate public purpose. Our analysis of c. 774’s legislative history and text leads us to conclude that the Legislature’s adoption of an administrative mechanism designed to supersede, when necessary, local restrictive requirements and regulations, including zoning by-laws and ordinances, in order to promote the construction of low and moderate income housing in cities and towns is a constitutionally valid exercise of the Legislature’s zoning power which was properly implemented in the proceedings before us.

In each case, a decree shall be entered in the Superior Court affirming the decision of the Housing Appeals Committee.

„ 7 7 So ordered.

The opinion of the Court was delivered by

WILENTZ, C.J.

This is the return, eight years later, of Southern Burlington County N.A.A.C.P. v. Township of Mount Laurel, 67 N.J. 151 (1975) (Mount Laurel I). We set forth in that case, for the first time, the doctrine requiring that municipalities’ land use regulations provide a realistic opportunity for low and moderate income housing. The doctrine has become famous. The Mount Laurel case itself threatens to become infamous. After all this time, ten years after the trial court’s initial order invalidating its zoning ordinance, Mount Laurel remains afflicted with a blatantly exclusionary ordinance. Papered over with studies, rationalized by hired experts, the ordinance at its core is true to nothing but Mount Laurel’s determination to exclude the poor. *199Mount Laurel is not alone; we believe that there is widespread non-compliance with the constitutional mandate of our original opinion in this case.

To the best of our ability, we shall not allow it to continue. This Court is more firmly committed to the original Mount Laurel doctrine than ever, and we are determined, within appropriate judicial bounds, to make it work. The obligation is to provide a realistic opportunity for housing, not litigation. We have learned from experience, however, that unless a strong judicial hand is used, Mount Laurel will not result in housing, but in paper, process, witnesses, trials and appeals. We intend by this decision to strengthen it, clarify it, and make it easier for public officials, including judges, to apply it.

This case is accompanied by five others, heard together and decided in this opinion.1 All involve questions arising from the *200 Mount Laurel doctrine. They demonstrate the need to put some steel into that doctrine. The deficiencies in its application range from uncertainty and inconsistency at the trial level to inflexible review criteria at the appellate level. The waste of judicial energy involved at every level is substantial and is matched only by the often needless expenditure of talent on the part of lawyers and experts. The length and complexity of trials is often outrageous, and the expense of litigation is so high that a real question develops whether the municipality can afford to defend or the plaintiffs can afford to sue.

There is another side to the story. We believe, both through the representations of counsel and from our own research and experience, that the doctrine has done some good, indeed, perhaps substantial good. We have tried to make the doctrine clearer for we believe that most municipal officials will in good faith strive to fulfill their constitutional duty. There are a number of municipalities around the State that have responded to our decisions by amending their zoning ordinances to provide realistic opportunities for the construction of low and moderate *201income housing.2 Further, many other municipalities have at least recognized their obligation to provide such opportunities in their ordinances and master plans. Finally, state and county government agencies have responded by preparing regional housing plans that help both the courts and municipalities themselves carry out the Mount Lam-el mandate. Still, we are far from where we had hoped to be and nowhere near where we should be with regard to the administration of the doctrine in our courts.

These six cases not only afford the opportunity for, but demonstrate the necessity of reexamining the Mount Laurel doctrine. We do so here. The doctrine is right but its administration has been ineffective.

A brief statement of the cases may be helpful at this point. Mount Laurel II results from the remand by this Court of the original Mount Laurel case. The municipality rezoned, purportedly pursuant to our instructions, a plenary trial was held, and the trial court found that the rezoning constituted a bona fide attempt by Mount Laurel to provide a realistic opportunity for the construction of its fair share of the regional lower income housing need. Reading our cases at that time (1978) as requiring no more, the trial court dismissed the complaint of the N.A.A.C.P. and other plaintiffs but granted relief in the form of a builder’s remedy, to a developer-intervenor who had attacked the total prohibition against mobile homes. Plaintiffs’ appeal of the trial court’s ruling sustaining the ordinance in all other respects was directly certified by this Court, as ultimately was defendant’s appeal from the grant of a builder’s remedy allowing construction of mobile homes. We reverse and remand to determine Mount Laurel’s fair share of the regional need and for further proceedings to revise its ordinance; we affirm the grant of the builder’s remedy.

*202In Caputo v. Township of Chester, two resident landowners of that Morris County township had long sought rezoning of their property to allow residential construction with densities greater than those previously permitted. After much negotiation the municipality rezoned but not to plaintiffs’ satisfaction. Plaintiffs therefore commenced an action challenging the validity of the ordinance, an action that ultimately was based upon our decision in Mount Laurel. The trial court held that the ordinance was invalid but refused to grant a builder’s remedy to the would-be-developer, who appealed. Chester apparently acceded to the court’s ruling, no appeal having been taken. We granted direct certification of the developer's appeal. The only issues before us in that matter are the propriety of the denial of a builder’s remedy and of the invalidation of a five acre minimum lot requirement in a single family zone. We affirm that denial, reverse the ruling as to the minimum lot requirement, and reverse and remand the cause for further proceedings limited to the issue of the satisfaction of Chester’s present indigenous need for lower income housing.

In Glenview Development Co. v. Franklin, again a developer sought both to invalidate the zoning ordinance and to obtain a builder’s remedy. The trial court held that this rural Hunterdon County township was not subject to the Mount Laurel obligation because it was deemed not to be a “developing” community, from which ruling the developer appealed. We certified the matter directly. We affirm the trial court’s ruling but reverse and remand the cause for further proceedings limited to the issue of the satisfaction of Franklin’s present indigenous need for lower income housing.

In Round Valley, Inc. v. Clinton, the usual two-pronged attack by the developer (declaration of invalidity and builder’s remedy) was successful at the trial level, where the court entered a judgment invalidating the ordinance and appointing a master to assure its appropriate revision and to assist in effectuating a builder’s remedy. On appeal, the Appellate Division held that the judgment below did not pass upon the validity of the *203ordinance but simply granted a builder’s remedy; that such action by a trial court was not authorized by the Mount Laurel doctrine, since it consisted of little more than the grant of a variance without complying with the statutory requirements. The Appellate Division also indicated that the sale of a portion of the developer’s tract between the trial and the appeal rendered most of the matters moot, since the portion sold was that designed by the developer for all of its multi-family units. The matter is before us on the developer’s appeal. We reverse the Appellate Division and remand for further proceedings to determine fair share and thereafter to revise the ordinance; in effect we sustain the trial court’s finding that the ordinance is invalid and its appointment of a master to aid in its revision, as well as its award of the builder’s remedy subject to the conditions set forth herein.

Urban League of Essex Co. v. Mahwah was the latest of numerous attempts to force the construction of housing for those who work in Mahwah in Bergen County. The trial court held that the Mahwah zoning ordinance complied with the Mount Laurel doctrine in that it allowed for the construction of “least cost” housing although admittedly no lower income housing could possibly be built there. Plaintiffs, being both those who live in and out of Mahwah and who work there, believing that the ordinance could more realistically provide an opportunity for the construction of lower income housing, appealed to the Appellate Division and we certified the matter directly. We reverse and remand for further proceedings concerning Mah-wah’s fair share and thereafter to revise the ordinance.

In Urban League of Greater New Brunswick v. Carteret, the zoning ordinances of all the municipalities in Middlesex County were initially challenged. The majority of the eases were settled during trial (through the revision of ordinances to comply with the trial court’s directives, or, in the cases of Perth Amboy, Carteret and Dunellen, through judgments of the trial court holding that no Mount Laurel obligation existed). Only seven municipalities appealed from the judgment of the trial court. *204As to those seven, the trial court held that their zoning ordinances violated the Mount Laurel doctrine. The court determined the regional need and then allocated to each municipality a sufficient part of the regional need so that, were it built, that municipality’s lower income housing would constitute the same percentage of its total housing as the lower income housing of the county constituted of its total housing. The balance of the regional need allocable to those seven municipalities was divided among them equally. On appeal the Appellate Division reversed, holding, among other things, that the trial court’s determination was improperly “formulaic” and that the formula was incorrect. There was no remand for proceedings in accordance with the Appellate Division’s opinion. The appeal to us is from that Appellate Division decision. We reverse the judgment of the Appellate Division but remand to the trial court for further proceedings concerning region, regional need, and fair share and thereafter for revision of the various ordinances.

This opinion is divided into three sections. Section I contains a brief history of the Mount Laurel doctrine with a discussion of the major implementation problems addressed in this opinion; a statement of the constitutional basis for the doctrine and the appropriate scope of the judicial power to enforce it; and a summary of the more significant rulings in today’s opinion. In Section II, we resolve the substantive issues raised by the six cases before us and set forth the obligations imposed upon municipalities and trial courts by the Mount Laurel doctrine. Finally, in Section III we apply these rulings to dispose of the six cases themselves.

I.

Background

A. History of the Mount Laurel Doctrine

In Mount Laurel I, this Court held that a zoning ordinance that contravened the general welfare was unconstitutional. We pointed out that a developing municipality violated that consti*205tutional mandate by excluding housing for lower income people; that it would satisfy that constitutional obligation by affirmatively affording a realistic opportunity for the construction of its fair share of the present and prospective regional need for low and moderate income housing. 67 N.J. at 174.3 This is the core of the Mount Laurel doctrine. Although the Court set forth important guidelines for implementing the doctrine, their application to particular cases was complex, and the resolution of many questions left uncertain. Was it a “developing” municipality? What was the “region,” and how was it to be determined? How was the “fair share” to be calculated within that region? Precisely what must that municipality do to “affirmatively afford” an opportunity for the construction of lower income housing? Other questions were similarly troublesome. When should a court order the granting of a building permit (i.e., a builder’s remedy) to a plaintiff-developer who has successfully challenged a zoning ordinance on Mount Laurel grounds? How should courts deal with the complicated procedural aspects of Mount Laurel litigation, such as the appointment of experts and masters, the joinder of defendant municipalities, and the problem of interlocutory appeals? These have been the principal questions that New Jersey courts have faced *206in attempting to implement the Mount Laurel mandate, and the principal questions dealt with in this opinion. We begin by examining how some of these questions have been dealt with up to now.

Two years after Mount Laurel I, in Oakwood at Madison, Inc. v. Township of Madison, 72 N.J. 481 (1977), this Court once again faced the exclusionary zoning issue. We ruled that “fair share” allocations need not be “precise” or based on “specific formulae” to win judicial approval. Id. at 498-99. Instead, the Court explained, a court should look to the “substance” of a challenged zoning ordinance and the “bona fide efforts” of a municipality to remove exclusionary barriers in order to determine whether that municipality had met its Mount Laurel burden.

With regard to the definition of the “region” from which fair share allocations were to be made, the majority cited with approval the trial court’s formulation of a region as the “ ‘area from which, in view of available employment and transportation, the population of the township would be drawn, absent invalidly exclusionary zoning.’” Id. at 537; quoting Oakwood at Madison, Inc. v. Township of Madison, 128 N.J.Super. 438, 441 (Law Div.1974). We distinguished this very general standard for determining region from the situation with which we would be confronted if a state planning body promulgated a plan that divided the whole state into regions.

Madison also addressed the nature of a municipality’s “affirmative” duty to encourage the construction of lower income housing. The Court reaffirmed that affected municipalities must provide realistic opportunities for their fair share of lower income housing, and required the municipality to provide density bonuses for the construction of multi-bedroom units, while reserving judgment, however, on other affirmative measures. 72 N.J. at 517-18 (a density bonus allows the developer to build more units per acre if certain kinds of units áre included in the project).

*207An important aspect of the Court’s decision was the award of a builder’s remedy to the plaintiff-developer. The Court emphasized that the plaintiff, for “six years” and through “two trials and on this extended appeal,” had “borne the stress and expense of this public interest litigation.” Id. at 549-50. The Court admonished, however, that this kind of remedy should “ordinarily be rare.” Id. at 551-52 n. 50.

Finally, the Court introduced the important concept of “least cost” housing, i.e., housing built at the least cost possible, even though not inexpensive enough for lower income occupancy. Recognizing that even with subsidies and affirmative devices some municipalities simply might not be able to provide a realistic opportunity for the construction of lower income housing, the Court held that under those and only those circumstances such municipalities could meet their obligation with “least cost” housing. Id. at 512-13.

Later in the same year that Madison was decided, the Court determined which municipalities were subject to the Mount Laurel fair share obligation. Pascack Ass’n, Ltd. v. Washington Twp., 74 N.J. 470 (1977); Fobe Associates v. Demarest, 74 N.J. 519 (1977). In Pascack, the Court held that “fully developed, single-family residential” communities such as Washington Township did not have any Mount Laurel obligation. This holding was reaffirmed in Fobe where the Court upheld the decision of the Demarest Board of Adjustment denying a variance sought for multi-family housing given the fact that a “developed” municipality like Demarest did not have a Mount Laurel obligation.4

*208B. Constitutional Basis for Mount Laurel and the Judicial Role

The constitutional basis for the Mount Laurel doctrine remains the same. The constitutional power to zone, delegated to the municipalities subject to legislation, is but one portion of the police power and, as such, must be exercised for the general welfare. When the exercise of that power by a municipality affects something as fundamental as housing, the general welfare includes more than the welfare of that municipality and its citizens: it also includes the general welfare — in this case the housing needs — of those residing outside of the municipality but within the region that contributes to the housing demand within the municipality. Municipal land use regulations that conflict with the general welfare thus defined abuse the police power and are unconstitutional. In particular, those regulations that do not provide the requisite opportunity for a fair share of the region’s need for low and moderate income housing conflict with *209the general welfare and violate the state constitutional requirements of substantive due process and equal protection. Mount Laurel I, 67 N.J. at 174 and 181.

That is the constitutional rationale for the Mount Laurel doctrine. The doctrine is a corollary of the constitutional obligation to zone only in furtherance of the general welfare. The doctrine provides a method of satisfying that obligation when the zoning in question affects housing.

It would be useful to remind ourselves that the doctrine does not arise from some theoretical analysis of our Constitution, but rather from underlying concepts of fundamental fairness in the exercise of governmental power. The basis for the constitutional obligation is simple: the State controls the use of land, all of the land. In exercising that control it cannot favor rich over poor. It cannot legislatively set aside dilapidated housing in urban ghettos for the poor and decent housing elsewhere for everyone else. The government that controls this land represents everyone. While the State may not have the ability to eliminate poverty, it cannot use that condition as the basis for imposing further disadvantages. And the same applies to the municipality, to which this control over land has been constitutionally delegated.

The clarity of the constitutional obligation is seen most simply by imagining what this state could be like were this claim never to be recognized and enforced: poor people forever zoned out of substantial areas of the state, not because housing could not be built for them but because they are not wanted; poor people forced to live in urban slums forever not because suburbia, developing rural areas, fully developed residential sections, seashore resorts, and other attractive locations could not accommodate them, but simply because they are not wanted. It is a vision not only at variance with the requirement that the zoning power be used for the general welfare but with all concepts of *210fundamental fairness and decency that underpin many constitutional obligations.5

*211Subject to the clear obligation to preserve open space and prime agricultural land, a builder in New Jersey who finds it economically feasible to provide decent housing for lower income groups will no longer find it govemmentally impossible. Builders may not be able to build just where they want — our parks, farms, and conservation areas are not a land bank for housing speculators. But if sound planning of an area allows the rich and middle class to live there, it must also realistically and practically allow the poor. And if the area will accommodate factories, it must also find space for workers." The specific location of such housing will of course continue to depend on sound municipal land use planning.

While Mount Laurel I discussed the need for “an appropriate variety and choice of housing,” 67 N.J. 179, the specific constitutional obligation addressed there, as well as in our opinion here, is that relating to low and moderate income housing. Id. All that we say here concerns that category alone; the doctrine as we interpret it has no present applicability to other kinds of housing. See Pascack, 74 N.J. at 480. It is obvious that eight years after Mount Laurel I the need for satisfaction of this *212doctrine is greater than ever. Upper and middle income groups may search with increasing difficulty for housing within their means; for low and moderate income people, there is nothing to search for.6

No one has challenged the Mount Laurel doctrine on these appeals. Nevertheless, a brief reminder of the judicial role in this sensitive area is appropriate, since powerful reasons suggest, and we agree, that the matter is better left to the Legislature. We act first and foremost because the Constitution of our State requires protection of the interests involved and because the Legislature has not protected them. We recognize the social and economic controversy (and its political consequences) that has resulted in relatively little legislative action in this field. We understand the enormous difficulty of achieving a political consensus that might lead to significant legislation enforcing the constitutional mandate better than we can, legislation that might completely remove this Court from those controversies. But enforcement of constitutional rights cannot await a supporting political consensus. So while we have always preferred legislative to judicial action in this field, we shall continue — until the Legislature acts — to do our best to uphold the constitutional obligation that underlies the Mount Laurel *213doctrine. That is our duty. We may not build houses, but we do enforce the Constitution.7

We note that there has been some legislative initiative in this field. We look forward to more. The new Municipal Land Use Law explicitly recognizes the obligation of municipalities to zone with regional consequences in mind, N.J.S.A. 40:55D-28(d); it also recognizes the work of the Division of State and Regional Planning in the Department of Community Affairs (DCA), in creating the State Development Guide Plan (1980) (SDGP), which plays an important part in our decisions today. Our deference to these legislative and executive initiatives can be regarded as a clear signal of our readiness to defer further to more substantial actions.

The judicial role, however, which could decrease as a result of legislative and executive action, necessarily will expand to the extent that we remain virtually alone in this field. In the absence of adequate legislative and executive help, we must give meaning to the constitutional doctrine in the cases before us *214through our own devices, even if they are relatively less suitable. That is the basic explanation of our decisions today.

C. Summary of Rulings

Our rulings today have several purposes. First, we intend to encourage voluntary compliance with the constitutional obligation by defining it more clearly. We believe that the use of the State Development Guide Plan and the confinement of all Mount Laurel litigation to a small group of judges, selected by the Chief Justice with the approval of the Court, will tend to serve that purpose. Second, we hope to simplify litigation in this area. While we are not overly optimistic, we think that the remedial use of the SDGP may achieve that purpose, given the significance accorded it in this opinion. Third, the decisions are intended to increase substantially the effectiveness of the judicial remedy. In most cases, upon determination that the municipality has not fulfilled its constitutional obligation, the trial court will retain jurisdiction, order an immediate revision of the ordinance (including, if necessary, supervision of the revision through a court appointed master), and require the use of effective affirmative planning and zoning devices. The long delays of interminable appellate review will be discouraged, if not completely ended, and the opportunity for low and moderate income housing found in the new ordinance will be as realistic as judicial remedies can make it. We hope to achieve all of these purposes while preserving the fundamental legitimate control of municipalities over their own zoning and, indeed, their destiny.

The following is a summary of the more significant rulings of these cases:

(1) Every municipality’s land use regulations should provide a realistic opportunity for decent housing for at least some part of its resident poor who now occupy dilapidated housing. The zoning power is no more abused by keeping out the region’s poor than by forcing out the resident poor. In other words, each municipality must provide a realistic opportunity for decent *215housing for its indigenous poor except where they represent a disproportionately large segment of the population as compared with the rest of the region. This is the case in many of our urban areas.

(2) The existence of a municipal obligation to provide a realistic opportunity for a fair share of the region’s present and prospective low and moderate income housing need will no longer be determined by whether or not a municipality is “developing.” The obligation extends, instead, to every municipality, any portion of which is designated by the State, through the SDGP as a “growth area.” This obligation, imposed as a remedial measure, does not extend to those areas where the SDGP discourages growth — namely, open spaces, rural areas, prime farmland, conservation areas, limited growth areas, parts of the Pinelands and certain Coastal Zone areas. The SDGP represents the conscious determination of the State, through the executive and legislative branches, on how best to plan its future. It appropriately serves as a judicial remedial tool. The obligation to encourage lower income housing, therefore, will hereafter depend on rational long-range land use planning (incorporated into the SDGP) rather than upon the sheer economic forces that have dictated whether a municipality is “developing.” Moreover, the fact that a municipality is fully developed does not eliminate this obligation although, obviously, it may affect the extent of the obligation and the timing of its satisfaction. The remedial obligation of municipalities that consist of both “growth areas” and other areas may be reduced, based on many factors, as compared to a municipality completely within a “growth area.”

There shall be a heavy burden on any party seeking to vary the foregoing remedial consequences of the SDGP designations.

(3) Mount Laurel litigation will ordinarily include proof of the municipality’s fair share of low and moderate income housing in terms of the number of units needed immediately, as well as the number needed for a reasonable period of time in the future. *216“Numberless” resolution of the issue based upon a conclusion that the ordinance provides a realistic opportunity for some low and moderate income housing will be insufficient. Plaintiffs, however, will still be able to prove a prima facie case, without proving the precise fair share of the municipality, by proving that the zoning ordinance is substantially affected by restrictive devices, that proof creating a presumption that the ordinance is invalid.

The municipal obligation to provide a realistic opportunity for low and moderate income housing is not satisfied by a good faith attempt. The housing opportunity provided must, in fact, be the substantial equivalent of the fair share.

(4) Any future Mount Laurel litigation shall be assigned only to those judges selected by the Chief Justice with the approval of the Supreme Court. The initial group shall consist of three judges, the number to be increased or decreased hereafter by the Chief Justice with the Court’s approval. The Chief Justice shall define the area of the State for which each of the three judges is responsible: any Mount Laurel case challenging the land use ordinance of a municipality included in that area shall be assigned to that judge.

Since the same judge will hear and decide all Mount Laurel cases within a particular area and only three judges will do so in the entire state, we believe that over a period of time a consistent pattern of regions will emerge. Consistency is more likely as well in determinations of regional housing needs and allocations of fair share to municipalities within the region. Along with this consistency will come the predictability needed to give full effect to the Mount Laurel doctrine. While determinations of region and regional housing need will not be conclusive as to any municipality not a party to the litigation, they shall be given presumptive validity in subsequent litigation involving any municipality included in a previously determined region.

*217The Chief Justice will analyze all pending Mount Laurel litigation to determine which, if any, should be transferred to one of the three Mount Laurel judges. As for the cases pending before us, given the knowledge acquired by the judges of the particular facts of the case, each will be remanded to the judge who heard the matter below with the exception of Round Valley, Inc. v. Clinton and Urban League of Greater New Brunswick v. Carteret, since neither of the judges who determined those matters remains on the trial bench.

(5) The municipal obligation to provide a realistic opportunity for the construction of its fair share of low and moderate income housing may require more than the elimination of unnecessary cost-producing requirements and restrictions. Affirmative governmental devices should be used to make that opportunity realistic, including lower-income density bonuses and mandatory set-asides. Furthermore the municipality should cooperate with the developer’s attempts to obtain federal subsidies. For instance, where federal subsidies depend on the municipality providing certain municipal tax treatment allowed by state statutes for lower income housing, the municipality should make a good faith effort to provide it. Mobile homes may not be prohibited, unless there is solid proof that sound planning in a particular municipality requires such prohibition.

(6) The lower income regional housing need is comprised of both low and moderate income housing. A municipality’s fair share should include both in such proportion as reflects consideration of all relevant factors, including the proportion of low and moderate income housing that make up the regional need.

(7) Providing a realistic opportunity for the construction of least-cost housing will satisfy a municipality’s Mount Laurel obligation if, and only if, it cannot otherwise be satisfied. In other words, it is only after all alternatives have been explored, all affirmative devices considered, including, where appropriate, a reasonable period of time to determine whether low and moderate income housing is produced, only when everything has *218been considered and tried in order to produce a realistic opportunity for low and moderate income housing that least-cost housing will provide an adequate substitute. Least-cost housing means what it says, namely, housing that can be produced at the lowest possible price consistent with minimal standards of health and safety.

(8) Builder’s remedies will be afforded to plaintiffs in Mount Laurel litigation where appropriate, on a case-by-case basis. Where the plaintiff has acted in good faith, attempted to obtain relief without litigation, and thereafter vindicates the constitutional obligation in Mount Laurel -type litigation, ordinarily a builder’s remedy will be granted, provided that the proposed project includes an appropriate portion of low and moderate income housing, and provided further that it is located and designed in accordance with sound zoning and planning concepts, including its environmental impact.

(9) The judiciary should manage Mount Laurel litigation to dispose of a case in all of its aspects with one trial and one appeal, unless substantial considerations indicate some other course. This means that in most cases after a determination of invalidity, and prior to final judgment and possible appeal, the municipality will be required to rezone, preserving its contention that the trial court’s adjudication was incorrect. If an appeal is taken, all facets of the litigation will be considered by the appellate court including both the correctness of the lower court’s determination of invalidity, the scope of remedies imposed on the municipality, and the validity of the ordinance adopted after the judgment of invalidity. The grant or denial of a stay will depend upon the circumstances of each case. The trial court will appoint a master to assist in formulating and implementing a proper remedy whenever that course seems desirable.

(10) The Mount Laurel obligation to meet the prospective lower income housing need of the region is, by definition, one that is met year after year in the future, throughout the years *219of the particular projection used in calculating prospective need. In this sense the affirmative obligation to provide a realistic opportunity to construct a fair share of lower income housing is met by a “phase-in” over those years; it need not be provided immediately. Nevertheless, there may be circumstances in which the obligation requires zoning that will provide an immediate opportunity — for instance, zoning to meet the region’s present lower income housing need. In some cases, the provision of such a realistic opportunity might result in the immediate construction of lower income housing in such quantity as would radically transform the municipality overnight. Trial courts shall have the discretion, under those circumstances, to moderate the impact of such housing by allowing even the present need to be phased in over a period of years. Such power, however, should be exercised sparingly. The same power may be exercised in the satisfaction of prospective need, equally sparingly, and with special care to assure that such further postponement will not significantly dilute the Mount Laurel obligation.

We reassure all concerned that Mount Laurel is not designed to sweep away all land use restrictions or leave our open spaces and natural resources prey to speculators. Municipalities consisting largely of conservation, agricultural, or environmentally sensitive areas will not be required to grow because of Mount Laurel. No forests or small towns need be paved over and covered with high-rise apartments as a result of today’s decision.

As for those municipalities that may have to make adjustments in their lifestyles to provide for their fair share of low and moderate income housing, they should remember that they are not being required to provide more than their fair share. No one community need be concerned that it will be radically transformed by a deluge of low and moderate income developments. Nor should any community conclude that its residents will move to other suburbs as a result of this decision, for those “other suburbs” may very well be required to do their part to provide the same housing. Finally, once a community has *220satisfied its fair share obligation, the Mount Laurel doctrine will not restrict other measures, including large-lot and open area zoning, that would maintain its beauty and communal character.

Many of these points will be discussed later in this opinion. We mention them now only to reassure all concerned that any changes brought about by this opinion need not be drastic or destructive. Our scenic and rural areas will remain essentially scenic and rural, and our suburban communities will retain their basic suburban character. But there will be some change, as there must be if the constitutional rights of our lower income citizens are ever to be protected. That change will be much less painful for us than the status quo has been for them.

II.

Resolution of the Issues

A. Defining the Mount Laurel Obligation

In Oakwood v. Madison, this Court held that it was sufficient in Mount Laurel litigation for courts to look to the “substance” of challenged zoning ordinances and to the existence of “bona fide efforts” by municipalities to meet their obligations. 72 N.J. at 499. It was hoped that this test would adequately protect the constitutional rights of lower income persons while at the same time minimizing the role of the courts in this area. Unfortunately, experience has taught us that this formulation is too vague to provide adequate guidance for either trial courts or municipalities. As the Mount Laurel II and Mahwah cases demonstrate, the Madison test does not ensure sufficient judicial scrutiny of zoning ordinances. Even those that plainly fail to meet the requisites of the Mount Laurel doctrine may pass the test of Madison.

Therefore, proof of a municipality’s bona fide attempt to provide a realistic opportunity to construct its fair share of lower income housing shall no longer suffice. Satisfaction of the Mount Laurel obligation shall be determined solely on an *221objective basis: if the municipality has in fact provided a realistic opportunity for the construction of its fair share of low and moderate income housing, it has met the Mount Laurel obligation to satisfy the constitutional requirement; if it has not, then it has failed to satisfy it.8 Further, whether the *222opportunity is “realistic” will depend on whether there is in fact a likelihood — to the extent economic conditions allow — that the lower income housing will actually be constructed. Plaintiff’s case will ordinarily include proof of the municipality’s fair share of the regional need and defendant’s proof of its satisfaction. Good or bad faith, at least on this issue, will be irrelevant. The numberless approach encouraged in Madison, where neither plaintiffs nor defendants are required to prove a fair share number, is no longer acceptable.

The numberless approach is to be distinguished, however, from presumptive facial invalidity. Plaintiff may continue to prove (in addition to or instead of proving the fair share obligation of the municipality) that the land use regulations fail to provide a realistic opportunity for low and moderate income housing or that they contain “expressly prescribed requirements or restrictions which preclude or substantially hinder it.” Mount Laurel I, 67 N.J. at 180-81. As before, such a showing shall create a prima facie case of a failure to satisfy the Mount Laurel obligation. The municipality shall then have the heavy burden of demonstrating, by a preponderance of the evidence, its fair share and its satisfaction of that share, or any justifica*223tion of its failure. It shall not be sufficient in such cases to show merely that there are one, two or three zones that purport to contain provisions for multi-family dwellings: what is needed where facial invalidity is relied on by the plaintiff is a definite presentation of facts by the defendant-municipality that shows that it has satisfied its fair share obligation.

In the remainder of Section II, we will restate what “fair share” means and what municipalities and courts must do to ensure that Mount Laurel obligations are met. Section IIB, using the DCA’s SDGP as the remedial standard, sets forth which municipalities have a prospective fair share obligation. Section IIC describes various ways by which the prospective fair share of municipalities may be calculated. Section IID describes the mechanisms municipalities must use to meet their Mount Laurel obligations. Section HE outlines the remedies available to trial courts to ensure compliance with our mandate. Finally, Section IIF emphasizes the importance of judicial management in making Mount Laurel effective.

B. Determining the Mount Laurel Obligation: Use of the State Development Guide Plan

The initial question in every Mount Laurel case is whether the municipality is subject to the Mount Laurel obligation. In its initial formulation in Mount Laurel I, this Court described the characteristics of Mount Laurel, implying that any municipality with similar characteristics would have the obligation announced in that opinion. Mount Laurel I, 67 N.J. at 160. Those municipalities are referred to as “developing municipalities.” Id. at 190. All subsequent litigation concerning the doctrine treated this preliminary determination as a condition precedent to its applicability (although there were pointed disagreements suggesting the developing-developed distinction inequitable, e.g., Pascack, 74 N.J. at 494-95 (Schreiber, J., concurring)), and the particular factors descriptive of Mount Laurel set forth in that opinion became fixed as the “six criteria of a developing municipality”:

*224A developing municipality (1) has a sizeable land area, (2) lies outside the central cities and older built-up suburbs, (3) has substantially shed rural characteristics, (4) has undergone great population increase since World War II or is now in the process of doing so, (5) is not completely developed and (6) is in the path of inevitable future residential, commercial and industrial demand and growth. [Glenview Development Co., 164 N.J.Super. at 567-68].

These criteria are discussed as if each must be satisfied in order for a municipality to be “developing.”

There are various drawbacks to this approach to the critical question of determining the existence of the obligation. Uncertainty is one of them. Ideally a municipality, and its governing body, should know without question whether it is subject to the Mount Laurel remedy, for without that knowledge municipalities that are “borderline” (between developing and non-developing) cannot be expected to comply with an obligation that may very well not exist. Given the foreseeable political pressures, governing bodies in that situation are almost certain to take the position either that the constitutional remedy or obligation does not apply, or that if it does apply, it has not been violated, or that their responsibility to the municipality and its residents requires that the issues be determined in litigation.

Of at least equal importance, the criteria will not necessarily result in the imposition of the obligation in accordance with sound planning. There may be areas that fit the “developing” description that should not yield to “inevitable future residential, commercial and industrial demand and growth.” Those areas may contain prime agricultural land, open spaces and areas of scenic beauty; apart from these their development might impose unacceptable demands on public investment to extend the infrastructure required to support such growth. Indeed, to some extent the very definition of “developing” suggests results that are quite the opposite of sound planning, for the whole purpose of planning is to prevent or deflect what would otherwise be “inevitable.”

Lacking any official guidance, however, as to the state’s plans for its own future, its own determination of where development should occur and where it should not, and what kind of develop*225ment, this Court fashioned its own remedial planning guide in the form of a definition of “developing.” It was obvious to anyone who studied the matter that such definition of the Mount Laurel responsibility furnished no guarantee that if lower income housing resulted, it would be built where it should be built, i.e., where a comprehensive plan for the State of New Jersey might indicate such development was desirable. We proceeded in spite of this drawback since, given the constitutional requirement and the lack of any assurance that such a statewide plan would be forthcoming, there appeared no justification for delay.

We now have a satisfactory alternative. The State Development Guide Plan (May 1980) promulgated pursuant to N.J.S.A. 13:1B-15.52, provides a statewide blueprint for future development. Its remedial use in Mount Laurel disputes will ensure that the imposition of fair share obligations will coincide with the State’s regional planning goals and objectives.

The SDGP represents the only official determination of the state’s plan for its own future development and growth. It is substantially similar, in concept and approach, to various regional planning documents by other entities, such as the Tri-State Regional Planning Association, Delaware Valley Regional Planning Association, the Regional Plan Association, and the Middle-sex, Somerset, Mercer Regional Study Council, Inc., which have the goal of guiding all new development within their planning jurisdictions. The SDGP resulted from an intensive study of all aspects of New Jersey’s current growth and development considered in conjunction with the “physical assets” of the state: its natural resources, open spaces, farmland, “infrastructure” (transportation, sewage facilities, water supplies and facilities), including the location of present intensive development, employment centers, community facilities, recreation areas, etc.9 By *226using proven sound planning concepts the Division of State and Regional Planning, statutorily charged with the obligation (N.J. S:A. 13:1B-15.52), developed a master plan (the SDGP and the Concept Map) for the purpose of guiding the future growth and development of this state.

The SDGP divides the state into six basic areas: growth, limited growth, agriculture, conservation, pinelands and coastal zones (the pinelands and coastal zones actually being the product of other protective legislation).10 While it does not purport to draw its lines so finely as to delineate actual municipal boundaries or specific parcels of land, the concept map, through the county maps, makes it quite clear how every municipality in the state should be classified (see Appendix). By clearly setting forth the state’s policy as to where growth should be encouraged and discouraged, these maps effectively serve as a blueprint for the implementation of the Mount Laurel doctrine. Pursuant to the concept map, development (including residential development) is targeted for areas characterized as “growth.” The Mount Laurel obligation should, as a matter of sound judicial discretion reflecting public policy, be consistent with the state’s plan for its future development. Consequently, *227the obligation should apply in these “growth” areas, and only in these areas, subject to the exceptions mentioned infra at 240-243.11

The use of the SDGP for this purpose is consistent with the statute authorizing its preparation and with its actual use by the Legislature, counties, municipalities, the Federal government and the Division of State and Regional Planning within the Department of Community Affairs. The administrators who carried out the legislative requirement to prepare such a plan “... for the future improvement and development of the State,” N.J.S.A. 13:113-15.52 a.(2), interpreted the statute to require a plan that would guide and influence the location of future development, including residential development. Channeling the development impetus of the Mount Laurel doctrine into “growth areas” is precisely the kind of use of the plan that was intended by those who prepared it.

The statute requires the Division of State and Regional Planning to “[pjromote programs to insure the orderly development of the State’s physical assets by ... preparing and maintaining a comprehensive guide plan and long term development and capital improvement program for the future improvement and development of the State ... . ” N.J.S.A. 13:1B-I5.52(a), (a)(2). The same section of the statute requires the Division to “as-*228sembl[e] and analyzje] pertinent facts as to existing development conditions and trends.” N.J.S.A. 13:1B-I5.52(a)(l). The plan is “comprehensive,” its intent is to “guide ... the future ... development of the State,” and its purpose of assuring the orderly development of the State’s “physical assets” (land, open spaces, infrastructure — all of the State’s natural and man-made “physical resources” (SDGP at ii)) is to be achieved by, among other things, “stimulating, assisting and coordinating local, county and regional planning activities.” N.J.S.A. 13:1B-15.-52(a)(4).12

The Division of State and Regional Planning completed its Horizon Plan and Ten Million Plan in the 1960’s, see SDGP, Preface at 1, and by 1975 was well along in its studies and work aimed at the preparation of the State Development Guide Plan. That work was proceeding through the efforts of the State Planning Task Force started in the early 1970’s.

In 1975, the Legislature recognized and supported this effort to guide the further development of the state in accordance with that comprehensive guide plan. Presumably with knowledge of the Division’s ongoing work in preparing the SDGP, it required all municipalities to consider the relationship of their master plans to the SDGP, each master plan to “include a specific policy statement indicating the relationship of the proposed development of the municipality as developed in the master plan to ... [the State Development Guide Plan].” N.J.S.A. 40:55D-28(d). While it did not mandate conformance of the municipal master plan or the development of the municipality to the SDGP, the legislative intent was clear: municipalities were encouraged to *229guide their development in conformance with the state plan to make it more likely that through voluntary municipal action, the future development of the entire state would be in accordance with comprehensive sound planning.

This legislatively mandated use of the SDGP is found in the Municipal Land Use Law, L.1975, c. 291, N.J.S.A. 40:55D-1 to -92, in which the Legislature explicitly recognized the importance of regional planning and the need to integrate each municipality’s development with the development of the state as a whole. Among the purposes of the Act are the following: “to encourage municipal action to guide the appropriate use or development of all lands in this State,” “to ensure that the development of individual municipalities does not conflict with the development and general welfare of neighboring municipalities, the county and the State as a whole,” “to promote the establishment of appropriate population densities and concentrations that will contribute to the well-being of persons, neighborhoods, communities and regions ...,” “to provide sufficient space in appropriate locations for a variety of agricultural, residential, recreational, commercial and industrial uses and open space, both public and private, according to their respective environmental requirements in order to meet the needs of all New Jersey citizens,” and “to promote the conservation of open space and valuable natural resources and to prevent urban sprawl and degradation of the environment through improper use of land.” N.J.S.A. 40:55D-2.

Among the many devices found in the law to achieve these purposes is the municipal master plan. That plan, which must relate to the SDGP, is “to guide the use of lands within the municipality.” Thus, it is essentially a plan to help determine, control, and provide locations for the municipality’s future growth. N.J.S.A. 40:55D-28(a). Among other things, master plans require “[a] land use plan element ... showing the existing and proposed location, extent and intensity of development of land to be used in the future for varying types of residential, commercial, industrial, agricultural, recreational, ed*230ucational and other public and private purposes or combination of purposes, and ... including a statement of the standards of population density and development intensity recommended for the municipality.” N.J.S.A. 45:55D-28 b(2).

When the Division of State and Regional Planning circulated its first draft of the SDGP in 1977 and then (after broad consultation with citizens throughout the state involved in the planning process, including county and regional planning agencies) released the SDGP in May 1980, it knew that the areas it had carved out of the state and designated for “growth,” “limited growth,” “conservation,” and “agricultural” would be used by municipalities in determining where development in fact should and should not occur. The SDGP was not only a document that could be used to decide where growth should be encouraged and discouraged, permitted, and prohibited, but a document that the Legislature said must be used for that purpose. And so it has been used by municipalities in accordance with the Municipal Land Use Law and in many other ways.

The primary function of the SDGP is to determine where growth, including residential growth, should be encouraged or discouraged.

The State Development Guide Plan is a policy statement about the State’s future growth and development. It contains a concept map which shows spacially where growth should be either discouraged, encouraged or delayed, and reflects the need to balance conservation areas, agricultural land and water resource imperatives with opportunities for further economic and residential expansion. [SDGP at ii].

Speaking of population and employment projections that indicate that the already-developed “pattern of more widespread suburbanization will continue into the 1980’s,” and noting that “this could result in the development of substantial amounts of open land for residential, commercial and industrial uses,” the SDGP describes the challenge that it is designed to help meet:

A major challenge faced by the people of New Jersey is how to guide this projected growth so that open space and environmental quality are retained while, at the same time, good residential areas are made available, needed employment opportunities are created, and public investments are efficiently utilized and developed. [SDGP at 2],

*231Housing is one of the major factors discussed in the Plan, along with population distribution and growth, the economy, energy, urban areas, infrastructure, environmental quality and natural resources. The Plan notes that “a major challenge in the coming years will be to provide a variety of housing opportunities in appropriate locations for New Jersey’s expanding population.” SDGP at 7. Referring to the “suburbanization process” that followed World War ÍI as “expensive and wasteful,” the Plan notes “a need now in New Jersey to alter this unplanned pattern of spread development. A compact development pattern for the future can serve to promote the utilization of the existing infrastructure and service system in an economical way .... It is now suggested that a major portion of the State’s development efforts should be directed to areas within and contiguous to existing development.” SDGP at 25. One of its growth management strategies advocates “a suitable balance between conservation and growth in New Jersey with space for both the conservation of agricultural and critical environmental areas and for residential and economic growth.” SDGP at 26.

"The overall strategy of the Guide Plan is to attain a sharper focus in governmental efforts directed to urban, suburban and open space areas so that the proposed patterns of conservation and development can be realized.” SDGP at 80. The strategy is one of “discouraging population expansion” in “Limited Growth Areas,” SDGP at 91, “to refrain from public investment in growth inducing facilities” in “Agricultural Areas,” SDGP at 90, to utilize “acquisitions and regulatory control, the withholding of major public investments for growth-inducing facilities to deter development” in conservation areas, SDGP at 87, and to target public investments for new growth-inducing facilities to “Growth Areas,” and within those areas “encourage housing development in proximity to jobs, commercial areas and public transportation,” and “provide a variety of housing types so that households of varying sizes and incomes can find suitable housing,” SDGP at 86, “Growth Areas” being those parts of the *232State previously defined as being “particularly suitable for development.” SDGP at 47.13

*233The remedial use by this Court of the SDGP as the primary standard to determine the locus of the Mount Laurel obligation, and consequently to determine where development (in this case housing) should be encouraged and, as importantly, its use to assure that the Mount Laurel doctrine does not encourage development in conflict with the State’s comprehensive plan, is thus the kind of use of the SDGP contemplated by the Legislature in various statutes, and by the Plan itself.

For instance, it is clear that municipal master plans, pursuant to the statutory mandate, have considered the SDGP; that many seem to view it as a helpful guide; that some conscientiously attempt to conform their proposed development to that suggested in the Plan; that others comply with it out of a concern that needed public funds will not be forthcoming unless they do; and that others simply note their consideration in a pro forma manner. The overriding fact, however, is that the SDGP is being used, to a greater or lesser extent, by municipalities in planning for their future development, and in particular is being used to determine where future development, including housing, should be located by referring to those areas of the SDGP classified as “growth areas.”

While its impact is not clear, the SDGP is being used by the state in commenting on all applications for major subdivi*234sions exceeding either 150 acres or 500 dwelling units. The Municipal Land Use Law requires that the Division of State and Regional Planning be notified of hearings on such applications (N.J.S.A. 40:55D-12(g)), the Legislature presumably intending that the location of such proposed developments be reviewed by the Division to determine if they conform with sound statewide comprehensive planning- — and ultimately if they conform with the comprehensive guide plan that the Legislature required all master plans to consider. The obvious purpose of this provision is to enable the Division to advise local agencies, before they act on such applications, of the relationship of the proposed development to statewide comprehensive planning and of the recommendations of the Division concerning either approval or disapproval, and of the conditions that might be considered in connection with such applications. The Division has used the SDGP “as a reference in its review of major subdivisions,” SDGP at iii, “to evaluate the suitability of major subdivision proposals [of this kind]” in order to “assess [ ] major development proposals in terms of statewide priorities and policies .. . [and to] shar[e] such assessments with the private sector and local governments concerned,” SDGP at 80 (in other words to let the municipality and the proposed developer know of the Division’s recommendations concerning the proposed development based on the SDGP). The direction by the Legislature provides a practical support for its declared policy that municipal land use regulations shall be applied in accordance with regional and statewide planning objectives, for it brings the Division of State and Regional Planning and the SDGP directly onto the stage where the development decisions are made.

The SDGP is also used by the state and its agencies in reviewing their own “functional plans.” “Some agencies have found the plan useful and have incorporated its major recommendations within their own programs. Some progress has been made, though on an informal basis, toward establishing a unified statewide land use and investment policy.” SDGP at iii. The Governor’s Office of Policy and Planning, created in 1978, *235designed to assure that the policies of state departments are complementary and mutually enforcing, has “given impetus to the movement toward coordinated comprehensive land use policies.” Id. The SDGP is “designed to assist the Governor’s Office of Policy and Planning and the various cabinet committees it serves,” id. at iv, in this function. The clear implication is that in accordance with its intent, the SDGP is being used to help guide state investment policies, capital growth strategies, and overall programmatic policies.

The SDGP has been used, since 1977, by the Division in its review of and for its comments on “applications for federal assistance processed through the Project Notification and Review System.” SDGP at iii. This is a system designed to obtain comments that will enable the federal government to determine whether its grant programs are in conflict with comprehensive planning projects of regions and states, to what extent in conflict, all for the purpose of determining whether or not to approve or reject an application, or to attach varying conditions to it. Many of the projects for which federal aid is sought are development-inducing (sewage construction, roads, water treatment, etc.); they may determine as much as any other governmental policy where residential development will occur and where it will not.

A brochure issued by the Division in September 1981 notes that its review of a federal aid application is to determine, among other things, the application’s “conformance with the State Development Guide Plan” as mandated by the federal legislation. Case studies in the brochure show the extent to which the SDGP’s classification of the State into growth areas and others has actually been used in influencing decisions affecting the State’s development.14

*236The lessons of history are clear, even if rarely learned. One of those lessons is that unplanned growth has a price: natural resources are destroyed, open spaces are despoiled, agricultural land is rendered forever unproductive, and people settle without regard to the enormous cost of the public facilities needed to support them. Cities decay; established infrastructures deteriorate for lack of funds; and taxpayers shudder under a financial burden of public expenditures resulting in part from uncontrolled migration to anywhere anyone wants to settle, roads leading to places they should never be — a pattern of total neglect of sensible conservation of resources, funds, prior public investment, and just plain common sense. These costs in New Jersey, the most highly urbanized state in the nation, are staggering, and our knowledge of our limited ability to support them has become acute. More than money is involved, for natural and man-made physical resources are irreversibly damaged. Statewide comprehensive planning is no longer simply desirable, it is a necessity recognized by both the federal and state governments.

Based on all of the foregoing, we are able to fashion judicial relief through means not available to us when we established the “developing municipality” remedial doctrine. These considerations, founded in sound public policy relating to comprehensive planning, are compelling in favor of a remedial solution that imposes the Mount Laurel obligation only in those areas designated as “growth areas” by the SDGP. For reasons shortly to be noted, we have decided to allow some limited variation from that rule. The point here is that we see every reason to modify what is generally regarded as one of the doctrines of Mount Laurel I, namely, that the Mount Laurel obligation applies only in developing municipalities, and no reason, either in the constitutional doctrine or in the Mount Laurel case itself, not to do so.

*237That we are not inhibited by the Constitution from making this change is apparent when one analyzes the constitutional obligation itself. Mount Laurel 1 held that in the exercise of the zoning power a municipality could not constitutionally limit to its own citizens those whose housing needs it would consider, but was required to consider the housing needs of all of the citizens of the region of which that municipality was a part. Put differently, the zoning power that the State exercised through its municipalities would have constitutional validity only if regional housing needs were addressed by the actions of the municipalities in the aggregate. The method selected by this Court in Mount Laurel I for achieving that constitutionally mandated goal was to impose the obligation on those municipalities that were “developing.” Clearly, however, the method adopted was simply a judicial remedy to redress a constitutional injury. Achievement of the constitutional goal, rather than the method of relief selected to achieve it, was the constitutional requirement.

Since our imposition of the Mount Laurel obligation on municipalities containing growth areas as defined by the SDGP (rather than on “developing” municipalities) is just as clearly related to achieving the constitutional goal, it is equally constitutionally valid. Furthermore, it is significantly preferable for other reasons. The constitutional obligation of the State of New Jersey in exercising its zoning power through its municipal subdivisions to provide a realistic opportunity for lower income housing for its citizens can just as well be met by requiring housing in municipalities in conformance with sound planning concepts as with judicially devised characterizations that may or may not advance other important policies of the state.

As this Court pointed out in Mount Laurel I, it might be “sounder” to have one municipality in the region have more lower income housing than another “because of greater availability of suitable land, location of employment, accessibility of public transportation or some other significant reason” — in other *238words for the combination of reasons that add up to a sound planning decision. Though it might be sounder, we reluctantly concluded that “every municipality [in a region] must bear its fair share of the regional burden.” 67 N.J. at 189 (emphasis supplied). We thought then that our hands were tied, that we could not distribute the Mount Laurel obligation in accordance with sound planning criteria both because of the method of distributing the tax burden in New Jersey and because zoning was not permitted on a regional basis. Id. at 189 & n. 22. Today, however, zoning in accordance with regional considerations is not only permissible, it is mandated as noted above. Furthermore, while we are far from achieving tax equality among all the municipalities of the state, our present programs of State aid to education (financed through an income tax that was not in effect at the time of our decision in Mount Laurel I) are designed to reduce significantly the differential school tax burden between municipalities that accept residential development and those that do not. As we view it, therefore, there is no reason today not to impose the Mount Laurel obligation in accordance with sound planning concepts, no reason in our Constitution to make every municipality a microcosm of the' entire state in its housing pattern, and there are persuasive reasons based on sound planning not to do so.

The Constitution of the State of New Jersey does not require bad planning. It does not require suburban spread. It does not require rural municipalities to encourage large scale housing developments. It does not require wasteful extension of roads and needless construction of sewer and water facilities for the out-migration of people from the cities and the suburbs. There is nothing in our Constitution that says that we cannot satisfy our constitutional obligation to provide lower income housing and, at the same time, plan the future of the state intelligently.

Sound planning requires that municipalities containing “growth areas” have a Mount Laurel obligation and that, together, all of those municipalities affirmatively provide a *239realistic opportunity for the construction of sufficient lower income housing to meet the needs of New Jersey’s lower income population. And, as among those municipalities containing “growth areas,” the Constitution does not prohibit further distinctions, some municipalities being required to take more than others because a combination of factors suggests that they are more suitable for such development. The thought that “suitability” may determine and validate distinctions in uses between municipalities was expressed by Chief Justice Vanderbilt in Duffcon Concrete Products v. Borough of Cresskill, 1 N.J. 509 (1949), one of the first cases to evaluate a zoning ordinance in the context of regional characteristics and needs.

There may be inequities between and among these municipalities located within growth areas, as there undoubtedly are between all of them and municipalities outside of growth areas, for the tax and other burdens caused by the location of lower income housing will not be fairly spread. The state, however, has made its decision as to where this development should occur. If location in accordance with that state plan has adverse economic consequences, it would be appropriate for the state, rather than this Court, to correct them.

As noted above, we have decided not to make the SDGP the absolute determinant of the locus of the Mount Laurel obligation. Our reluctance to give it conclusive effect is based on the fact that while it has the legitimacy of legislative authorization, the Legislature has neither explicitly authorized its use for Mount Laurel purposes nor mandated that the actual use of land, as permitted in zoning ordinances, conform to the SDGP. Given these circumstances, we deem it prudent to allow parties to attempt to persuade the trial court, in a particular case, that the SDGP should not determine whether the Mount Laurel doctrine applies to the particular municipality involved in the case. While we believe important policy considerations are involved in our decision not to make the SDGP conclusive, we think it even more important to point out that it will be the *240unusual case that concludes the locus of the Mount Laurel obligation is different from that found in the SDGP. Subject to those cases, we hold that henceforth, only those municipalities containing “growth areas” as shown on the concept map of the SDGP (or any official revision thereof) shall be subject to the Mount Laurel prospective need obligation.15

Any party in Mount Laurel litigation seeking a ruling that varies the locus of the Mount Laurel obligation from the SDGP growth areas will have to prove one of the following: (1) accepting the premises of the SDGP, the conclusion that the municipality includes any growth area, or as much growth area as is shown on the concept map, is arbitrary and capricious, or, alternatively, the conclusion that the municipality does not contain any growth area whatsoever is arbitrary and capricious; (2) since the preparation of the concept map (or any revision thereof) the municipality has undergone a significant transformation that renders the SDGP’s characterization of it inappropriate, admitting that at the time of the preparation of the SDGP and the concept map (or any revision thereof) the classification of the municipality was correct; or (3) (and this exception shall apply only if the concept map is not revised before January 1,1985) subsequent to the date of this decision the municipality, containing no “growth area,” encourages or allows commercial, *241residential or industrial development or, if it contains some “growth area,” encourages or allows development outside of that area.

The foregoing exceptions will allow a party to have the court impose a Mount Laurel obligation on a municipality that has no growth area as shown on the concept map, or to impose a greater Mount Laurel obligation by, in effect, proving that the growth area should be enlarged, or, conversely, to relieve a municipality from any Mount Laurel obligation even though the concept map shows it as including a “growth area,” or to diminish the obligation by proving that the “growth area” shown on the concept map should be cut down.

The first exception recognizes the possibility of errors on the part of the planning group that prepared the SDGP. No trial court should, however, simply substitute its judgment for the state’s planners’ under that exception. Not only must the evidence show that the conclusion and the classification were arbitrary and capricious, but the party challenging the characterization must contend with the obvious fact that lines must be drawn somewhere and that merely to show that one municipality containing a “growth area” is remarkably similar to a neighboring one that includes no “growth area” is not enough: the party must show that it was arbitrary and capricious not to place the line somewhere else.

The second exception requires proof of substantial change. Those who prepared the SDGP and the concept map obviously realized that conditions would change after its publication, that-planning is a dynamic process, and that plans like the SDGP must remain current. Changes, therefore, sufficient to warrant reclassification of a municipality in whole or in part should be addressed not by the court, but by the Division when it revises the SDGP. The second exception, however, recognizes the possibility that prior to such a revision a municipality may change sufficiently so that it is inappropriate to retain its present SDGP classification. If a municipality that is substan*242tially rural changes only to the extent of an added industrial use and a fairly large residential subdivision, that might or might not constitute a substantial change, depending on all of the circumstances; if in addition there was further development of its infrastructure and several new substantial places of work and residential subdivisions, that municipality’s SDGP classification should probably be changed. Furthermore, if the trial court finds that subsequent to the date of this decision the municipality has encouraged or allowed development, it should more readily conclude that the challenged SDGP “non-growth” characterization has become inappropriate. We do not intend to allow the SDGP to be used as a wall behind which municipalities may create or expand exclusionary developments.

The third exception recognizes that if the planning process does not remain a continuing one, the categories set forth in the SDGP might become unrealistic and certainly would lose a considerable degree of their legitimacy. It is one thing for a court to defer to the judgment of the planners, even where it disagrees; it is another to defer to a document that is clearly out of date where deferral might frustrate a constitutional obligation. In order for it to remain a viable remedial standard, we believe that the SDGP should be revised no later than January 1,1985 (and, in the absence of proof of a more appropriate period, every three years thereafter).16 If it is not, then courts shall have considerable discretion to vary the locus of the Mount Laurel obligation from that shown on the present SDGP concept map. For instance, if, after the date of this decision, a municipality containing no growth area allows the construction *243of a significant industrial use creating significant employment opportunities, that would be sufficient to justify a court in imposing a Mount Laurel remedy on that municipality as if a portion of it had been characterized as “growth area”; the same conclusion would follow if such a municipality, after the date of this decision, encourages or allows the construction of a residential subdivision, or if, though unsuccessful, it attempts to attract development of either kind or of a commercial nature. Such relative ease of variance from the SDGP shall cease, however, when the SDGP is thereafter brought up to date by a future revision.

We believe that this use of the SDGP is the best way to satisfy the requirements of our Constitution consistent with the requirements of sound planning. If events indicate, however, that this new direction given to the Mount Laurel doctrine is somehow inadequate, or needs further revision or refinement, the Court remains open to any party to advance such a contention. While there are numerous advantages to certainty in this area, it is much too complex to be dogmatic about almost anything. Flexibility is needed here, for our work is partially legislative in character. We do it not by choice, but because of our understanding of our Constitution and the Legislature’s failure to act.

As noted before, all municipalities’ land use regulations will be required to provide a realistic opportunity for the construction of their fair share of the region’s present lower income housing need generated by present dilapidated or overcrowded lower income units, including their own. Municipalities located in “growth areas” may, of course, have an obligation to meet the present need of the region that goes far beyond that generated in the municipality itself; there may be some municipalities, however, in growth areas where the portion of the region’s present need generated by that municipality far exceeds the municipality’s fair share. The portion of the region’s present need that must be addressed by municipalities in growth areas *244will depend, then, on conventional fair share analysis, some municipality’s fair share being more than the present need generated within the municipality and in some cases, less. In non-growth areas, however (limited growth, conservation, and agricultural), no municipality will have to provide for more than the present need generated within the municipality, for to require more than that would be to induce growth in that municipality in conflict with the SDGP.

It is our intention by this decision generally to channel the entire prospective lower income housing need in New Jersey into “growth areas.” It is clear that that is what the SDGP intends and there is nothing to indicate that those areas are not more than sufficient to accommodate such growth for the foreseeable future.17

The SDGP does not purport to apply its growth, limited growth, conservation and agricultural classifications to lands subject to the jurisdiction of the Division of Coastal Resources under the Coastal Area Facility Review Act, N.J.S.A. 13:19-1 to -21, or the Pinelands Commission under the Pinelands Protection Act, N.J.S.A. 13:18A-1 to -29. While the maps that form part of the SDGP and some of its text suggest such classifica*245tion,18 it is clear that the Division intended to defer to these operational state planning agencies. In referring to the Coastal Area Facilities Review Act and its administration, the SDGP notes that “the designations and growth patterns recommended by the Department of Environmental Protection ... have been incorporated.” SDGP at 43. As for the Pinelands, it states that “[t]he Concept Map classifies the entire Pinelands Area as Conservation. When an approved management plan has been completed by the Pinelands Commission, appropriate amendments to the Guide Plan and the Concept Map will be made.” SDGP at 67.

In the case of both the Pinelands and Coastal Areas, state agencies have been created with direct responsibility and power to classify areas for purposes of encouraging and discouraging growth, indeed with power to prohibit it completely. Our review of the present plans and policies of both the Division of Coastal Resources and the Pinelands Commission indicates that their classification process is well-advanced and most complex. Since the relationship of the work of these agencies, and of their classification of the area subject to their jurisdiction, to the SDGP was neither argued nor briefed, we decline to decide in this litigation which municipalities within their bounds are subject to the Mount Laurel doctrine.

Trial judges in Mount Laurel cases involving municipalities located either in the Pinelands or the coastal zone shall consider in detail the classification systems involved to determine whether imposition of the Mount Laurel doctrine would be consistent *246with the regional planning goals of the agency, and whether the constitutional obligation will under any circumstances override those goals.

We realize that the construction of lower income housing in these coastal and pinelands areas where the Mount Laurel doctrine does apply will require approvals of agencies in addition to the municipality, that the approval procedure can be difficult and time consuming, and that what may appear as a realistic opportunity in a zoning ordinance may turn out not to be so by virtue of the position or regulations of the Division or Commission. These complexities necessarily arise from this double-layered system of municipal and state agency regulation designed to assure the greatest protection and coherence in the development of these highly sensitive environmental areas.19

Given the remedial function now assigned by this Court to the SDGP in determining the imposition of the Mount Laurel obligation, its admission in evidence in particular litigation poses no evidentiary problems whatsoever. It is admissible subject to Evid.R. 67 which requires “[authentication of the original or a copy of a writing ... before it may be received in evidence.” See also Evid.R. 63(15). Once the document is thus proven to be the authentic State Development Guide Plan, no hearsay problems exist. The document is not introduced to prove the truth of any fact contained in it. It is introduced because the Plan itself is a fact. By virtue of our opinion today, the State Development Guide Plan’s delineation of growth areas will in *247most cases conclusively determine the existence and location for the imposition of the Mount Laurel obligation.20

Our use of the State Development Guide Plan for the purpose of determining where Mount Laurel applies does not, of course, guarantee that lower income housing will be constructed in the future solely pursuant to this comprehensive rational plan for the development of New Jersey. It simply tends to assure that the judiciary will not contribute to irrational development, diseordánt with the state’s own vision of its future, by encouraging it in areas that the state has concluded should not be developed, areas more suitable for other purposes, or by inadvertently leading municipalities to encourage lower income housing in such areas. There is nothing, however, that prevents municipalities from encouraging growth, including residential growth, in areas designated by the SDGP as limited growth, agricultural or conservation areas. Uninhibited by any statutory restrictions, municipalities may, for a variety of reasons, plan their future in a manner totally inconsistent with the state’s plan, bringing factories, retail shopping centers, large-scale housing developments, including lower income housing, into areas where their presence runs completely counter to the objectives of the SDGP. Except for protective legislation (such as that pertaining to the Pinelands and certain coastal areas) limited to particular ecologically sensitive areas, the state has imposed no proscriptions against development. While conformi*248ty of the constitutional obligation to the design of the Plan unquestionably advances the state’s purpose, the absence of such proscriptions against development may, in the long run, undermine the regional planning objectives of the SDGP whether we limit the Mount Laurel obligation to growth areas or not.21

C. Calculating Fair Share

The most troublesome issue in Mount Laurel litigation is the determination of fair share. It takes the most time, produces the greatest variety of opinions, and engenders doubt as to the meaning and wisdom of Mount Laurel. Determination of fair share has required resolution of three separate issues: identifying the relevant region, determining its present and prospective housing needs, and allocating those needs to the municipality or municipalities involved. Each of these issues produces a morass of facts, statistics, projections, theories and opinions sufficient to discourage even the staunchest supporters of Mount Laurel. The problem is capable of monopolizing counsel’s time for years, overwhelming trial courts and inundating reviewing courts with a record on review of superhuman dimensions.

*249We have had enough experience with Mount Laurel litigation to warrant procedural modifications designed, over a period of time, to simplify these determinations. The procedural modification provided in this opinion (confining all Mount Laurel litigation to a limited number of judges) is well within conventional judicial techniques.

The first hint of the troubles ahead in determining fair share is found in Mount Laurel I:

The composition of the applicable “region” will necessarily vary from situation to situation and probably no hard and fast rule will serve to furnish the answer in every case. Confinement to or within a certain county appears not to be realistic, but restriction within the boundaries of the State seems practical and advisable. [67 N.J. at 189-90],

The concurrence was more specific. Referring to the three issues mentioned above, it noted that:

[A]ll of these steps involve difficult factual determinations based upon expert testimony and statistical evidence. It may well be appropriate for the court to appoint independent experts or consultants for its assistance or to invite participation by the Department of Community Affairs as amicus curiae.... [C]on-flicting decisions within a given region would be highly undesirable .... ” [67 N.J. at 216 (Pashman, J., concurring) (citations omitted) ].

By the time we reached Madison, the full import of what sounded like benign flexibility in Mount Laurel I had become apparent. Ruling that the trial court need not make findings either of a specific region or of the precise fair share of a municipality within that region, we noted:

Firstly, numerical housing goals are not realistically translatable into specific substantive change in a zoning ordinance by any technique revealed to us by our study of the data before us. There are too many imponderables between a zone change and the actual production of housing on sites as zoned, not to mention the production of a specific number of lower cost units in a given period of time. Municipalities do not themselves have the duty to build or subsidize housing. Secondly, the breadth of approach by the experts to the factor of the appropriate region and to the criteria for allocation of regional housing goals to municipal “subregions” is so great and the pertinent economic and sociological considerations so diverse as to preclude judicial dictation or acceptance of any one solution as authoritative. For the same reasons, we would not mandate the formula approach as obligatory on any municipality seeking to correct a fair share deficiency. [72 N.J. at 499],

Again:

The formulation of a plan for the fixing of the fair share of the regional need for lower income housing attributable to a particular developing municipality, *250although clearly envisaged in Mount Laurel, 67 N.J. at 162, 189-190, involves highly controversial economic, sociological and policy questions of innate difficulty and complexity. Where predictive responses are called for they are apt to be speculative or conjectural. [Madison, 72 N.J. at 533],

A reading of Madison with its comprehensive coverage not only of the facts and expert opinions in that case but also of the voluminous literature with its wealth of contrasting doctrine and approaches is still the most convincing argument now, as it was to us then, to modify the doctrine. Confronted with the overwhelming demonstration provided by Madison that the factors that make up the Mount Laurel doctrine are simply too complex and too interwoven with social, political and economic issues to permit judicial resolution, we ruled, bluntly, that they need not be resolved. We pointed out, not once, but on numerous occasions, that the problem is better addressed by others. (“[T]he basic underlying social problem is far better addressed by administrative action than litigation.” 72 N.J. at 499 n. 5. Distinguishing fair share determinations by administrative planning agencies from courts adjudicating Mount Laurel disputes, we also noted that “[t]he correlative disadvantages of a court adjudicating an individual dispute are obvious.” Id. at 533.)

One possible resolution of the fair share issue by an administrative agency was contained in “A Revised Statewide Housing Allocation Report for New Jersey” promulgated by the New Jersey Division of State and Regional Planning in May 1978. As noted in Madison, a preliminary draft of the HAR had been circulated, and but for its status as “tentative and subject to further public hearings and review,” we suggested we might regard the regions and fair share allocations contained therein as “meriting prima facie judicial acceptance.” Madison, 72 N.J. at 538.22 Subsequent to our decision in Madison the report was updated (although still designated for public review and comment) to May 1978. Without in any way implying support for *251or approval of the legitimacy, method or results of that report, it was the kind of administrative action that arguably provided a means not only for resolving the litigated issues, but for achieving a much more substantial degree of voluntary compliance with Mount Laurel.

On May 4,1982, the Executive Orders that provided the legal justification for the HAR (Executive Order No. 35, April 2,1976, and No. 46, December 8, 1976) were rescinded by Executive Order No. 6. Any regulations adopted and promulgated under Executive Orders No. 35 and 46 were to be null and void. While the basis for such rescission (“said plan has proven to be inadequate to effectively meet the housing needs of the citizenry of New Jersey”; the orders “have proven inadequate and ineffective in meeting their stated goal”; Executive Order No. 6) did not explicitly foreclose use of the HAR in Mount Laurel litigation, we believe such use would not be in keeping with the spirit of the Governor’s Executive Order, which is to render the HAR a nullity. Since the Executive’s authorization and approval provided the legitimacy for the use of such report in Mount Laurel litigation, our authorization of its further use after that rescission would be clearly inappropriate.

With the rescission of Executive Orders 35 and 46 and the consequent unavailability of the HAR for Mount Laurel litigation, we remain confronted with precisely the same problem we faced in Madison. In that ease, it was the determination of fair share that was disturbing, a determination that was comprised of the region, its need, and the allocation of that need. One must necessarily have some humility, based upon the past eight years of experience with Mount Laurel, in expressing confidence about its practical effect, but of this we are confident: our approach in Madison does not work and is not likely to.

When we relieved the parties and the court of the obligation to determine with precision the region, its need, and the fair share of the municipality, we underestimated the pressures that weigh against lower income housing. Given those pressures, the *252test of whether the ordinance in substance provided some opportunity for either lower income or least-cost housing, and whether bona fide efforts had been made to minimize cost-generating requirements in a reasonable area of the municipality, has proven insufficient. The temptation for municipalities after our decision in Madison to ignore the Mount Laurel obligation or to provide the absolute minimum of apparently realistic opportunity for some lower income housing apparently became irresista-ble. Some of its results are before us today. Trial courts interpreted Madison as shifting the burden of compliance from the judiciary to the municipality and looked sympathetically on ordinances that arguably constituted a bona- fide effort to comply. Sometimes, when the litigation, was concluded, no one would know what the fair share of that municipality was, for no one had been required to determine it. There was no standard that municipalities could apply if they wanted to comply. This recognition of the complexities of the issue and the consequent reformulation of the rule in Madison unfortunately provided no relief from the complexities of Mount Laurel litigation, for parties could, and often did, continue to prove region, need, and fair share with the same profusion of facts and expert opinions but without knowing whether the court would regard the evidence as persuasive or even relevant.

In summary, in spite of our intentions, Madison has led to little but a sigh of relief from those who oppose Mount Laurel.

The situation must be remedied. In the absence of executive or legislative action to satisfy the constitutional obligation underlying Mount Laurel, the judiciary has no choice but to enforce it itself. Enforcement, to be effective, will require firm judicial management.

The difficulty in making the Mount Laurel obligation a reality is perhaps unique, for it consists of determining the obligation as much as enforcing it. Litigation that at its conclusion leaves everyone in doubt as to just what the constitutional obligation was and just how it was complied with has, in the aggregate, *253the effect of leaving the constitutional obligation itself in doubt. Until the regions of New Jersey, their present and prospective lower income housing needs, and the allocation of those needs among all of the municipalities of the state charged with the Mount Laurel obligation are determined, uncertainty will prevail, and the weakness of the constitutional doctrine will continue. We intend today to begin a process aimed at ultimately eliminating the uncertainty that surrounds these issues.

The restriction of Mount Laurel litigation to three judges should simplify and perhaps, in time, substantially eliminate the issues of “region” and “regional need” from litigation. Of the three major issues in this area, their determination is most susceptible to judicial treatment. We implied as much in Madison when we contrasted “[t]he technical details of the basis for fair share allocations of regional goals among municipalities, pertaining as they do to an area of considerable complexity and theoretical diversity ...” with the “gross regional goal,” 72 N.J. at 536, and when we noted that “[tjhere is much greater diversity among the experts [concerning “fair share allocation”] than in relation to determining pertinent regions.” Id. at 541. We also noted that the determination of region was more important in achieving the goals of Mount Laurel than the fair share allocation itself (“harm to the objective of securing adequate opportunity for lower income housing is less likely from imperfect allocation models than from undue restriction of the pertinent region,” id.).

As noted above, following the release of this opinion the Chief Justice, with the approval of the Court, will name three judges (the number to be changed if necessary) who will thereafter handle all Mount Laurel litigation (except some of the cases before us now and except for other pending matters, as to which transfer to those three judges will be at the discretion of the Chief Justice). Each of these three judges will be exclusively responsible for a particular area of the state; any litigation challenging a land use regulation of a municipality in *254that judge’s area on Mount Laurel grounds shall be assigned to that judge.

We anticipate that after several cases have been tried before each judge, a regional pattern for the area for which he or she is responsible will emerge. Ultimately a regional pattern for the entire state will be established, as will a fairly consistent determination of regional needs on both an area and statewide basis. Given that only three judges are involved, it is also not unreasonable to assume that the method for determining the municipality’s fair share of the regional need will be consistent within the judge’s area and tend to promote consistency throughout the state.

The determination of region and regional need by any of these judges shall be presumptively valid as to all municipalities included in the region unless the judge hearing the matter indicates otherwise for reasons stated in his or her decision.23 Given the importance of these determinations, municipalities not named as parties- may attempt to intervene or the court may require their joinder if, all things considered, it is thought advisable that such a municipality be bound by the determination even though such joinder may complicate the litigation. The extent of such litigation, whether non-party municipalities should be allowed to participate, and whether they should be joined as parties, shall all be within the discretion of the court, who will be better able to balance the various considerations involved. While it is possible that many municipalities may seek to enter such litigation, we believe that as a practical matter most will be content to abide by litigation in which others are involved. There may be exceptions, but in most cases municipalities will realize that the determination of region will probably not be substantially affected by their participation in the litigation; that while the variations in the definition and deline*255ation of the region are infinite, its general contours are fairly predictable; and that the same applies to some extent to regional need.

In short we foresee that within several years the fair share question will be confined to the allocation issue. Our use of the SDGP should end practically all disputes over the existence of the Mount Laurel obligation and, in relatively short time, adjudication by the three judges should end most disputes over region and regional need. In practically all cases the only issue (other than the adequacy of the housing opportunity provided by the ordinance) that may require serious litigation is a particular municipality’s fair share of that need. And even' as to that issue, the housing allocation methodologies previously adopted should simplify it considerably. It is possible, of course, that the presumptively valid region and regional need determinations may be seriously contested, but we doubt it. Except for municipalities on the outer edges of a region, the regional determinations are not likely to be significantly varied by the judges, given the desirability of consistency and predictability; only the strongest evidence is likely to lead to substantial change. If the importance given to the initial litigation by the rule of presumptive validity results in a case more complex and with more parties than it might otherwise have been, we believe the ultimate potential gains are well worth it.

As implied above in our summary of rulings (Ruling 4, supra at 216-217), additional benefits may follow from this procedure. Each of the three judges will become more and more adept in handling Mount Laurel litigation, in defining and narrowing the issues early in the litigation, in expediting the case, in determining when an expert should be appointed by the court and when a master should be named to aid the municipality in its revision of a zoning ordinance, and ultimately in devising remedies suitable for the complete redress of exclusionary zoning.

*256Such Mount Laurel litigation shall proceed using the previously accepted definitions and methods of determining region, regional need, and fair share. For example, we indicated in Madison our general approval of Judge Furman’s definition of region (72 N.J. at 537), slightly modified, as “that general area which constitutes, more or less, the housing market area of which the subject municipality is a part, and from which the prospective population of the municipality would substantially be drawn, in the absence of exclusionary zoning.” Id. at 543. A trial court’s acceptance of any variant of this definition should be premised on special circumstances. Certainly in its ultimate determination the court may consider the factors mentioned in Justice Pashman’s concurring opinion in Mount Laurel I, 67 N.J. at 215 n. 16. We will not attempt here to provide any further guidance for the determination of regional need, but leave that to the experts, including the experts appointed by the trial courts pursuant to our opinion.

As for fair share, however, we offer some suggestions.24 Formulas that accord substantial weight to employment opportunities in the municipality, especially new employment accompanied by substantial ratables, shall be favored; formulas that have the effect of tying prospective lower income housing needs to the present proportion of lower income residents to the total population of a municipality shall be disfavored; formulas that have the effect of unreasonably diminishing the share because of a municipality’s successful exclusion of lower income housing in the past shall be disfavored.

In determining fair share, the court should decide the proportion between low and moderate income housing unless *257there are substantial reasons not to do so. The provisions and devices needed to produce moderate income housing may fall short of those needed for lower. Since there are two fairly distinct lower income housing needs, an effort must be made to meet both.

The proportion between the two is, inevitably, a matter for expert testimony. It will depend, as does the fair share itself, on a complex mix of factors. We note, without comment, the trial court’s use in Carteret, for this purpose, of the actual proportion between the low and moderate income population of the county. 142 N.J.Super. at 36-37: The point here is that it is an issue that should be addressed in passing on the adequacy of land use regulations (and revisions thereof) as well as builder’s remedies. Cf. Madison, 72 N.J. at 549 & n. 48, 551 (builder’s remedy conditioned on allocation of “at least 20% of the units to low or moderate income families”; in fact, allocation would be satisfied by moderate income units only).

The ultimate outcome of such litigation in most cases shall be a determination by the court of a precise region, a precise regional present and prospective need, and a precise determination of the present and prospective need that the municipality is obliged to design its ordinance to meet.

We recognize that the tools for calculating present and prospective need and its allocation are imprecise and further that it is impossible to predict with precision how many units of housing will result from specific ordinances. What is required is the precision of a specific area and specific numbers. They are required not because we think scientific accuracy is possible, but because we believe the requirement is most likely to achieve the goals of Mount Laurel.

While it would be simpler in these cases to calculate a municipality’s fair share by determining its own probable future population (or some variant thereof), such a method would not be consistent with the constitutional obligation (although it is a factor that could be considered in a fair share calculation in the

*258absence of other proof). Municipal population projections are based on many factors, but in no case that we know of do they include a value judgment that such municipality should bear its fair share of the region’s lower income housing need. In fact, in most cases, we believe, one of the factors necessarily involved in such municipal population projections is the prior and probable future effect of the municipality’s exclusionary zoning. If, because of that exclusionary zoning, a suburban municipality with substantial developable land has a very, very small probable growth as shown by the most reliable population projections (resulting in part from its very small past growth caused by exclusionary zoning), it should not be allowed to evade its obligation by basing its fair share of the lower income housing need on that small projected population growth. On the other hand, when that municipality is considered as part of the region and the region’s population growth is projected, a value judgment is made, based upon the Mount Laurel obligation, that may result in a substantially greater fair share for that municipality and indeed may have the effect of changing what would otherwise be the population projection for that municipality.

It may be that the overall population projections for the State of New Jersey and for its various regions are somewhat affected by the aggregate impact of exclusionary zoning — that is something for experts to determine. Even so, when gross population projections are used for a region, it is more likely that the total lower income housing need will be included and much more likely that whatever lower housing income need is in fact included will be distributed fairly, not in accordance with prior patterns of exclusionary zoning but in accordance with suitability for such housing.

D. Meeting the Mount Laurel Obligation

1. Removing Excessive Restrictions and Exactions

In order to meet their Mount Laurel obligations, municipalities, at the very least, must remove all municipally created *259barriers to the construction of their fair share of lower income housing. Thus, to the extent necessary to meet their prospective fair share and provide for their indigenous poor (and, in some cases, a portion of the region’s poor), municipalities must remove zoning and subdivision restrictions and exactions that are not necessary to protect health and safety.25

It may be difficult for a municipality to determine how to balance the need to reduce the costs of its regulations against the need to adequately protect health and safety, just as it may be difficult for a court to determine when a municipality has reduced these costs enough. There are, however, relatively objective guides that can help both the municipality and the court. Particularly helpful, though in no way conclusive as to what the minimum standards should be in a particular community, are the Department of Housing and Urban Development’s Minimum Property Standards and the suggestions as to minimum zoning and subdivision standards made by the Rutgers Center for Urban Policy Research in Housing Costs, supra at 212 n. 6. With these and other such guides, plus specific evidence submitted by the parties, we believe that a court can determine whether municipally-imposed housing costs have been sufficiently reduced.

Once a municipality has revised its land use regulations and taken other steps affirmatively to provide a realistic oppor*260tunity for the construction of its fair share of lower income housing, the Mount Laurel doctrine requires it to do no more. For instance, a municipality having thus complied, the fact that its land use regulations contain restrictive provisions incompatible with lower income housing, such as bedroom restrictions, large lot zoning, prohibition against mobile homes, and the like, does not render those provisions invalid under Mount Laurel. Obviously, if they are otherwise invalid — for instance if they bear no reasonable relationship to any legitimate governmental goal — they may be declared void on those other grounds. But they are not void because of Mount Laurel under those circumstances. Mount Laurel is not an indiscriminate broom designed to sweep away all distinctions in the use of land. Municipalities may continue to reserve areas for upper income housing, may continue to require certain community amenities in certain areas, may continue to zone with some regard to their fiscal obligations: they may do all of this, provided that they have otherwise complied with their Mount Laurel obligations.

2. Using Affirmative Measures

Despite the emphasis in Mount Laurel I on the affirmative nature of the fair share obligation, 67 N.J. at 174, the obligation has been sometimes construed (after Madison) as requiring in effect no more than a theoretical, rather than realistic, opportunity. As noted later, the alleged realistic opportunity for lower income housing in Mount Laurel II is provided through three zones owned entirely by three individuals. There is absolutely no assurance that there is anything realistic in this “opportunity”: the individuals may, for many different reasons, simply not desire to build lower income housing. They may not want to build any housing at all, they may want to use the land for industry, for business, or just leave it vacant. It was never intended in Mount Laurel I that this awesome constitutional obligation, designed to give the poor a fair chance for housing, be satisfied by meaningless amendments to zoning or other ordinances. “Affirmative,” in the Mount *261 Laurel rule, suggests that the municipality is going to do something, and “realistic opportunity” suggests that what it is going to do will make it realistically possible for lower income housing to be built. Satisfaction of the Mount Laurel doctrine cannot depend on the inclination of developers to help the poor. It has to depend on affirmative inducements to make the opportunity real.

It is equally unrealistic, even where the land is owned by a developer eager to build, simply to rezone that land to permit the construction of lower income housing if the construction of other housing is permitted on the same land and the latter is more profitable than lower income housing. One of the new zones in Mount Laurel provides a good example. The developer there intends to build housing out of the reach of the lower income group. After creation of the new zone, he still is allowed to build such housing but now has the “opportunity” to build lower income housing to the extent of 10 percent of the units. There is absolutely no reason why he should take advantage of this opportunity if, as seems apparent, his present housing plans will result in a higher profit. There is simply no inducement, no reason, nothing affirmative, that makes this opportunity “realistic.” For an opportunity to be “realistic” it must be one that is at least sensible for someone to use.

Therefore, unless removal of restrictive barriers will, without more, afford a realistic opportunity for the construction of the municipality’s fair share of the region’s lower income housing need, affirmative measures will be required.26

*262There are two basic types of affirmative measures that a municipality can use to make the opportunity for lower income housing realistic: (1) encouraging or requiring the use of available state or federal housing subsidies, and (2) providing incentives for or requiring private developers to set aside a portion of their developments for lower income housing. Which, if either, of these devices will be necessary in any particular municipality to assure compliance with the constitutional mandate will be initially up to the municipality itself. Where necessary, the trial court overseeing compliance may require their use. We note again that least-cost housing will not ordinarily satisfy a municipality’s fair share obligation to provide low and moderate income housing unless and until it has attempted the inclusionary devices outlined below or otherwise has proven the futility of the attempt.

a. Subsidies

Because the kinds of lower income housing subsidies available are subject to change — and have in fact changed *263often — it is more important to establish the municipality’s general Mount Laurel obligation concerning subsidies than its required role as to any particular existing subsidy. The importance of defining that obligation may depend at any particular time on the then extent and impact of available subsidies; if anything, the quantity of housing subsidies varies even more than the kind. For example, the amount of lower income housing subsidies now available is substantially less than several years ago, and there is no indication that subsidies for lower income housing construction are likely to increase in the near future. They are, nevertheless, apparently a permanent part of the housing scene; the long-term importance of defining the municipality’s Mount Laurel obligation in relation to such subsidies is that the construction of lower income housing is practically impossible without some kind of governmental subsidy.

While Mount Laurel I did not come to grips with this issue, its clear import was that the Court at that time expected Mount Laurel, in addition to adopting “appropriate zoning ordinance amendments,” to take “whatever additional action encouraging the fulfillment of its fair share of the regional need for low and moderate income housing [as might be] necessary and advisable.” 67 N.J. at 192. The Court noted that Mount Laurel — at the least — had a “moral obligation” to provide lower income housing through a local housing agency. Simply facilitating subsidies granted by other levels of government is municipal action of a significantly lesser dimension, well within that contemplated by this Court when it noted that “[s]hould Mount Laurel not perform as we expect, further judicial action may be sought.” Id. at 192. The Court had previously recognized the necessity for such subsidies and optimistically stated its expectation that housing would become a reality “through private or public enterprises, or both.” Id. at 188 n. 21. Whatever the Court may have meant then, Madison made it clear that not only was public housing a “moral obligation” at most (rather than “at least”), but that some “devices” — which may in fact be absolutely essential if federal or state subsidies are to be forthcoming— *264“must be summarily rejected,” referring to tax concessions, among other devices. Oakwood v. Madison, 72 N.J. at 546. We were not, however, directly faced in Madison with the extent of a municipality’s obligation, if any, to facilitate lower income housing subsidies available from other levels of government.

The implication of the observation that lower income housing cannot be built without subsidies is that if the Mount Laurel principle requires municipalities to provide a realistic opportunity for such housing through their land use regulations but leaves them free to prevent subsidies through non-action, that obligation is a charade. Mount Laurel was never intended to produce the perfect model of a just zoning ordinance; it was intended to provide a realistic opportunity for the construction of lower income housing.

We do not suggest that a municipality would be required to create a housing authority to meet its Mount Laurel obligation. We do, however, expect municipal officials in appropriate cases to do more than pass land use regulations conforming to Mount Laurel I. Where appropriate, municipalities should provide a realistic opportunity for housing through other municipal action inextricably related to land use regulations.

On occasion, what is needed to obtain a subsidy may be as simple as a “resolution of need” stating that “there is a need for moderate income housing” in the municipality. N.J.S.A. 55:14J-6(b). In addition to the “resolution of need,” the most important federal program for providing lower income housing subsidies (the section 8 low and moderate income housing program; 42 U.S.C. § 1437f (1982 Supp.)) requires in New Jersey, as a practical matter, that the municipality grant tax abatements to developers. See N.J.S.A. 55:14J — 8(f).27

*265In evaluating the obligation that the municipality might be required to undertake to make a federal or state subsidy available to a lower income housing developer, the fact that some financial detriment may be incurred is not dispositive. Satisfaction of the Mount Laurel obligation imposes many financial obligations on municipalities, some of which are potentially substantial. By contrast, a tax abatement for a low or moderate income housing project will have only a minimal effect on the public fisc. Thus viewed, the asserted fiscal reasons justifying the failure to provide a tax abatement may be nothing more than a red herring. The direct and immediate financial impact of a tax abatement agreement between the municipality and the developer may be unimportant when compared with increases in municipal and school district costs caused by the advent of lower income housing. The trial court in a Mount Laurel case, therefore, shall have the power to require a municipality to cooperate in good faith with a developer’s attempt to obtain a subsidy and to require that a tax abatement be granted for that purpose pursuant to applicable New Jersey statutes where that abatement does not conflict with other municipal interests of greater importance.

b. Inclusionary Zoning Devices

There are several inclusionary zoning techniques that municipalities must use if they cannot otherwise assure the construction of their fair share of lower income housing. Although we will discuss some of them here, we in no way intend our list to be exhaustive; municipalities and trial courts are encouraged to *266create other devices and methods for meeting fair share obligations.28

The most commonly used inclusionary zoning techniques are incentive zoning and mandatory set-asides. The former involves offering economic incentives to a developer through the relaxation of various restrictions of an ordinance (typically density limits) in exchange for the construction of certain amounts of low and moderate income units. The latter, a mandatory set-aside, is basically a requirement that developers include a minimum amount of lower income housing in their projects.

(i) Incentive Zoning

Incentive zoning is usually accomplished either through a sliding scale density bonus that increases the permitted density as the amount of lower income housing provided is increased, or through a set bonus for participation in a lower income housing program. See Fox & Davis, 3 Hastings Const. L.Q. 1015, 1060-62 (1977).

Incentive zoning leaves a developer free to build only upper income housing if it so chooses. Fox and Davis, in their survey of municipalities using inclusionary devices, found that while developers sometimes profited through density bonuses, they were usually reluctant to cooperate with incentive zoning programs; and that therefore those municipalities that relied exclu*267sively on such programs were not very successful in actually providing lower income housing. Id. at 1067.29

Sole reliance on “incentive” techniques (or, indeed, reliance exclusively on any one affirmative device) may prove in a particular case to be insufficient to achieve compliance with the constitutional mandate.

(i) Mandatory Set-Asides

A more effective inclusionary device that municipalities must use if they cannot otherwise meet their fair share obligations is the mandatory set-aside.30 According to the De*268partment of Community Affairs, as of 1976 there were six municipalities in New Jersey with mandatory set-aside programs, which varied from a requirement that 5 percent of developments in a certain zone be composed of low and moderate income units (Cherry Hill, Camden County) to a requirement that between 15 and 25 percent of all PUDs be reserved for low and moderate income housing (East Windsor, Mercer County). Housing Handbook, supra at 201 n. 2 at 12-16.31 Apparently, judging from the Handbook itself and from responses to our inquiries at oral argument, lower income housing is in fact being built pursuant to these mandatory requirements.

The use of mandatory set-asides is not without its problems: dealing with the scarcity of federal subsidies, maintaining the rent or sales price of lower income units at lower income levels over time, and assuring developers an adequate return on their investments. Fox and Davis found that the scarcity of federal subsidies has greatly undermined the effectiveness of mandatory set-asides where they are triggered only when a developer is able to obtain such subsidies. Fox & Davis, supra, 3 Hastings Const. L.Q. at 1065-66. Where practical, a municipality should use mandatory set-asides even where subsidies are not available.32

*269As several commentators have noted, the problem of keeping lower income units available for lower income people over time can be a difficult one. See id. at 1034-36; Housing Handbook, supra at 201 n. 2, at 11; Eleven, supra at 266 n. 28, at 1445. Because a mandatory set-aside program usually requires a developer to sell or rent units at below their full value so that the units can be affordable to lower income people, the owner of the development or the initial tenant or purchaser of the unit may be induced to re-rent or re-sell the unit at its full value.

This problem, which municipalities must address in order to assure that they continue to meet their fair share obligations, can be dealt with in two ways. First, the developer can meet its mandatory quota of lower income units with lower cost housing, such as mobile homes or “no-frills” apartments, which may be affordable by lower income families at close to the units’ market value. The other, apparently more common, approach for dealing with the re-sale or re-rent problem is for the municipality to require that re-sale or re-rent prices be kept at lower income levels. For example, the Cherry Hill ordinance requires that there be “regulations which reasonably assure that the dwelling units be occupied by [lower income persons].” Similarly, the Franklin Township (Somerset County) ordinance requires that a developer demonstrate in writing that the rentals and prices of lower income units remain low enough to benefit lower income persons. Housing Handbook, supra at 201 n. 2, at 11. These provisions appear to place the burden on the developer to devise the precise mechanism for maintaining the units at lower income levels.

A more sophisticated approach, considered by Princeton Township, would have established a two part control mechanism. First, disposition covenants would have been created for all the lower income units binding the owners and renters of such units to sell or rent only at lower income levels. Second, a Public Trust would have been created whose trustees would have administered the covenants and determined what would be *270“lower income levels” over time. See Housing Proposal, supra at 266 n. 28, at 68-70.

Mandatory set-asides can be rendered ineffective if a developer builds all its conventional units first and then reneges on the obligation to build the lower income units. To avoid this problem, municipalities and courts should require that a developer phase-in the lower income units as the development progresses. That is, if a developer is required to set aside 20 percent of a development for lower income units, 20 percent of each stage of the development should be lower income, to the extent this is practical.

In addition to the mechanisms we have just described, municipalities and trial courts must consider such other affirmative devices as zoning substantial areas for mobile homes and for other types of low cost housing and establishing maximum square footage zones, i.e., zones where developers cannot build units with more than a certain footage or build anything other than lower income housing or housing that includes a specified portion of lower income housing. In some eases, a realistic opportunity to provide the municipality’s fair share may require over-zoning, i.e., zoning to allow for more than the fair share if it is likely, as it usually is, that not all of the property made available for lower income housing will actually result in such housing.

Although several of the defendants concede that simply removing restrictions and exactions is unlikely to result in the construction of lower income housing, they maintain that requiring the municipality to use affirmative measures is beyond the scope of the courts’ authority. We disagree. Before directly answering their objections, we note, as we did in Robinson v. Cahill, 69 N.J. 133 (1975), that unless an “appropriate remedy” is formulated to “redress a violation of [constitutional] rights,” our Constitution “embodies rights in a vacuum, existing only on paper.” Id. at 147, quoting Cooper v. Nutley Sun Printing Co., Inc., 36 N.J. 189, 197 (1961). See also Swann v. Charlotte-Meck- *271 lenburg Bd. of Ed., 402 U.S. 1, 15, 91 S.Ct. 1267, 1275, 28 L.Ed.2d 554 (1971). If it is plain, and it is, that unless we require the use of affirmative measures the constitutional guarantee that protects poor people from municipal exclusion will exist “only on paper,” then the only “appropriate remedy” is the use of affirmative measures.

The specific contentions are that inclusionary measures amount to a taking without just compensation and an impermissible socio-economic use of the zoning power, one not substantially related to the use of land. Reliance is placed to some extent on Board of Supervisors v. DeGroff Enterprises, Inc., 214 Va. 235, 198 S.E.2d 600 (1973), to that effect. We disagree with that decision. We now resolve the matter that we left open in Madison, 72 N.J. at 518-19. We hold that where the Mount Laurel obligation cannot be satisfied by removal of restrictive barriers, inclusionary devices such as density bonuses and mandatory set-asides keyed to the construction of lower income housing, are constitutional and within the zoning power of a municipality.

In Taxpayers Ass’n of Weymouth Twp. v. Weymouth Twp., 80 N.J. 6 (1976), we upheld a zoning ordinance that allowed mobile homes in a particular zone, limited however to “elderly” persons or families. Our decision was based, in part, on the conclusion that meeting the special housing needs of the elderly served the general welfare, and that the restriction of this use to a mobile home district satisfied the requirement that the zoning ordinance be related to the physical use of the land. See also Shepard v. Woodland Twp. Comm. & Planning Bd., 71 N.J. 230 (1976); DeSimone v. Greater Englewood Housing Corp. No. 1, 56 N.J. 428 (1970).33

*272The rationale of Weymouth could, under appropriate circumstances, sustain a zoning ordinance that restricted a particular district exclusively for mobile homes for the elderly (the actual restriction allowed other uses). If that is permissible, then the comparable special need of lower income families for housing, and its impact on the general welfare, could justify a district limited to such use and certainly one of lesser restriction that requires only that multi-family housing within a district include such use (the equivalent of a mandatory set-aside). Since the objective here goes beyond serving the special needs of a particular class of citizens for the general welfare and extends to the fulfillment of a constitutional obligation, the constitutionality of such devices, and the power of the municipality to impose them, is even clearer.

The contention that generally these devices are beyond the municipal power because they are “socio-economic” is particularly inappropriate. The very basis for the constitutional obligation underlying Mount Laurel is a belief, fundamental, that excluding a class of citizens from housing on an economic basis (one that substantially corresponds to a socio-economic basis) distinctly disserves the general welfare. That premise is essential to the conclusion that such zoning ordinances are an abuse of the zoning power and are therefore unconstitutional.

It is nonsense to single out inclusionary zoning (providing a realistic opportunity for the construction of lower income hous*273ing) and label it “socio-economic” if that is meant to imply that other aspects of zoning are not. Detached single family residential zones, high-rise multi-family zones of any kind, factory zones, “clean” research and development zones, recreational, open space, conservation, and agricultural zones, regional shopping mall zones, indeed practically any significant kind of zoning now used, has a substantial socio-economic impact and, in some cases, a socio-economic motivation. It would be ironic if inclu-sionary zoning to encourage the construction of lower income housing were ruled beyond the power of a municipality because it is “socio-economic” when its need has arisen from the socioeconomic zoning of the past that excluded it.

Looked at somewhat differently, having concluded that the constitutional obligation can sometimes be satisfied only through the use of these inelusionary devices, it would take a clear contrary constitutional provision to lead us to conclude that that which is necessary to achieve the constitutional mandate is prohibited by the same Constitution. In other words, we would find it difficult to conclude that our Constitution both requires and prohibits these measures.

We find the distinction between the exercise of the zoning power that is “directly tied to the physical use of the property,” Madison, 72 N.J. at 517, and its exercise tied to the income level of those who use the property artificial in connection with the Mount Laurel obligation, although it obviously troubled us in Madison,34 The prohibition of this kind of affirm*274ative device seems unfair when we have for so long allowed large lot single family residence districts, a form of zoning keyed, in effect, to income levels. The constitutional obligation itself is not to build three bedroom units, or single family residences on very small lots, or high-rise multi-family apartments, but rather to provide through the zoning ordinance a realistic opportunity to construct lower income housing. All of the physical uses are simply a means to this end. We see no reason why the municipality cannot exercise its zoning power to achieve that end directly rather than through a mass of detailed regulations governing the “physical use” of land, the sole purpose of which is to provide housing within the reach of lower income families. We know of no governmental purpose relating to zoning that is served by requiring a municipality to ingeniously design detailed land use regulations, purporting to be “directly tied to the physical use of the property,” but actually aimed at accommodating lower income families, while not allowing it directly to require developers to construct lower income units. Indirection of this kind has no more virtue where its goal is to achieve that which is permitted — indeed, constitutionally mandated — than it has in achieving that which is prohibited.

3. Zoning for Mobile Homes

As the cost of ordinary housing skyrockets for purchasers and renters, mobile homes become increasingly important as a source of low cost housing. The evidence clearly supports a finding that mobile homes are significantly less expensive than site-built housing. See Legislature’s Mobile Home Study Commission, Report and Recommendations (1980) (finding that while *275it would take a household income of at least $21,000 to afford a medium priced site-built home in 1979, an income of $11,700 would be sufficient for a family to afford a mobile home and lot); Housing Handbook, supra, at 32. We agree fully with the finding of Judge Wood in Mount Laurel II that mobile homes are “economically available for persons of low and moderate income.” 161 N.J.Super. at 357. Therefore, subject to the qualifications noted hereafter, we rule that municipalities that cannot otherwise meet their fair share obligations must provide zoning for low-cost mobile homes35 as an affirmative device in their zoning ordinances.

Townships such as Mount Laurel that now ban mobile homes do so in reliance upon Vickers v. Gloucester, 37 N.J. 232 (1962), in which this Court upheld such bans. Vickers, however, explicitly recognized that changed circumstances could require a different result. Id. at 250. We find that such changed circumstances now exist. As Judge Wood found in Mount Laurel II, mobile homes have since 1962 become “structurally sound [and] attractive in appearance.” 161 N.J.Super. at 357. Further, since 1974, the safety and soundness of mobile homes have been regulated by the National Mobile Home Construction and Safety Standards Act, 42 U.S.C. 5401 (1974). Vickers, therefore, is overruled; absolute bans of mobile homes are no longer permissible on the grounds stated in that case. Strong support for this new rule can be found in the report of the Legislature’s Mobile Home Study Commission, Report and Recommendations, at 10-*27611, and in the DCA’s Housing Handbook, supra at 201 n. 2, at 32-33.36

Lest we be misunderstood, we do not hold that every municipality must allow the use of mobile homes as an affirmative device to meet its Mount Laurel obligation, or that any ordinance that totally excludes mobile homes is per se invalid. Insofar as the Mount Laurel doctrine is concerned, whether mobile homes must be permitted as an affirmative device will depend upon the overall effectiveness of the municipality’s attempts to comply: if compliance can be just as effectively assured without allowing mobile homes, Mount Laurel does not command them; if not, then assuming a suitable site is available, they must be allowed.

Insofar as the arbitrariness of a total exclusion is concerned, such conclusion will depend upon the facts and circumstances of each case, regardless of the Mount Laurel doctrine. While the question is not directly before us, there may be municipalities whose development is such that the otherwise inoffensive appearance of a mobile home park may be quite offensive. There may be municipalities whose only vacant land has been legitimately set aside for commercial, industrial or residential uses other than mobile homes, where such planning is quite legitimate. But just as Vickers is hereby overruled to the extent that it held that any developing municipality may totally exclude mobile homes, we hold that such attempt at a total exclusion will have to be justified by the same doctrines that *277would justify a total exclusion of apartment houses, townhouses, or even single family residences. We recognize the propriety of aesthetic considerations in zoning, but the “subjective sensibilities” of present residents are not a sufficient basis for the exclusion. See Vickers, 37 N.J. at 269 (Hall, J., dissenting).

4. Providing “Least Cost” Housing

There may be municipalities where special conditions such as extremely high land costs make it impossible for the fair share obligation to be met even after all excessive restrictions and exactions, i.e., those not essential for safety and health, have been removed and all affirmative measures have been attempted. In such cases, and only in such cases, the Mount Laurel obligation can be met by supplementing whatever lower income housing can be built with enough “least cost” housing to satisfy the fair share. Least cost housing does not, however, mean the most inexpensive housing that developers will build on their own; it does not mean $50,000-plus single family homes and very expensive apartments. Least cost housing means the least expensive housing that builders can provide after removal by a municipality of all excessive restrictions and exactions and after thorough use by a municipality of all affirmative devices that might lower costs. Presumably, such housing, though unaffordable by those in the lower income brackets, will be inexpensive enough to provide shelter for families who could not afford housing in the conventional suburban housing market. At the very minimum, provision of least cost housing will make certain that municipalities in “growth” areas of this state do not “grow” only for the well-to-do.

The form that “least cost” housing will take can vary with the particular characteristics of individual municipalities. Municipalities that must resort to “least cost” housing to meet their Mount Laurel obligations should, if appropriate, zone significant areas for housing that most closely approaches lower income housing, e.g., mobile homes. Furthermore, “overzoning” *278for such housing will greatly increase the likelihood that some of these units, even if not “lower income,” will be affordable by those close to the top of the moderate income bracket. See Madison, 72 N.J. at 519.

It is important for us to emphasize here that unless it meets the stringent “least cost” requirements set out above, middle income housing will not satisfy the Mount Laurel obligation. This is so despite claims by some defendant-municipalities that the provision of such middle income housing will allow less expensive housing to “filter down” to lower income families. The problem with this theory is that the housing that has been built and is now being built in suburbs such as Mount Laurel is rapidly appreciating in .value so that none of it will “filter down” to poor people. Instead, if the only housing constructed in municipalities like Mount Laurel continues to be middle and upper income, the only “filter down” effect that will occur will be that housing on the fringes of our inner cities will “filter down” to the poor as more of the middle class leave for suburbs, thereby exacerbating the economic segregation of our cities and suburbs. See A. Downs, supra at 221 n. 8, at 9-12. Only if municipalities like Mount Laurel begin now to build lower income or least cost housing will some part of their housing stock ever “filter down” to New Jersey’s poorer families. See Madison, 72 N.J. at 513-14 & n. 22.

E. Judicial Remedies

If a trial court determines that a municipality has not met its Mount Laurel obligation, it shall order the municipality to revise its zoning ordinance within a set time period to comply with the constitutional mandate; if the municipality fails adequately to revise its ordinance within that time, the court shall implement the remedies for noncompliance outlined below; and if plaintiff is a developer, the court shall determine whether a builder’s remedy should be granted.

*2791. Builder’s Remedy

Builder’s remedies have been one of many controversial aspects of the Mount Laurel doctrine. Plaintiffs, particularly plaintiff-developers, maintain that these remedies are (i) essential to maintain a significant level of Mount Laurel litigation, and the only effective method to date of enforcing compliance; (ii) required by principles of fairness to compensate developers who have invested substantial time and resources in pursuing such litigation; and (iii) the most likely means of ensuring that lower income housing is actually built. Defendant municipalities contend that even if a plaintiff-developer obtains a judgment that a particular municipality has not complied with Mount Laurel, that municipality, and not the developer, should be allowed to determine how and where its fair share obligation will be met.

In Madison, this Court, while granting a builder’s remedy to the plaintiff appeared to discourage such remedies in the future by stating that “such relief will ordinarily be rare.” 72 N.J. at 551-52 n. 50. Experience since Madison, however, has demonstrated to us that builder’s remedies must be made more readily available to achieve compliance with Mount Laurel. We hold that where a developer succeeds in Mount Laurel litigation and proposes a project providing a substantial amount of lower income housing,37 a builder’s remedy should be granted unless the municipality establishes that because of environmental or *280other substantial planning concerns, the plaintiff’s proposed project is clearly contrary to sound land use planning. We emphasize that the builder’s remedy should not be denied solely because the municipality prefers some other location for lower income housing, even if it is in fact a better site. Nor is it essential that considerable funds be invested or that the litigation be intensive.

Other problems concerning builder’s remedies require discussion. Care must be taken to make certain that Mount Laurel is not used as an unintended bargaining chip in a builder’s negotiations with the municipality, and that the courts not be used as the enforcer for the builder’s threat to bring Mount Laurel litigation if municipal approvals for projects containing no lower income housing are not forthcoming. Proof of such threats shall be sufficient to defeat Mount Laurel litigation by that developer.

It is within the power of trial courts to adjust the timing of builder’s remedies so as to cushion the impact of these developments on municipalities where that impact would otherwise cause a sudden and radical transformation of the municipality. This adjustment is analogous to the phasing-in of the satisfaction of present and prospective need mentioned in Ruling 10, supra at 218-219.

The trial court (and the master, if one is appointed) should make sure that the municipal planning board is closely involved in the formulation of the builder’s remedy. This does not mean that the planning board should be permitted to delay or hinder the project or to reduce the amount of lower income housing required. However, with this caveat, the trial court and master should make as much use as they can of the planning board’s expertise and experience so that the proposed project is suitable for the municipality.

Finally, we emphasize that our decision to expand builder’s remedies should not be viewed as a license for unnecessary *281litigation when builders are unable, for good reason, to secure variances for their particular parcels (as Judge Muir suggested was true in the Chester Township case). Trial courts should guard the public interest carefully to be sure that plaintiff-developers do not abuse the Mount Laurel doctrine. Where builder’s remedies are awarded, the remedy should be carefully conditioned to assure that in fact the plaintiff-developer constructs a substantial amount of lower income housing. Various devices can be used for that purpose, including prohibiting construction of more than a certain percentage of the non-lower income housing until a certain amount of the lower income housing is completed.

2. Revision of the Zoning Ordinance: the Master .

If the trial court determines that a municipality’s zoning ordinance does not satisfy its Mount Laurel obligation, it shall order the defendant to revise it. Unless it is clear that the requisite realistic opportunity can be otherwise provided, the trial court should direct the municipality to incorporate in that new ordinance the affirmative devices discussed above most likely to lead to the construction of lower income housing. The trial court shall order the revision to be completed within 90 days of its original judgment against the municipality. For good cause shown, a municipality may be granted an extension of that time period.

To facilitate this revision, the trial court may appoint a special master to assist municipal officials in developing constitutional zoning and land use regulations.38 The use of such special masters, sometimes called “hybrid” masters, is not un*282common in litigation resulting in some form of institutional change.39 See, e.g., T. Eisenberg & S. Yeazell, “The Ordinary and the Extraordinary in Institutional Litigation,” 93 Harv.L. Rev. 465 (1980); T. Mayo, “Exclusionary Zoning, Remedies, and the Expansive Role of the Court in Public Law Litigation,” 31 Syracuse L.Rev. 755 (1980); “Special Project — The Remedial Process in Institutional Reform Litigation,” 78 Colum.L.Rev. 784, 794 (1978); Berger, “Away from the Court House and Into the Field: The Odyssey of a Special Master,” 78 Colum.L.Rev. 707 (1978); “The Wyatt Case: Implementation of a Judicial Decree Ordering Institutional Change,” 84 Yale L.J. 1338, 1344 (1975). These impartial experts use their skills to help the parties formulate a remedy that will comply with the trial court’s order and supply information that the parties may not have available to them. 78 Colum.L.Rev. at 794. They differ from traditional masters, whose roles are usually limited to serving as fact-finders and supervising procedural tasks, id. at 805, in that special masters work with the parties to devise a remedy that will meet with the court’s approval. Id. at 805-06; 84 Yale L.J. at 1344.

While the appointment of a master is discretionary, we believe that such appointment is desirable in many cases where the court orders a revision of the land use regulations, especially if that revision is substantial. We do not view the appointment of *283a master as punitive in the least; it is not designed to settle scores with recalcitrant municipalities. The point here is that we intend that the appointment of masters be viewed by the court as a readily available device, one to be liberally used. In our view the master is of potential help to all concerned: to the municipality, to the plaintiffs, to the court and counsel. He or she is an expert, a negotiator, a mediator, and a catalyst — a person who will help the municipality select from the innumerable combinations of actions that could satisfy the constitutional obligation, the one that gives appropriate weight to the many conflicting interests involved, the one that satisfies not only the Constitution but, to some extent, the parties as well.

Where the court, however, has solid reason — more than faith — to believe that the municipality will promptly attempt to rezone in accordance with the spirit of the court’s order, the better part of judicial discretion would be not to appoint a master.

This form of supervision is neither as intrusive nor as novel as it might seem. It is not overly intrusive since the municipality itself develops the ordinance with the advice and assistance of the special master and the participation of the other parties. 78 Colum.L.Rev. at 809. The final result, of course, is subject to the trial court’s approval. Id. Nor is it especially novel. In addition to the increasing use of special masters in the implementation of remedies in institutional litigation, courts necessarily intrude into parties’ affairs in all litigation — that is the very nature of a lawsuit and its consequences. Such intrusions have traditionally taken the form of supervising a party’s business, whether as a result of bankruptcy, probate, or corporate litigation; compelling parties to appear as witnesses to testify, which may entail considerable disruption of those persons’ lives and affairs; creating special tribunals; and in many other ways becoming involved itself with the lives and activities of the parties. 93 Harv.L.Rev. at 474k-92. We have however become accustomed to seeing courts and their delegates function in those “traditional” roles and therefore do not object to their *284activities, while the use of special masters is a relatively new remedial device.

The special master may also free the court from unwise direct over-involvement in the revision of the zoning ordinance, saving the court’s time and eliminating the need for the court to develop even greater expertise in the intricacies of land use regulation. The municipality receives the assistance of an expert in the zoning field to aid in its revision process. Obviously the court must select a planning expert who has had no previous connection with the litigation.

The master will work closely not only with the governing body but with all those connected with the litigation, including plaintiffs, the board of adjustment, planning board and interested developers. He or she will assist all parties in discussing and negotiating the requirements of the new regulations, the use of affirmative devices, and other activities designed to conform to the Mount Laurel obligation. The parties will presumably give the master’s suggestions great weight, since the revised ordinance will be submitted to the master for his or her review and recommendations prior to its submission .ip the court. During the course of the revision process, the master will report periodically to the court on the progress of the revision process. At the end of the 90 day period, on notice to all the parties, the revised ordinance will be presented in open court and the master will inform the court under oath, and subject to cross-examination, whether, in his or her opinion, that ordinance conforms with the trial court’s judgment. That opinion, however, is not binding on the trial court. The master’s powers are limited to rendering opinions, proposing findings, issuing recommendations, and assisting the court in other similar ways as it may direct.40 See, e.g., Fidelity Union Trust Co. v. Ritz Holding Co., 126 N.J.Eq. 148 (Ch. 1939). It is the trial court *285that must ultimately determine, independently, whether or not the municipality has conformed to its judgment and to the Mount Laurel doctrine.

The municipality may elect to revise its land use regulations and implement affirmative remedies “under protest.” If so, it may file an appeal when the trial court enters final judgment of compliance. Until that time there shall be no right of appeal, as the trial court’s determination of fair share and non-compliance is interlocutory. Stay of the effectiveness of an ordinance that is the basis for a judgment of compliance where the ordinance was adopted “under protest” shall be determined in accordance with the usual rules. Proceedings as ordered herein (including the obligation of the municipality to revise its zoning ordinance with the assistance of the special master) will continue despite the pendency of any attempted interlocutory appeals by the municipality.

3. Remedies for Non-Compliance

If within the time allotted by the trial court a revised zoning ordinance is submitted by the defendant municipality that meets the municipality’s Mount Laurel obligations, the trial court shall issue a judgment of compliance. If the revised ordinance does not meet the constitutional requirements, or if no revised ordinance is submitted within the time allotted, the trial court may issue such orders as are appropriate, including any one or more of the following:

(1) that the municipality adopt such resolutions and ordinances, including particular amendments to its zoning ordinance, and other land use regulations, as will enable it to meet its Mount Laurel obligations;

(2) that certain types of projects or construction as may be specified by the trial court be delayed within the municipality until its ordinance is satisfactorily revised, or until all or part of its fair share of lower income housing is constructed and/or firm commitments for its construction have been made by responsible developers;

(3) that the zoning ordinance and other land use regulations of the municipality be deemed void in whole or in part so as to relax or eliminate building and use restrictions in all or selected portions of the municipality (the court may condition this remedy upon failure of the municipality to adopt resolutions or ordinances mentioned in (1) above); and

*286(4) that particular applications to construct housing that includes lower income units be approved by the municipality, or any officer, board, agency, authority (independent or otherwise) or division thereof.

In determining remedies for non-compliance, the trial court may use the assistance and advice of a master subject to the guidelines set forth above.

The remedies permitted herein upon judgment of non-compliance go beyond what had previously been allowed by this Court in Mount Laurel cases. They were clearly anticipated by the Court, however, in Madison, where we explicitly approved and adopted remedies far beyond our actions in Mount Laurel I. As we noted in Madison:

In Mount Laurel we elected not to impose direct judicial supervision of compliance with the judgment “in view of the advanced view of zoning law as. applied to housing laid down by [the] opinion.” 67 N.J. at 192. The present case is different. The basic law is by now settled. Further, the defendant was correctly advised by the trial court as to its responsibilities in respect of regional housing needs in October 1971, over five years ago. [Oakwood at Madison Inc. v. Township of Madison,] 117 N.J.Super. 11. It came forth with an amended ordinance which has been found to fall short of its obligation. Considerations bearing upon the public interest, justice to plaintiffs and efficient judicial administration preclude another generalized remand for another unsupervised effort by the defendant to produce a satisfactory ordinance. The focus of the judicial effort after six years of litigation must now be transferred from theorizing over zoning to assurance of the zoning opportunity for production of least cost housing. [72 N.J. at 552-53].

That step itself was anticipated in Mount Laurel I, when we said that “[sjhould Mount Laurel not perform as we expect, further judicial action may be sought by supplemental pleading in this cause.” 67 N.J. at 192.

It is now five years beyond Madison. The direct orders we issued to the municipality then, 72 N.J. at 553, may appropriately now be issued by trial courts initially and with complete specificity. And that which we intimated in Madison might be the ultimate outcome after so many years of litigation— adoption by the trial court of a master’s recommendations to achieve “compliance,” id. at 553-54 — may now be the appropriate initial judicial remedy at the trial level.

*287We adhere to the belief that where conventional remedies are adequate to vindicate a right, they should be employed, that it is unwise to devise remedies that partake more of administrative and legislative than of judicial power where traditional remedies will do. Judicial legitimacy may be at risk if we take action resembling traditional executive or legislative models; but it may be even more at risk through failure to take such action if that is the only way to enforce the Constitution.41

In short, there being a constitutional obligation, we are not willing to allow it to be disregarded and rendered meaningless by declaring that we are powerless to apply any remedies other than those conventionally used. We intend no discourse on the history of judicial remedies, but suspect that that which we deem “conventional” was devised because it seemed perfectly adequate in view of the obligation it addressed. We suspect that the same history would show that as obligations were recognized that could not be satisfied through such conventional remedies, the courts devised further remedies, and indeed the history of Chancery is as much a history of remedy as it is of obligation. The process of remedial development has not yet been frozen.

We should be clear as to what is new here and what is conventional and about the extent of remedial change, regardless of the labels used. The use of a master to aid in resolution of a dispute is not new. Indeed, here it is not a remedy at all *288but a method of aiding the parties in complying with a court order.42

When the court orders that an ordinance be amended, it does very little different from ordering that a variance be granted, actions taken by our courts in New Jersey for many years. It does very little different from declaring that a zoning ordinance is invalid on equal protection grounds, the effect of that often being not simply to allow a plaintiff to use his property in a manner not permitted by the ordinance, but to give the same right to an entire class. The ordinance is effectively amended to permit a use explicitly excluded, or in some cases to exclude one explicitly permitted. Sometimes the action of the court comes even closer to ordering, indeed declaring, that an ordinance has been changed, see West Point Island Ass’n v. Township Committee of Dover Twp., 54 N.J. 339 (1969), where this Court, in effect, affirmed the decision of a trial court ordering a municipality to take certain action, which action could be taken only by the adoption of a resolution that the municipality had not adopted. As noted above, we did not hesitate, in Madison, to order amendment of the municipal zoning ordinance. Similarly, in Lusardi v. Curtis Point Property Owners Ass'n, 86 N.J. 217 (1981), relying on the judiciary’s power to regulate zoning in the public interest, we effectively modified an ordinance that conflicted with the state’s policy of affording recreational opportunities on the Atlantic seafront for as many citizens as possible.

The scope of remedies authorized by this opinion is similar to those used in a rapidly growing area of the law commonly referred to as “institutional litigation” or “public law litiga*289tion.”43 While it may not have been appropriate at the time of Mount Laurel to employ those remedies, regularly used in such public law litigation, we clearly recognized “the further extent of judicial power in the field” by citing the lower court’s decision in Pascack, 131 N.J.Super. 195 (Law Div.1974), a case in which the panoply of remedies appropriate in . institutional litigation was used. What we said in Mount Laurel in reference to remedy eight years ago was that such remedies were “not appropriate at this time, particularly in view of the advanced view of zoning law as applied to housing laid down by this opinion .... ” 67 N. J. at 192. That view is no longer “advanced,” at least not in this state. It is eight years old. Our warning to Mount Laurel — and to all other municipalities — that if they do “not perform as we expect, further judicial action *290may be sought . ..,” id. at 192, will seem hollow indeed if the best we can do to satisfy the constitutional obligation is to issue orders, judgments and injunctions that assure never-ending litigation but fail to assure constitutional vindication.

4. Summary of the Remedial Stage

The remedies authorized today are intended to achieve compliance with the Constitution and the Mount Laurel obligations without interminable trials and appeals. Municipalities will not be able to appeal a trial court’s determination that its ordinance is invalid, wait several years for adjudication of that appeal, and then, if unsuccessful, adopt another inadequate ordinance followed by more litigation and subsequent appeals. We intend by our remedy to conclude in one proceeding, with a single appeal, all questions involved. There will be either a judgment of compliance (from which a municipality that acted “under protest” may appeal with or without stays) signifying the trial court’s conclusions that there are land use regulations and affirmative devices in place conforming to the constitutional obligation; or there will be a judgment containing one or more of many orders available in the event of non-compliance along with the action of the municipality conforming to such orders. On appeal, the appellate court will have before it everything needed to determine fully the issues.

It may ultimately turn out on appeal, of course, that the trial court’s initial determination that the ordinance before it failed to comply with Mount Laurel was incorrect. In that case, all of the steps subsequently taken by the municipality to comply at the trial level may have been wasted energy. Our requirement of this procedure, however, is based upon our belief that much more time has been (and would continue to be) wasted, and much less compliance effected, as a result of the multiple appeals that have been allowed in the past. In the most unusual circumstances stays may be granted either by the trial or appellate courts and interlocutory appeals taken (or *291attempted); furthermore, there may even be circumstances in which the trial court declines to handle the litigation in one package. It may, for instance, enter as final judgment (upon certification pursuant to £.4:42-2) what would otherwise be an interlocutory order invalidating the ordinance before it. It should ordinarily do so only where it entertains substantial doubts as to the correctness of its position and concludes that on balance an immediate appeal is clearly preferable to any procedures that might otherwise follow the interlocutory judgment of invalidation.

We intend to administer the Mount Laurel doctrine effectively. It is complex. Its administration is important not simply to those seeking lower income housing, but to the municipalities as well. We have no desire to deprive municipalities of their right to litigate each and every determination affecting their interests, but we believe that the present procedures, allowing numerous appeals, retrials, and ordinarily resulting in substantial delay in meeting the obligation, do not strike the proper balance. While we cannot totally satisfy both the plaintiffs’ and defendants’ interests, we think the procedures required above come closer than those that have existed in the past to achieving a just balance of all the polices involved.

That balance also requires modification of the role of res judicata in these cases. Judicial determinations of compliance with the fair share obligation or of invalidity are not binding under ordinary rules of res judicata since circumstances obviously change. In Mount Laurel cases, however, judgments of compliance should provide that measure of finality suggested in the Municipal Land Use Law, which requires the reexamination and amendment of land use regulations every six years. Compliance judgments in these cases therefore shall have res judicata effect, despite changed circumstances, for a period of six years, the period to begin with the entry of the judgment by the *292trial court.44 In this way, municipalities can enjoy the repose that the res judicata doctrine intends, free of litigious interference with the normal planning process.

F. Judicial Management

This topic is listed under “Resolution of the Issues,” the subject heading of one of the three major portions of the opinion. Strictly speaking, it is not an “issue” at all. Judicial management of a Mount Laurel trial, however, is as important to the constitutional obligation as our substantive rulings today. Confusion, expense, and delay have been the primary enemies of constitutional compliance in this area. This problem needs the strong hand of the judge at trial as much as the clear word of the opinion on appeal.

The judges assigned to Mount Laurel cases should confer with counsel as soon as the pleadings are complete, or before, if that seems desirable. Those conferences should result in special orders that establish definite limits and schedules for discovery, determine what motions will be required and when they shall be prepared and argued, what experts will be used and on what subject, when the pretrial shall be held and for what purposes, and whether the court should retain an expert for trial purposes (testifying as to region, regional need, fair share, etc.) and/or for remedial purposes (as a master). The trial court should remain in close touch with counsel, keeping itself informed as to the status of the case on a regular basis and calling status conferences whenever they seem needed. There is no reason why, if such procedures are used, the factual and legal issues in the matter cannot be identified early in the litigation; nor any reason why issues of region, regional need, fair share, and compliance cannot be promptly disposed of, followed immediately by resolution of the remedial stage of the litigation. The

*293trial court should use any aids that may sensibly dispose of this litigation fairly, practically, promptly, and effectively. There are experts in this field who are prepared to testify,45 who have studied this subject matter for many years, and who will not be in the pay of any of the parties, although their general bias may be well known. They should be liberally used by the trial court. They include those in the public sector, in particular in the Department of Community Affairs, as well as those working in the private sector. As for the compensation of court-appointed experts — and compensation will not always be required — the trial court should determine that matter at the time the expert is retained. One or more of the parties will have to pay; on occasion the ultimate liability may await the outcome of the litigation.

We hope that individualized case management, the constant growth of expertise on the part of the judges in handling these matters, the simplification and elimination of issues resulting both from our rulings and from the active involvement of judges early in the litigation, and the requirement that, generally, the matter be disposed of at the trial level in its entirety before any appeal is allowed, will result in an example of trial efficiency that needs copying, not explaining.

III.

Resolution of the Cases

A. Mount Laurel II

1. The 1976 Revised Zoning Ordinance

Plaintiffs here are the same as in Mount Laurel I (Southern Burlington County N.A.A.C.P., Camden County C.O.R.E., Camden County N.A.A.C.P. and several individuals). They are, or represent, lower income persons seeking housing in Mount Lau*294rel.. They are joined by Davis Enterprises, a mobile home developer permitted, on remand after our original decision, to intervene as plaintiff. They attack Mount Laurel’s amended zoning ordinance passed in response to Mount Laurel I, which invalidated those portions of the existing zoning ordinance inconsistent with it. The remand in that decision required adoption of an amended ordinance within 90 days (or as extended by the trial court) and allowed the plaintiffs to file a supplemental complaint 30 days after such adoption if they wished to challenge the amended ordinance. Mount Laurel I, 67 N.J. at 191. The amended ordinance was adopted on April 19, 1976, 13 months, rather than 90 days, after our opinion. The action now before us was filed shortly thereafter.

After ruling that all developing municipalities were required, through their land use regulations, affirmatively to provide a realistic opportunity for lower income housing, we had held Mount Laurel’s zoning ordinance invalid “only to the extent and in the particulars set forth in this opinion.” Mount Laurel I, 67 NJ. at 191.46 Instead of attempting to amend those specific deficiencies, Mount Laurel simply added three new zones to meet its fair share obligation, presumably assuming that such action would conform to the underlying intent of our ruling.

We find that the amended ordinance falls far short of what was required, that it neither corrects the particular deficiencies of the prior ordinance nor otherwise affirmatively provides a realistic opportunity for Mount Laurel’s fair share of lower *295income housing. It is little more than a smoke screen that attempts to hide the Township’s persistent intention to exclude housing for the poor.

In our original decision we gave Mount Laurel the opportunity to amend its ordinance. Stating that “[w]e trust it will do so in the spirit we have suggested ...,” Mount Laurel I, 67 N.J. at 192, we declined to impose any judicial supervision over the municipality’s efforts to comply. Our trust was ill plaeed. Therefore, to assure compliance with our mandate, all further proceedings to conform to today’s decision shall be strictly supervised by the trial court, including not only any further litigation that may be required by this opinion, but all municipal action needed to conform to this and the trial court’s judgment.

The original Mount Laurel ordinance under trial court review in 1972 provided for several zones: an industrial zone (about 30 percent of the land — although there is some dispute about that), a retail business zone (1.2 percent of the land), and five residential zones that included approximately 10,000 of the Township’s 14,000 acres. Four of those five residential zones allowed only detached single family residences, the requirements being such as would prevent low and moderate income families from buying them. The fifth residential zone was the only one allowing multi-family housing, but it consisted of only 200 of the 10,000 residential acres, was designed for single-ownership development of the entire zone, was limited to senior citizens, and contained restrictions and requirements that again brought the price or rental beyond the means of low and moderate income retirees. Additionally, Mount Laurel had approved four planned unit developments (PUDs) pursuant to N.J.S.A. 40:55-54 to -67 (subsequently repealed by the Municipal Land Use Law),47 *296which, when completed, would provide for approximately 10,000 units by the year 2000. While allowing multi-family housing, these PUDs (which were carved out of the industrial and three residential zones) were similarly too expensive for lower income families. The resolutions authorizing them noted that they would “attract a highly educated and trained population base ...,” Mount Laurel I, 67 N.J. at 168, and it was clear that “only persons of medium and upper income [were] sought as residents.” Id. at 167.

This Court’s conclusion that no low or moderate income housing could be built in Mount Laurel was based on numerous restrictions and cost-generating provisions contained in its land use regulations as well as on the total absence (except for the PUDs and the senior citizen housing) of any provision for multi-family housing. Those restrictions included large lot zoning (or more accurately, the absence of small lot zoning), limitations on multi-bedroom homes and units, penalties for a large number of children per unit, contribution requirements for mul-ti-family units, excessive mínimums for frontages, setbacks, front yards, and home sizes, overall density máximums for sections and projects, paving requirements, and more. This Court specifically declared some of these restrictive devices invalid.

Nothing has really changed since the date of our first opinion, either in Mount Laurel or in its land use regulations. The record indicates that the Township continues to thrive with added industry, some new businesses, and continued growth of middle, upper middle, and upper income housing.48 As far as *297lower income housing is concerned, from the date of that opinion to today (as far as the record before us shows) no one has yet constructed one unit of lower income housing — nor has anyone even tried to.49 Mount Laurel’s lower income housing effort has been either a total failure or a total success — depending on its intention.

We realize that given today’s economy, especially as it affects housing, the failure of developers to build lower income housing does not necessarily prove that a town’s zoning ordinances are unduly restrictive. One might have expected, however, that in the eight years that have elapsed since our decision, Mount Laurel would have something to show other than this utter cipher — that is, unless one looked at the amended ordinance.

Mount Laurel’s notion of providing a realistic opportunity to build lower income housing has led to the rezoning of less than one-fourth of one percent of its land (about 20 out of 14,700 acres). This miniscule acreage consists of three zones, R-5, -6, and -7, each one owned by a different individual (apparently not residential developers in the cases of R-5 and R-6) who may very well elect never to take “advantage” of the alleged opportunity to build lower income housing.

The zone designated R-5, consisting of 13 acres, allows the construction of townhouses and garden apartments with a maxi*298mum of 10 units per acre. It is owned by an industrial developer, is totally surrounded by industrially zoned land, virtually isolated from residential uses, has no present access to other parts of the community, no water or sewer connections nearby, is in the path of a proposed high speed railroad line, and is subject to possible flooding. It would be hard to find (other than R-6) a less suitable parcel for lower income or indeed any kind of housing. Furthermore, as one of plaintiffs’ experts pointed out, no experienced industrial developer would allow this parcel to become a pocket of protesting residents objecting to his planned industrial uses surrounding them.

The R-6 zone is for detached single family residences on 6,000 square foot lots, which is an effort to comply with the Mount Laurel I requirement that there be some residential development permitted on “very small lots.” Mount Laurel I, 67 N.J. at 187. It includes, however, only 7.45 acres. It has an extremely serious drainage problem, lying so low compared to the surrounding area that it would cost $10,000 per acre, according to plaintiffs’ experts, to raise it so as to minimize that problem. In addition, there are no water or sewer connections nearby. There are cost-generating requirements concerning parking, street widths, and others, that will subsequently affect the price of homes, if they are ever to be built. The size of the zone itself is so small that it is highly unlikely that any developer would consider building low and moderate income housing there, for the necessary economies of scale could never be achieved. Defendant’s planner estimated that only 30 units could be built in this zone, and conceded that under no circumstances would anything be built for five to six years since there would be no sewer or water access available until then. Lower income housing on this tract is a phantom.

The R-7 zone is somewhat more complex. It does not consist of any specific land but rather is defined as being a maximum of 10 percent of the units to be built in Section VII of an existing approved PUD known as Larchmont. The only thing certain *299about this zone is that there will be no construction started until 1984, according to defendant’s planner.

R-7 is really not a zone at all, but rather a waiver by Mount Laurel of certain restrictions and requirements that would otherwise have been imposed on the Larchmont Section VII units. As noted in Mount Laurel I, the price of units scheduled to be built in this development would be far beyond the reach of low or moderate income families because of these restrictions. Plaintiffs’ planner concluded that even without them, the developer would have no incentive to build lower income housing. In other words, R-7 provides the “realistic opportunity for the construction of lower income housing” by allowing, not requiring, such construction by an existing developer whose plans for middle and upper income housing are already intact and who will be permitted to build precisely what he had intended to build at a higher profit than could be realized from lower income housing.

Unless something changes radically, it is certain that no builder will construct lower income housing in R-7. There is no evidence that the present developer has any intention to do so, especially in light of the benefits available to him when he builds upper and middle income housing in the R-7 zone.

Mount Laurel’s view of these zones is, of course, somewhat different. The Township itself, however, concedes that at the very most the three new zones could accommodate only 131 units of lower income housing. Their belief that this would be sufficient to comply with our mandate was based upon an analysis of Mount Laurel’s fair share obligation that we find to be wholly inadequate.

To determine its fair share, Mount Laurel first conducted an on-site study to determine its indigenous lower income housing need and concluded that this was 103 units based upon the number of deteriorated or dilapidated units in the Township and the number of lower income families presently residing in *300Mount Laurel paying rent beyond their means.50 Mount Laurel then calculated its fair share of the prospective regional lower income housing need to the year 2000 as 515 units. The Township then incorrectly assumed that its indigenous housing obligation was part of its prospective need obligation and therefore concluded that its total obligation until the year 2000 was 515 units, 103 of which met its present indigenous need.

The final step of the Township’s fair share analysis was a determination that these 515 units should be phased in through the year 2000. The Township would allow immediately for the 103 units to meet its indigenous need, and permit the remaining 412 units in segments of 17 per year through 2000. The Township maintained at trial that once its three new zones were filled to capacity with lower income housing, it would create additional zones where necessary to meet its yearly quota.

The shortcomings of Mount Laurel’s fair share calculations are obvious: Mount Laurel adopted an approach calculated to lead to the smallest possible share of lower income units in order to keep such units out, as it has done successfully for so long. The Township began by adopting the estimate of its region’s lower income housing need derived by another entity, the Delaware Valley Regional Planning Commission. We will for the moment assume the correctness of that Authority’s conclusion that 22,900 units represents the prospective need to the year 2000 for Burlington County, although one could make a case — as plaintiffs’ experts did — that that figure should be closer to 40,000.

It is the allocation of this need among the municipalities of the county that is so blatantly self-serving. The sole factor used by Mount Laurel’s planners in allocating this regional need for 22,900 lower income units was “developable land.” Its studies indicated that Mount Laurel had 5,936 acres of such land, and *301Burlington County 263,282 acres, and concluded that this 2.25 percent ratio, when applied to the county need of 22,900 units through the year 2000, meant that Mount Laurel’s fair share was 515 units. Vacant developable land, at this point, may be regarded as land not legally committed to other uses. The formula, therefore, assigns the same share to 100 acres located 100 miles from Camden, totally unsuitable for lower income housing and totally devoid of any demand for such use, as it does to 100 acres 10 miles from the center of Camden, near shopping centers, transportation facilities, and highly suited for lower income housing and subject to intensive demand for such use. In fact Mount Laurel’s formula equates the highly desirable vacant acreage of Mount Laurel with that of the Pine Barrens. Its analysis would lead to the conclusion that Bass River, located far from any activity that would suggest, at least at the present time, its desirability for lower income housing and with a present population of 1,000, has a “fair share” of 2,500 lower income units while Mount Laurel, located at the heart of the regional housing demand, and with a population of 17,614, has a fair share of 515 units.

As one of the plaintiffs’ experts noted, the Mount Laurel plan is “not a fair share plan, virtually by definition.”51 More *302importantly, Mount Laurel failed to take certain critical factors into consideration in its allocation formula, namely, the land’s suitability and the need for lower income housing. As' for suitability, plaintiffs’ expert noted that the distance from Camden was a fair measure of suitability (and perhaps need) in this analysis. She also noted that any respectable fair share formula took into consideration the “need” for lower income housing, usually using some kind of employment statistic as a measurement of that need, reflecting the obvious fact that people want to live near where they work. Plainly, consideration of these factors and reduction of the weight given to vacant developable land would substantially increase Mount Laurel’s fair share allocation.52

Furthermore, the measurements were made for different times — Burlington’s vacant “developable” land being taken as of 1970, and Mount Laurel’s as of 1976, guaranteeing a consequently smaller fair share for Mount Laurel. As Mount Laurel’s planner noted, use of the same year for both, namely 1976, would have increased Mount Laurel’s fair share only from 515 units to 687 units. This was, according to the planner, “just not significant [enough] to really worry about.”

In sum, we find that Mount Laurel’s 1976 revised zoning ordinance fails completely to comply with the mandate of Mount Laurel I. The three new zones created by the revised ordinance do not provide a “realistic opportunity” for the construction of any lower income housing. Further, even if those zones could realistically accommodate as much lower income housing as Mount Laurel claims (131 units), this would fall far short of Mount Laurel’s fair share of the prospective regional lower *303income housing need — as would the 515 units that Mount Laurel claims it will accommodate by the year 2000.

The conflicting testimony offered by Mount Laurel on these issues was unpersuasive. The Township planner responded to criticism that the use of vacant land as the exclusive criterion in determining fair share would result in allocating an enormous amount of lower income housing to the Pine Barrens (30 percent of the entire regional need, according to plaintiffs’ expert) by noting “that it is a very reasonable allocation method because the total units ... get equitably distributed over the entire county; and ... there are a lot of townships in the county land.” Given the likelihood that areas such as the Pine Barrens that are reserved as conservation zones by the SDGP should not have any prospective fair share allocations, we find wholly unacceptable Mount Laurel’s allocation formula, which would place a greater obligation on Pine Barrens municipalities.

Mount Laurel also tried to deflect criticism of its zoning ordinance by contending once again that its PUDs, which this Court in Mount Laurel I found totally unsuited for lower income housing, did in fact provide a realistic opportunity for such housing. For the reasons outlined in Mount Laurel I, 67 N.J. at 167-69, 182-83, we again reject this contention.53

Finally, Mount Laurel maintained that it had made an “honest attempt in an unknown area” and that the courts should defer to a municipality’s good faith effort to determine and meet its fair share obligation. It was this argument that seems to have particularly impressed the trial court when it upheld the Township’s revised zoning ordinance. Relying upon this Court’s statement in Madison, that the “entire problem” of formulating fair share plans “is essentially and functionally a legislative and administrative, not a judicial one,” 72 N.J. at 541-42 (footnote *304omitted), and having concluded that Mount Laurel acted in “good faith and with the express intent of compliance,” Mount Laurel II, 161 N.J.Super. at 344, the trial court upheld Mount Laurel’s fair share analysis and the resulting 1976 revised zoning ordinance.54

The trial court’s initial finding of compliance55 is understandable in light of Madison. Under the objective tests of today’s opinion, however, Mount Laurel’s revised ordinance is plainly insufficient to meet the Township’s Mount Laurel obligation. The plaintiffs not only show that Mount Laurel had failed *305to provide a realistic opportunity for the construction of its fair share of lower income housing, however that fair share is reasonably measured, but, even absent such evidence, also proved that the land use regulations of Mount Laurel remained facially invalid. The demonstration of “facial invalidity” does not depend upon an ordinance being totally bereft of provisions for multi-family dwellings; it is enough to show either that such provisions are woefully inadequate or are simply a smoke screen that diffuses the underlying exclusionary intent or effect.

Moreover, a much more significant burden, different in kind although not in weight, was cast on Mount Laurel in this case. Mount Laurel’s actions in this matter, commencing even prior to 1971 (for 1971 marks simply the date of the commencement of the original action, as distinguished from the date when Mount Laurel began to exclude lower income people, which undoubtedly occurred long before), require a modification of the rule that attaches presumptive validity to municipal ordinances. Its actions not only make such a presumption inappropriate, but, given the importance of the constitutional obligation, require just the reverse, namely, that the burden be cast on Mount Laurel to prove that its ordinances are valid.

Presumptive validity of governmental action serves many important values. It acts as the most effective check on judicial interference with executive and legislative actions. It is justified by the fact that those in government generally act within the powers granted to them and do so properly. Ultimately it represents an assertion of faith in government, including an obligation on the part of the governed to abide by the rules of constituted authority, for it casts a heavy imprint of validity on any governmental action challenged by an individual. Absent particular fundamental interests (such as freedom of speech) that may be impinged upon, any governmental action from the issuance of a parking ticket to the seizure of a steel plant is presumptively valid. The genius of our system of laws is that it is only a presumption, for both may be set aside upon *306proper proof that the presumption was unwarranted. The exception, however, is a rare one, for the presumption goes deep, and indirectly includes the assumption of any conceivable state of facts, rationally conceivable on the record, that will support the validity of the action in question.

Given the importance of the societal interest in the Mount Laurel obligation and the potential for inordinate delay in satisfying it, presumptive validity of an ordinance attaches but once in the face of a Mount Laurel challenge. Equal treatment requires at the very least that government be as fair to the poor as it is to the rich in the provision of housing opportunities. That is the basic justification for Mount Laurel. When that clear obligation is breached, and instructions given for its satisfaction, it is the municipality, and not the plaintiffs, that must prove every element of compliance. It is not fair to require a poor man to prove you were wrong the second time you slam the door in his face.

This ruling is similar to Kruvant v. Mayor & Council Twp. of Cedar Grove, 82 N.J. 435 (1980), where we announced a “time of decision” rule that precluded a municipality from blocking a particular use of land by continually adopting prohibitory ordinances, one just as invalid as the next. There was a time to stop, we said, and while it may have taken six ordinances in Kruvant before we called a halt to dilatory municipal action, the principle is the same: depending upon the circumstances, a time must come when the courts will cease to defer in the conventional manner to municipal action. In Kruvant, we refused to consider the most recently adopted municipal ordinance; here we refuse to accord presumptive validity to Mount Laurel’s revised ordinance. The delay in this matter exceeds that in Kruvant, for the rights of those represented by plaintiffs have been denied for at least ten years and undoubtedly far longer. Furthermore, the interests protected here dwarf those involved in Kruvant, as important as they might have been, for Kruvant *307involved only the economic interests of the plaintiffs. Here we have plaintiffs who assert interest in one of the basic necessities of life,56 and seek protection that, if denied, would similarly affect many, many other poor people.57

We therefore remand this matter to the trial court for further proceedings to determine Mount Laurel’s fair share, and upon such determination to require further actions by the municipality to assure the expeditious revision of Mount Laurel’s land use regulations (and other actions) all in accordance with this opinion. While we have held that the bona fides of Mount Laurel is irrelevant in determining its compliance with the underlying constitutional obligation, it is not irrelevant in determining the remedy adopted herein. Where, as here, there is evidence of lack of municipal good faith and/or interminable delay, trial courts must closely supervise orders designed to compel compliance. Here that supervision must include the appointment of a master.

2. The Builder’s Remedy

Davis Enterprises was permitted to intervene as plaintiff in this case after Mount Laurel I was decided. Davis proposed a 535 unit, 107 acre mobile home park for the Township. Davis committed itself to securing federal Section 8 subsidies for 20 percent of the units. Mount Laurel originally rejected the Davis project because its zoning ordinance barred all mobile homes. Although this rationale for excluding Davis is no longer tenable after our overturning of Vickers, supra at *308275 we must still decide whether Davis is entitled to the builder’s remedy it seeks.

The trial court granted the builder’s remedy, ordering the Mount Laurel Planning Board to consider the Davis application and review it in a “manner consistent with the least-cost housing principles enunciated in Oakwood v. Madison,” 161 N.J.Super. at 359. We affirm the grant of a builder’s remedy. It is clearly appropriate in this case under the new standard enunciated in this opinion. First, the Davis project will provide lower income housing for Mount Laurel. Beside the fact that mobile homes are generally much less costly than site-built housing, the trial court’s decision requires that Davis construct at least 20 percent of its units for lower income persons. In addition, the site chosen by Davis is plainly suited for mobile home development and Mount Laurel has presented no real evidence to the contrary. Finally, we feel that after ten years of litigation it is time that something be built for the resident and non-resident lower income plaintiffs in this case who have borne the brunt of Mount Laurel’s unconstitutional policy of exclusion.

The Mount Laurel Planning Board held hearings on the Davis project after this Court had already granted direct certification of the original trial court decision in Mount Laurel II. On May 8, 1980, after Davis had submitted a revised plan to the Board that reduced the number of proposed units from 535 to 456, the Board rejected the Davis project. On December 2, 1980, the trial court granted a motion made by Davis in aid of litigant’s rights, E.l:10-5, and ordered Mount Laurel to grant Davis a building permit on condition that Davis apply to HUD for Section 8 subsidies for 20 percent of its units. On March 17, 1981, we granted Davis’ motion for direct certification pending an appeal of the trial court’s order by the Township. We now affirm the December 2, 1980, order with the added condition that if Davis is not able to obtain the Section 8 subsidies being sought, the developer must use whatever other means are available to make certain that at least 20 percent of the units built *309are affordable by lower income households, with at least half of these being affordable by low income households.58

B. Caputo v. Chester

Plaintiffs, brothers Joseph and Aldo Caputo, who have owned the property in question since 1960, brought suit against defendants Chester Township and its planning board (Chester) after their efforts to obtain rezoning were unsuccessful. The suit was originally filed on July 22, 1975. By the next year it became apparent that Chester, having been advised by its counsel that its zoning ordinance violated the Mount Laurel doctrine, intended to effect a substantial rezoning of the entire Township. After Chester adopted a new zoning ordinance in October 1976, settlement efforts came close to terminating this litigation but ultimately collapsed. An amended complaint was filed December 16, 1976 and a trial was conducted over 25 days between October 11, 1977 and March 7, 1978. At that trial plaintiffs sought a declaration that the Chester ordinance was unconstitutional under the Mount Laurel doctrine along with appropriate relief, including an order that Chester rezone and include provisions in such rezoning that would enable plaintiffs to develop their land for multi-family units in accordance with their proposed plan (a builder’s remedy). The trial court, three years after the action was originally filed, in a 111 page decision held that the ordinance was unconstitutional and ordered rezoning in accordance with its opinion; the builder’s remedy was denied, however, primarily on environmental grounds, because the court concluded that Chester’s decision to allow multi-family dwellings elsewhere (and not allow them on plaintiffs’ property) was reasonable. The court specifically invalidated provisions of the *310ordinance that limited development to single family dwellings, and held that minimum five acre lot requirements were illegal per se. Plaintiffs appealed from the portion of the trial court’s decision denying the builder’s remedy; defendants appealed only from the invalidation of the five acre restriction — they did not appeal from the court’s determination that the ordinance violated Mount Laurel or from its order requiring rezoning. As a result, it would appear that the only issues before us involve the trial court’s refusal to grant a builder’s remedy and the legality of the five acre restrictions imposed by the Township.

Since Chester is characterized as a “limited growth” area by the SDGP, our decision to abide by the Plan’s policy of limiting development to growth areas requires us to note that were the trial court’s decision imposing a fair share obligation on Chester Township appealed, we would remand to enable the trial court to consider the effect of the SDGP in accordance with this opinion.59 It appears that close examination of the Township itself and of its zoning history lends support to the legitimacy, and perhaps the wisdom, of accepting the policy of the SDGP in limited growth municipalities as the suitable remedy to meet the Mount Laurel obligation.

Chester is in the southwest corner of Morris County, completely surrounding Chester Borough, which, by comparison, is fairly well developed. Chester has very little commerce or industry, and consists mostly of residences and farms. While its population growth in the 1970-80 decade was at a substantial rate (a 22 percent growth compared to a statewide increase of less than 3 percent), the total population and its density give a truer picture of the Township. The population in 1980 was 5,198, and its density 179.9 people per square mile, as compared to Morris County’s 866.9 and the state’s 983.3. Chester is described as being 42 percent developed and 58 percent undeveloped, al*311though we suspect that picture does not adequately convey its lack of development, since a very substantial portion of that 42 percent undoubtedly consists of vacant land (the one, two or five acres on which homes are found, as well as a very substantial portion of public parks). Of Chester’s more than 18,000 acres, 12.3 percent are residentially developed, 25.9 percent are developed for public purposes (almost completely consisting of park-lands), 19.6 percent are in agricultural use, 38.4 percent are undeveloped, and the remaining several percent are devoted to commercial, recreational, and industrial uses.60

Chester’s infrastructure, other than some of its highways and bus service, is far from well developed. Furthermore, its topography, water resources, and agricultural suitability are objective facts that support a policy of non-development. One portion of the Township consists of a plateau, a portion of which is develo-pable, while the balance consists of rugged, fairly hilly terrain, with steeply rising slopes, which is unsuitable for development. The two are divided by a “fault” marked by a ravine through which a stream flows bordered by steep hills and slopes. The fault, the stream, the steeply rising slopes and the consequent plateau and contrasting rugged terrain all traverse the Caputos’ tract. Chester contains the source and early portions of some of the streams that ultimately flow into and constitute the Raritan River. It also is located over substantial underground water supplies, which presumably provide the source of the streams themselves. The well-known effects of residential development — pollution of surface waters that run into streams, pollution of underground water supplies through sewage percolation, erosion of slopes during development (and thereafter) and conse*312quent pollution of streams — are particularly critical in Chester. If uncontrolled, such residential development could seriously affect this part of the Raritan River water basin, its downstream owners, the people who directly and indirectly depend upon the purity of the river, and the underlying water supplies. Because the Caputo tract is strategically located, its development would contribute to the effects mentioned above. Finally, Chester has within its borders a substantial portion of prime agricultural land, which no one disputes should remain free of development.

The record suggests that the duly elected and appointed officials of Chester, given all of these factors, had adopted a non-development policy. While they legitimately ascribed their decision to the many objective factors mentioned above (preservation of open spaces, agricultural land, maintenance of the purity of the watershed, unsuitability of terrain for development, dangers of erosion), it is clear that there was an additional equally important one: they simply wanted to keep Chester the way it was.

Some professional planners saw Chester in a different light. Without exception, every study of the area, no matter how defined, indicated that Morris County, including Chester, was, and would continue to become, an important focal point for ever-increasing development radiating from the general metropolitan region as well as from its own major centers of growth, Morristown and Dover. The demand for land in areas similar to and including Chester for all purposes — commercial, industrial and residential — was apparently intense. Whatever relevant projection might be considered, be it population growth, employment growth or ratable growth; Morris County and Chester, along with other county municipalities, would be shown as anticipating more businesses, factories, jobs, people, and ultimately homes. Presumably the projections were not simply mathematical formulas, but reflected an actual and ever-growing demand for land in the area. It was generally acknowl*313edged that the question was not whether Chester would be developed but how and under what controls.

Whatever the future may hold, it would appear that this development has still not really begun, for despite the 22 percent population growth during the last decade, it has been desultory, insofar as Chester is concerned, consisting only of a few homes built here and there over this ten year period. It may be assumed that the existing zoning regulations themselves have been one of the factors contributing to this very slow pace of growth.

Chester’s zoning history might almost be predicted from the foregoing statement of facts. The consistent overriding development pattern has been the construction of single family residences on very substantial parcels. A municipality that has more than 85 percent of its developed land (excluding publicly owned land, parklands and farms) in residential use, 95 percent of which consists of single family homes, could be expected to have a zoning ordinance in which more than 70 percent of the entire acreage of the Township is set aside for single family residences on either two acre or five acre plots. The zoning ordinance under attack, adopted in 1976, contains eight districts (as did the original ordinance of 1964). Of the four residential districts, three are limited to single family residences, one with two acre minimum lots (40 percent of the entire municipality), another with a five acre minimum (31 percent of the Township), and the third with a one acre minimum (0.5 percent of the total). The fourth residential zone permits, in addition to two acre single family residences, multi-family development at a density of five units per acre with a limit of ten bedrooms per acre. Furthermore, no more than 150 units may be built on any tract and no more than 300 in the entire Township. While cluster development was permitted, it was limited to tracts of at least 30 acres in the two acre zone and 50 acres in the five acre zone. Total acreage available for multi-family dwellings constitutes 1.5 percent of the Township. There were only three tracts so zoned, and two of them were not likely to be developed in the *314foreseeable future since they were owned by substantial long-term residents of the community.

The balance of Chester’s zoning consisted of parklands (22 percent), business (about 0.5 percent), industrial (1 percent) and residential/office (no acreage total or percentage determination was provided in the record).

The trial court found the ordinance facially invalid because it did not provide a realistic opportunity for the construction of a variety of housing, rejecting all of the justifications advanced by Chester. Additionally the court held the bedroom limitation in the multi-family zone and the five acre zoning provision invalid. It did not order any specific plan of rezoning but simply required revisions of the zoning, ordinance “consonant with this opinion,” noting that the five acre zone must be eliminated and that while there may be an allowance for low density zoning, small acre zoning and some lower income housing must be provided, along with provisions for “all variety and types of housing ... based upon the needs of the area”; bedroom restrictions and all cost generating restrictions should be removed in the multi-family zone and, generally, all of the provisions of the Municipal Land Use Law must be complied with.61

Municipalities that are in “growth” areas (unlike Chester) and have resulting fair share duties can satisfy them in a variety of ways, including density adjustments. We hold, however, that low density limitations like five acre lot minimums are not necessarily in violation of the Mount Laurel fair share obligation so long as municipalities are able to satisfy that obligation in spite of apparently “exclusionary” devices.62 *315Therefore, such devices are subject to the same level of scrutiny as other municipal regulations, once compliance with fair share requirements has been demonstrated. Moreover, we hold that the preservation of open spaces itself may, under proper circumstances, be sufficient justification for large lot zoning, including five acre zoning. Where a municipality’s zoning provides for its fair share of low and moderate income housing, as well as for other uses it deems appropriate, it is not obliged, in its other zones, to allow for the maximum density of construction that environmental factors will permit. In an area like Chester, it may decide that the value of preserving open space is sufficient to warrant such zoning. We therefore reverse the trial court’s ruling on five acre lot mínimums, intending thereby to leave Chester free to make its own decisions on the advisability of such zoning, subject to challenge on some basis other than Mount Laurel.

Similarly, the need for agriculture, open space, and, perhaps most importantly, for geographic and aesthetic heterogeneity and variety in different areas of this state convince us that the SDGP’s characterization of a presently predominantly rural area like Chester as “limited growth” is presumptively sound. More to the point, there is no showing even approaching that which would be required for us to replace the plan with our own planning judgments.63

The builder’s remedy sought by plaintiffs must of necessity be denied. One of the conditions for awarding such remedy is that the builder establish that the municipal ordinance fails to comply with the Mount Laurel obligation. Since we have held that Chester is not subject to that obligation (insofar as the prospective need for lower income housing is concerned), *316plaintiffs’ claim has not succeeded. Assuming, pursuant to the remand ordered hereafter, that Chester is required to further amend its ordinance or take other steps to provide a realistic opportunity for lower income housing for its indigenous poor, this denial of a builder’s remedy shall not be disturbed. We are satisfied that Chester established very substantial reasons for denying such a remedy. The environmentally sensitive nature of much of Chester’s lands, and the location of plaintiffs’ property within that sensitive area, were fully documented. We interpret the trial court’s decision as a determination not simply that there were better places in Chester for lower income housing, but that plaintiffs’ property was unsuitable for substantial environmental reasons. The record adequately supports that determination.

We therefore affirm the judgment of the trial court denying a builder’s remedy. Given the new principles set forth in this opinion, plaintiffs should be given the opportunity at the trial level to challenge the adequacy of the amended ordinance to meet Chester’s indigenously generated present need. We therefore reverse and remand for such further proceedings as plaintiffs may wish to pursue on that issue alone, in accordance with our opinion.

C. Glenview Development Co. v. Franklin Township

The SDGP’s treatment of agricultural areas is similar to that of limited growth municipalities: municipalities, like Franklin, classified as exclusively agricultural and/or conservation have no obligation to provide a fair share of the prospective regional housing need. The validity of this policy judgment is made clear by an examination of Franklin Township and its zoning history.

Franklin Township is a predominantly rural municipality in the center of Hunterdon County comprised of 23.3 square miles (14,718 acres), characterized as exclusively “agricultural” by the SDGP. It is located approximately 45 miles from Newark, 40 *317miles from Trenton, and 35 miles from Allentown, Pennsylvania. Seventy-six percent of the Township is under “farmland assessment” pursuant to the New Jersey Farmland Assessment Act of 1964, N.J.S.A. 54:4-23.1 to -23.24. Less than 3 percent of the Township is devoted to industrial and commercial uses. The population of Franklin, 2,294 in 1980, though substantially greater than its 1950 population of 1,255, has grown by only 140 since 1970. Franklin’s 1980 population density of 98.4 people per square mile is significantly fewer than Hunterdon County’s 203.4 and the state’s 983.3.

Franklin’s infrastructure network is relatively undeveloped. The Township itself has no public sewer or water utilities, although such services may be available from the adjacent Town of Clinton. Roads in the Township amount to only 2.58 linear miles per square mile, compared with a state mean of 10.24 linear miles per square mile. Mass transit is non-existent. There is only one public elementary school and one volunteer fire department.

The challenged Franklin zoning ordinance contains four zones: R-3 (agricultural, residential single family with three acre minimum, home occupations in a residence, public buildings, roadside stands, industrial) (12,798 acres or 87 percent of total); R-5 (same as R-3 with five acre minimum) (1,143 acres or 8 percent); Business (206 acres or 1.5 percent); and Flood Plain (only agricultural and recreational uses — no structures permitted) (279 acres or 2 percent). No multi-family dwellings are permitted. Plaintiff Glenview’s property is wholly within the R-3 zone.

Glenview owns 242 acres in Franklin that are presently zoned primarily for residential or agricultural uses on three acre or larger lots. On September 7, 1976, it brought an action in lieu of prerogative writ against Franklin challenging both the Township’s zoning ordinance as a whole and its application to plaintiff’s land. Glenview claimed that the ordinance failed to provide for Franklin’s fair share of the regional low and moderate income housing need and was therefore invalid under both *318 Mount Laurel and the Municipal Land Use Law (MLUL), N.J. S.A. 40:55D-1 to -92. On December 18,1978, two and one half years after the original complaint was filed and after a seven day trial, the trial court, in an opinion reported at 164 N.J.Super. 563 (1978), held that because Franklin Township has not yet shed its rural characteristics, it was not a “developing” municipality and therefore was not subject to the mandate of Mount Laurel. The court further held that no distinct obligation to provide low and moderate income hdusing was imposed on municipalities by certain sections of the MLUL. Finally, the court rejected plaintiff’s claim that the Township’s zoning ordinance was confiscatory as applied to plaintiff’s property.64

Clearly, imposition of a fair share requirement on Franklin at this time would be terribly costly and would result in tremendous change. The SDGP’s classification of Franklin as an area where growth should be discouraged appears logical. The SDGP’s conclusion was that high density development should be located elsewhere; that agricultural areas such as those found in Franklin should be preserved. We see no reason for the judiciary to disturb this conclusion of the state’s comprehensive plan. We therefore affirm the trial court’s determination that Franklin has no prospective fair share obligation. This does not, of course, relieve Franklin of its obligation to meet the present housing needs of its indigenous poor.

Plaintiff suggests that the MLUL is, in large part, a codification of the Mount Laurel fair share duty, particularly sections 55D-2(a), (d), and (e). Moreover, plaintiff claims that the MLUL extends the application of a fair share duty to all municipalities in the state, making the “developing” distinction obsolete. We do not believe that the MLUL in any way at*319tempts to define or codify the Mount Laurel obligation, although, as noted above, the MLUL explicitly recognizes the municipal obligation to zone with regional consequences in mind, and certain of its provisions strongly support the use of the SDGP to determine the locus of the obligation.

The legislative history of the MLUL leads us to conclude that while the Legislature was cognizant of our Mount Laurel I decision and drafted the MLUL so as not to conflict with it, it did not incorporate the doctrine into the legislation.

This legislative intent was made clear by the MLUL’s sponsor, Senator Greenberg, during a public hearing on the bill (S-3054):

The bill is a codification of five basic land use statutes enacted at different points in time, including the Planning Acts and the Zoning Enabling Acts, the Official Map and Building Act, the Plan Unit Development Act, and the Regional Planning Act.

Now as everyone in this room and throughout the State, I would suspect, knows by this time the Supreme Court of the State of New Jersey has rendered a decision which is commonly referred to as the Mt. Laurel Decision. So that the record is clear, this bill is not a response to that decision. This bill, as I have stated, is a procedural bill in concept and it should not be confused with any other bills which are pending dealing with the Mt. Laurel decision, one of which is mine and that may have given rise to some confusion. I have introduced a bill, the number of which is 3100,[65] which deals with a balanced housing concept. That is not the subject of this hearing, and I want to make that very, very clear. Public Hearing Before Senate County and Municipal Government Committee on Senate Bill No. 3054, April 3, 1975, at 2.

Other speakers reiterated Senator Greenberg’s point:

This Bill is void of any provision concerning the problem of exclusionary zoning as set forth in the recent New Jersey Supreme Court decision of Southern Burlington County, N.A.A.C.P., et al. v. Township of Mount Laurel, which was decided on March 24, 1975. [Id. at 32.] [Robert Molnar, Housing and Urban Affairs Committee of the New Jersey State Bar Association];

It should also be noted that S-3054 does not affect the zoning powers of a municipality with respect to that municipality’s power to determine use and density, lest the bill be confused with the recent Mt. Laurel Supreme Court decision. The key element of S-3054 is its procedural streamlining. [Id at 13.] [Stewart Hutt, General Counsel, New Jersey Builder’s Assn.]

*320In fact, Mr. Molnar expressed his Committee’s regret that the proposed MLUL did not address the substantive zoning duties announced in Mount Laurel I:

Although we realize this hearing is not on Senate Bill 3100, we think the time and attention of this Committee would be better directed to the provisions of that Bill in terms of meeting the issues raised by the Mount Laurel decision. Our Committee feels that the time has long passed for any half-way measures and that strong affirmative action will have to be taken by the New Jersey legislature so that these vital matters can be settled here where they rightly belong, not in the Courts by default. [Id at 33.]

Senator Greenberg’s response is self-explanatory:

SENATOR GREENBERG: Well, I appreciate your views. Since you were not here, let me just succinctly state again that this bill has been approximately one and a half years in the drafting at the direction of this Committee. It is true that the Mount Laurel decision was making its way up through the courts at that time. And almost simultaneously with the drafting of this bill, I have been engaged in the drafting of a bill dealing with housing which is presently pending in our Committee, to which you have also referred. I do not agree that the matters must be considered at the same time, and I have so stated. This is a bill which I deem to be procedural in nature and will permit, in my judgment, the implementation of any modifications which the Legislature sees fit to make as a result of the mandate directed to the Legislature in the Mount Laurel decision. I have very carefully attempted to read the Mount Laurel decision and I have only read it three times, so I am not quite yet sufficiently familiar with all of the ramifications to comment on it, nor do I think it’s appropriate today. But I have also re-read this bill that is under consideration today in the light of the Mount Laurel decision to determine whether or not there is anything in it which runs contrary to the concepts and principles of the Mount Laurel decision, and I do not find such conflict. I do find that local municipalities will have to function in some capacity with some element of home rule maintained in this State in the future but in conformity with the decision in the Mount Laurel opinion.

My view, and I speak for myself on this subject, is that this bill would permit those municipalities to function more effectively from a procedural point of view, regardless of any ultimate legislation that the Legislature may see fit to adopt in conformity with Mount Laurel. [Id. at 35-36.]

In other words, the MLUL was viewed as a procedural device. It did not interfere with the satisfaction of the constitutional duty. It was not a legislative effort aimed at creating balanced housing opportunities in New Jersey. See, e.g., The Comprehensive and Balanced Housing Plan Act, S-505 (1978). Given the realities of the political process discussed previously, the actual passage and enactment of the MLUL serve as further evidence of its limited purpose. We therefore agree with the trial court’s *321rejection of plaintiff’s claim that the MLUL legislative imposes a fair share duty on all municipalities.

As in Caputo v. Chester, we necessarily decide that no builder’s remedy is warranted here, since Glenview did not prevail either at the trial or appellate level. Furthermore, since its claim essentially was for Mount Laurel relief that would require Franklin to provide its fair share of the region’s prospective need, a builder’s remedy would be inappropriate even if, on further proceedings authorized below, it is determined that the ordinance does not adequately provide for Franklin’s present indigenous lower income housing needs.

We therefore affirm the judgment of the trial court denying a builder’s remedy. Given the new principles set forth in this opinion, plaintiff should be given the opportunity at the trial level to challenge the adequacy of the amended ordinance to meet Franklin’s indigenously generated present need. We therefore reverse and remand for such further proceedings as plaintiff may wish to pursue on that issue alone, in accordance with our opinion.

D. Round Valley v. Township of Clinton

In this case, we apply the Mount Laurel doctrine, as modified by our decision today, to what is essentially a rural but nonetheless “developing” community (as that term has been previously understood) that is partially comprised of a “growth area” as determined by the SDGP. This case also requires the application of our new standard for granting a builder’s remedy where the interests of the builder conflict with environmental concerns of the municipality, as well as a . review of the trial court’s appointment of a master in light of our rulings on that subject.

1. Procedural and Factual Setting

Plaintiff-developer, Round Valley, Inc., sought invalidation of the land use regulations of defendant Township of Clinton under the Mount Laurel doctrine (as well as on other grounds), along *322with a builder’s remedy. Its Mount Laurel claim was typical, its proofs directed at a showing of facial invalidity, along with a showing of Clinton’s actual fair share and the failure of its land use regulations to provide a realistic opportunity for construction of that fair share. The additional remedy sought was the development of a 790 acre, two parcel tract so as to permit construction of 3,500 units accommodating approximately 10,000 people. The trial court granted the developer the relief it sought, both invalidating the ordinance and granting a builder’s remedy. The Appellate Division reversed that judgment in all respects, and we certified the matter.

Clinton is a large municipality (about 22,000 acres) consisting in large part of farmland, public land, and residential uses. Though parts of it are more densely populated than most other portions of Hunterdon County, it is nevertheless appropriately characterized as “rural” (the total density being 244.3 people per square mile compared to the state’s 983.3). Its zoning ordinances for many years effectively prohibited all multi-family housing. As a matter of fact, there were only five to six multi-family units in the entire municipality at the time of trial. Similarly, low cost housing of any kind has been practically non-existent, 95 percent of the housing being beyond the reach of 75 percent of New Jersey residents.

Plaintiff’s tract was originally divided by Route 31 into an easterly portion (known as the Gobel tract) of 469 acres and a westerly portion (known as Beaverbrook) of about 321 acres. The easterly portion is well suited to many different kinds of residential development, including high density multi-family. The Beaverbrook tract, on the other hand, is hilly and therefore topographically not as suitable for development. Plaintiff’s proposed project would have placed most of the high density development on the Gobel tract. The project consisted of a planned unit development including a small commercial area, but devoted otherwise to construction of 3,500 new residential units: 90 percent of the units planned were townhouses and garden apartment units, with a minimum unit price of $29,900, *323and an average price range from $30,000 to $40,000, suitable at the lowest range for purchase by moderate income residents, or, with subsidization, by low income residents (market estimates in the region indicated that moderate income families could be accommodated in housing priced at $35,000 or below). If constructed, it would have added 10,000 people over a period of nine years to the population of Clinton, more than doubling its 1980 population of 7,345.

Plaintiff first presented its project to Clinton authorities in 1973. It was legitimately delayed for many reasons, including the changing development of the municipality itself as well as the changing law. In terms of zoning, what plaintiff sought was a rezoning of the Gobel tract. The Beaverbrook tract allowed residential development at the rate of three units per acre: plaintiff was apparently satisfied with that in terms of its overall project. The Gobel tract, however, allowed no residential development at all but was zoned for research, office and light manufacturing purposes (ROM). Plaintiff sought an amendment that would have given it the right to erect a planned unit development (a PUD option). When the proposed revisions of the land use regulations (pursuant to the Municipal Land Use Law) had progressed to the point where it became apparent to plaintiff that the amendment would not be forthcoming, plaintiff brought this lawsuit, filing the original complaint on April 15, 1975. The ordinance was in fact amended in 1977, shortly before trial began, in accordance with the Township’s plans; plaintiff’s complaint was then amended to assert all of its claims against the new ordinance.

Trial began on May 31,1977, and ended October 12,1977, with 29 days of hearings. The trial court found that since its alleged provision of multi-family housing was “camouflage” (if it allowed multi-family housing at all, it would have been much beyond the reach of lower income families), since it contained numerous excessive cost exactions, and since it was clearly over-zoned for commercial and industrial uses, the ordinance was facially invalid. Furthermore, it found that the Mount *324 Laurel doctrine applied to Clinton as a “developing" municipality and accepted plaintiff’s contention that Clinton’s prospective fair share obligation was between 2,833 and 3,457 units. It found that the existing (1977) land use regulations did not provide for such fair share, ordered revisions to accomplish same, and invalidated numerous cost exaction provisions in the form of specific revisions to the ordinance, namely, the inclusion of a PUD option on the Gobel tract. As to all of this, the court provided for the appointment of a master, not only to help implement the judgment and the ordinance revisions, but apparently actually to determine, for the court, whether the revisions complied with the judgment in all respects.

2. The Appellate Division Opinion

The judgment invalidating the zoning ordinance and awarding a builder’s remedy was stayed by the trial court pending appeal. The Appellate Division, in an opinion reported at 173 N.J.Super. 45 (1980), found that the trial court’s judgment did not invalidate the entire ordinance but simply accorded a builder’s remedy that was indistinguishable from a variance and reversed what it believed to be the only significant determination of the trial court, namely, the award of a builder’s remedy. It concluded that there was nothing before it regarding the ordinance as a whole. Id. at 52. It also reversed the trial court’s invalidation of specific portions of the zoning ordinance — the 120 percent performance bond, the two year maintenance bond, and the 50 acre minimum for a planned unit residential development. Id. at 53. It pointed out that the appointment of a master was “premature” in view of the fact that there had been no allowance for voluntary compliance by Clinton, apparently being of the opinion that until there had been some showing of intransigence on the part of the municipality, a master was inappropriate. Id. The court’s decision to reverse on the builder’s remedy was heavily influenced by the fact that after the trial court’s decision, Round Valley had sold the Gobel tract to Exxon for ROM uses. As the Appellate Division viewed it, this sale *325“renders most of the substantive issues presented on appeal moot.” Id. at 51.

We disagree with the Appellate Division’s characterization of the trial court opinion. While there is some ambiguity at the end of the trial court’s opinion and in its judgment, a reading of the entire record, the opinion and the judgment convinces us that the trial court intended to and did invalidate the entire Clinton land use regulations on Mount Laurel grounds. Furthermore, the trial court’s articulate and persuasive opinion leaves no doubt as to the correctness of its determinations that Clinton is subject to Mount Laurel (under the “developing” doctrine that then applied) and that its land use regulations were both facially invalid and actually inadequate in meeting its demonstrated fair share. Furthermore, the court’s appointment of a master was not only appropriate but most desirable under the circumstances, as was its award of a builder’s remedy. That relief, however, must be modified in view of the sale of the Gobel tract, as must the court’s delegation of certain powers to the master.

The Appellate Division’s interpretation of the trial court’s decision as addressing itself solely to the appropriateness of a builder’s remedy without invalidating the entire Clinton ordinance (or put differently, invalidating the ordinance only insofar as it applies to plaintiff’s lands) was based upon certain portions of the trial court’s opinion that were actually directed at indicating that the only portions of the ordinance immediately invalidated were those applicable to plaintiff’s lands. The trial court did not want to strip Clinton completely of any land use regulations pending the enactment of new and complying ordinances. That is what the trial court was referring to when it stated that the ordinances were invalid only as to plaintiffs lands “in the interim,” meaning until the new ordinances could be prepared and adopted.

There are innumerable indications in the trial court’s opinion and judgment that all of the zoning ordinances of Clinton were *326to be revised to conform to the opinion, which itself clearly invalidated them on Mount Laurel grounds, and that the functions of the master were not simply to assure the provision of builder’s remedy, but to implement the provisions of the opinion including those invalidating all of the ordinances.

The Appellate Division incorrectly believed that “plaintiff neither sought nor succeeded in having the township’s land use ordinances declared unconstitutional generally.” 173 N.J. Super, at 52. The fact is that in both its original and amended complaint, and in both the original and subsequent pretrial order, a declaration was sought that the land use regulations of Clinton were invalid under Mount Laurel. In the issues section of both pretrial orders the question is stated quite clearly. If indeed plaintiff had sought and the trial court had granted on Mount Laurel grounds a builder’s remedy only, without invalidating the ordinance, we would agree with the Appellate Division that such action would be improper. Mount Laurel is not to be used as a substitute for a variance. That, however, is not at all the case here.

Nor do we agree that the disposition of the Gobel tract after judgment in the trial court rendered most, or indeed any, of the issues moot. We are not sure whether the Appellate Division meant to rule in that fashion as a matter of law or whether it concluded that since plaintiff was satisfied at the trial level with the zoning in Beaverbrook and had now disposed of the only portion where it sought relief (namely, the Gobel tract), it no longer had any serious interest in the litigation. On the record before us, it is clear that plaintiff continues to seek a builder’s remedy that must now be limited to the Beaverbrook tract. We do not believe that plaintiff’s right to a builder’s remedy in cases of this kind must depend upon standing absolutely still until the litigation is completed. A sale of a portion, even a large portion, of plaintiff/developer’s land during the litigation does not necessarily mean that a remedy is no longer sought as to the remainder, and the mere fact that plaintiff *327continues to pursue the litigation indicates that indeed it is. Furthermore, a plaintiff in these actions, given the very broad rule we announce today on standing in Mount Laurel litigation, may continue to pursue an action simply to vindicate the Mount Laurel right without seeking a builder’s remedy.66

The only effect a disposition of a portion of a plaintiff’s land should have is on the appropriateness and form of the builder’s remedy. As we have noted elsewhere, a builder’s remedy is no longer to be considered “extraordinary.” It is to be given where appropriate, in view of our perception that it is one of the most effective tools for implementing Mount Laurel. A builder who has endured intensive litigation and succeeded in vindicating the Mount Laurel right in the interest of the public should not be deprived of his remedy simply because, during the course of litigation, he sold some of his land and thereby was required to revise his plans.67 It is true that such a disposition could well result in some further litigation, but if such further litigation were to be the basis for denying the builder his remedy, then the price of that remedy would consist not only of the enormous risks imposed by the litigation itself, but also of a freezing of all of the builder’s land from any advantageous disposition during the pendency of litigation lest it be totally disqualified from any builder’s remedy at all.

Plaintiff’s counsel has represented to this Court that plaintiff still seeks a builder’s remedy; that given the fact that the Gobel tract has been disposed of, the zoning of Beaverbrook is no *328longer satisfactory to it; that it seeks permission to build at higher densities on Beaverbrook than now permitted; that it understands it must comply with environmental regulations and generally construct an environmentally acceptable development; and that it must provide a substantial portion of low and moderate income units in that development. We will treat these questions later, as we will the specific problems of the builder’s remedy sought, but mention them now only as those factors bear on the Appellate Division’s view of the matter. Even if a builder’s remedy were inappropriate, for any reason, we would rule on the other issues before us as long as plaintiff demonstrated some legitimate interest in vindicating the Mount Laurel right.

In view of the conclusions stated above, we now deal with those issues that the Appellate Division dismissed as moot, namely, the correctness of the trial court’s findings concerning Clinton’s Mount Laurel obligation and the propriety of the remedy granted, including the appointment of a master.

3. Analysis of Clinton’s Fair Share Responsibility

(a) The Fair Share Determination

As we explained above, the “developing” characterization used by the trial court has been discarded as a remedial standard in favor of the “growth” characterization contained in the SDGP. Clinton Township, unlike the bulk of Hunterdon County, contains a substantial growth area, and is part of the SDGP’s “Clinton Corridor.” The “Current Development Character” of this Corridor is defined as follows:

This corridor extends westward from the Northeast Area along Interstate 78 to Clinton. The area includes older centers such as Somerville, Raritan and Clinton, but much of the land is either open or developed at very low densities. Many communities are within easy reach of northeastern New Jersey and New York employment centers by improved highways and interstates. [SDGP at 55],

In keeping with the planned growth policies of the SDGP, we hold that Clinton Township has the obligation to provide a realistic opportunity for the construction of its fair share of the *329region’s lower income housing need. A determination of the extent of that fair share obligation shall be made by the trial court on remand, consistent with the principles and procedures outlined in this opinion.

Our remand for a fair share determination does not reflect disagreement with the fair share findings below. On the contrary, those findings were well reasoned and solidly grounded in the evidence at trial. We remand on this issue for the sole purpose of allowing the trial court to determine whether our rulings concerning the SDGP have any impact on what would otherwise be an unassailable holding as to Clinton’s fair share. The SDGP placed approximately 40 percent of the township in a “growth” area, the balance being “limited growth” and “agriculture areas.” On remand the trial court should determine whether the fair share can be accommodated completely in the growth area consistent with sensible planning. If it can, then the fair share determination below shall stand; if not, it shall be revised appropriately. The trial court need not be concerned with the general growth pressure that any development in a “growth” area may exert on the neighboring “limited growth” or “agriculture” area, since those pressures are implicit in and presumably acceptable to the State Plan. They are, obviously, inevitable. Such general pressures, however, are to be distinguished from the site specific pressure of locating a large-scale development in such a fashion in a growth area as to make it highly likely that growth will occur where it is intended not to, namely, in the “limited growth” area. These matters are best left to the municipality and planners in redesigning the zoning ordinance. In that connection, the revised ordinance should obviously be tailored to encourage lower income housing only in the “growth” area.

Once the trial court has determined Clinton’s fair share obligation, it must ensure that this obligation will be met, using the remedies and procedures set forth above. We will comment separately here on the builder’s remedy and the master.

*330(b) The Builder’s Remedy

As previously explained, builder’s remedies will no longer be “rare” and will be granted as a matter of course where (i) the plaintiff-developer will provide a substantial amount of lower income housing, and (ii) the proposed project accords with sound land use planning. Satisfaction of both of these preconditions may be difficult in this case.

As noted above, the topography of the Beaverbrook tract (which plaintiff still owns) makes construction much more difficult than on the Gobel tract (which plaintiff sold), especially if costs are to be kept down so that construction of lower income housing is economically feasible. In fact, plaintiff had originally planned to construct its lower cost housing only on the Gobel tract and may be unable to construct it on Beaverbrook. If, on remand, the trial court finds that Round Valley in fact has no plans to build lower income housing on its remaining Beaver-brook tract, no builder’s remedy should be granted.

We emphasize that it will not be sufficient for plaintiff to show that its proposed development on Beaverbrook will supply “least cost housing.” Our prior observations that least cost housing cannot be used to satisfy the Mount Laurel obligation unless and until it is demonstrated that the numerous affirmative devices set forth above cannot produce actual lower income housing relates to the overall ordinance and its validity; it is only in that respect that “least cost housing” is an acceptable substitute for the Mount Laurel obligation. When it comes to a builder’s remedy, however, there is no substitute for low and moderate income housing. If the builder wants his remedy, he must prove to the satisfaction of the trial court that lower income housing, not just least cost housing, will actually result from it. Obviously, this may require very high density construction, subsidies, mandatory set-asides, and other devices; but whatever is required, lower income housing must result. What a “substantial portion” is will be up to the trial court. We do not mean to indicate that the presence of a substantial amount *331of “least cost housing” will not be influential in persuading the trial court that the prevision of a lesser amount of lower income housing will be satisfactory. One may influence the other.

As for the suitability of plaintiff’s tract, there was some evidence offered suggesting environmental problems in the construction of multi-family dwellings on this property.68 Therefore on remand there shall be further inquiry by the trial court as to the environmental factors affecting Beaverbrook in view of whatever plaintiff’s new plans may be, for those plans have not yet been formulated. When these plans have been completed and the trial court is satisfied that they will indeed provide a substantial portion of low and moderate income housing, the environmental issue shall then be determined, including, as the trial court did in its initial decision, requirements that the appropriate environmental agencies pass on the development. We emphasize again that the mere fact that there may be a better piece of land for this kind of development does not justify rejection of plaintiff’s builder’s remedy. It is only if the proposed development of Beaverbrook is contrary to sound planning principles, or represents a substantial environmental hazard, that it should be denied.

If, on remand, plaintiff is awarded a builder’s remedy, the trial court shall determine whether the scope of the proposed project would effect a “radical transformation” of Clinton and, *332if so, whether it should be “phased-in” at a rate slower than now planned by plaintiff.

(c) The Master

The appointment of a master by the trial court was not “premature,” contrary to the Appellate Division’s determination. As stated above, we believe that upon invalidation of an ordinance and the issuance of an order to rezone, trial courts may, in their discretion, freely appoint a master to aid in the implementation of their order, and need not await “voluntary compliance” on the part of the municipality, as suggested by the Appellate Division. We disapprove, however, of the apparent delegation of judicial power by the trial court to the master in this case. It must in the end be the judge and not the master who decides what actions will constitute compliance with our mandate. In this case it is quite clear, at least from the trial court’s language, that the court intended simply to adopt the master’s determination as to compliance.

We therefore reverse the Appellate Division judgment and remand to the trial court for proceedings consistent with this opinion.

E. Urban League of Essex County v. Township of Mahwah

1. Procedural and Factual Setting

Exclusionary zoning litigation involving Mahwah Township has, in one form or another, been going on for more than a decade.69 In the present action, plaintiffs, Urban League of *333Essex County, the North Jersey Community Union, and three individuals who seek housing in Mahwah (representing themselves and all others similarly situated), brought suit in February, 1972 against the northern Bergen County municipalities of Mahwah, Ramsey, Saddle River, and Upper Saddle River, alleging that these municipalities had failed to provide their fair share of low and moderate income housing as required by Mount Laurel. In December 1977, the Assignment Judge for Bergen County ordered a severance of the defendant municipalities. Plaintiffs decided to proceed initially against Mahwah only, and a trial was held in January and February of 1979.

On March 8, 1979, seven years after the plaintiffs had originally filed their complaint, the trial court, in an unreported decision, dismissed plaintiffs’ claims. While finding that Mah-wah Township was a “developing” municipality and therefore had an obligation to provide opportunities for lower income housing under Mount Laurel and Madison, the court concluded that this obligation was being met by the Township’s “bona fide” efforts to provide least cost housing through its new multi-family and mobile home zones. The court rejected plaintiffs’ contention that the housing that could be built in Mah-wah’s new zones was not “least cost” (it was being priced at $70,000 and above). It was the court’s finding that it was simply impossible to build housing at lower costs in northern Bergen County. Finally, the court refused to order Mahwah to take affirmative action to assure the construction of lower income housing, reasoning that such affirmative action was not required by this Court’s decisions in Mount Laurel and Madison.

On direct certification before this Court, plaintiffs’ basic contention is that Mahwah’s provision for very expensive multifamily dwellings should not be considered sufficient to meet the Township’s Mount Laurel obligation. Instead, they maintain *334that the trial court should have ordered Mahwah to take the necessary affirmative steps to insure that the housing built would actually be affordable to low and moderate income families. Finally, plaintiffs argue that the trial court erred by basing its finding that lower cost housing could not be built in Mahwah largely upon the judge’s own personal experience.

Although the municipality urges affirmance of the trial court’s decision, it disagrees with it in two respects. First, Mahwah had urged the trial court, and urges this Court now, to deny standing to the organizational plaintiffs in this case (the Urban League of Essex County and the North Jersey Community Union). Mahwah relies upon the United States Supreme Court’s holding in Warth v. Seldin, 422 U.S. 490, 95 S.Ct. 2197, 45 L.Ed.2d 343 (1975), that such organizations do not have standing in federal court to challenge allegedly exclusionary zoning ordinances. The defendant did not challenge the standing of the individual plaintiffs before the trial court because their standing had been approved in Urban League of Essex Cty. v. Township of Mahwah, 147 N.J.Super. 28 (App.Div.), certif. den., 74 N.J. 278 (1977). However, defendant now argues that this approval may no longer be justified because it was based primarily on the fact that the plaintiffs worked at the Ford plant in Mahwah, which is now closed.

The defendant also disagrees with the trial court’s characterization of Mahwah as a “developing” municipality within the ambit of the Mount Laurel doctrine. The Township claims that it is composed of two discrete areas, one of which is rural and the other of which is fully developed, both, therefore, outside the scope of Mount Laurel.

In urging this Court to affirm the trial court’s holding that it has met any Mount Laurel obligation it may have, the defendant emphasizes its agreement with the trial court’s refusal to order affirmative remedies. Mahwah maintains that such remedies would be either ineffective in actually producing lower income housing (e.g., density bonuses) or of questionable legality (e.g., *335mandatory set-asides). Mahwah further contends that the finding that lower cost housing cannot be built in northern Bergen County rests squarely on the evidence.

Mahwah is a 25.7 square mile (16,450 acre) municipality at the northwest corner of Bergen County. Because large areas are wilderness or devoted to public uses (e.g., Ramapo College), only 6,800 of these acres are developed or developable. Of these, between 1,400 and 1,700 are still vaeant. Mahwah is about 35 miles from Newark and New York City. Its population has been growing rapidly from 7,376 in 1960 to 10,800 in 1970 to 12,130 in 1980. The population density of 471.9 people per square mile is significantly lower than Bergen County’s 3,604 and the state’s 983.3.

Mahwah’s infrastructure, though not as extensive as that of communities closer to major urban areas, is relatively well developed. The Township has its own public water system and is in the first stages of providing public sewers for its developed areas. Mass transit is readily available between Mahwah and the urban centers to its southeast. The Township is traversed by at least two major highways, Routes 17 and 507, and has 2.77 linear miles of roads per square mile compared with 7.96 linear miles per square mile in Bergen County and a statewide mean of 10.24.

The challenged zoning ordinance, adopted in 1976, divides the Township into 11 districts. Six of these districts permit single-family housing only: R-80 (80,000 square foot lots); R-40 (40,000 square foot lots); R-20 (20,000 square foot lots); R-10 (10,000 square foot lots); R-5 (5,000 square foot lots); and C-80 (a Conservation Zone District that was largely undevelopable but that set aside certain low density zones that permit no more than one unit per 80,000 square foot lot).

Five districts permit multi-family housing. GA-200 permits garden apartments to be built on 20,000 square foot minimum lots. The number of units per lot depends on the number of bedrooms per unit: the more bedrooms, the fewer units. The *336ordinance also requires two off-street parking spaces per unit, half of which must be inside a garage. R-ll permits two-family dwellings to be built on 11,000 square foot lots. PRD-4 and PRD-6 allow for planned residential developments with overall densities, respectively, of four and six units, per acre. The planned residential development must be a specified mixture of single-family dwellings, garden apartments, and townhouses. Developers are required to amass a 50 acre lot as a prerequisite to developing any land within this zone, although the Planning Board has discretion to allow 25 acre plots to be added to existing developments. An underlying use of the PRD zones is R-20, without the requirement that the developer amass 50 acres. RM-6 permits mobile homes or, in the alternative, single-family houses on 40,000 square foot plots. Mobile home parks are permitted with 400,000 square foot minimum lot sizes for each park.

In addition to the above districts, there is a CED zone. This zone, which is covered by a separate ordinance, provides for timed commercial development and for the-construction of 565 housing units.

Mahwah’s zoning provisions, when coupled with the present economic situation, have resulted in extremely high housing costs in the Township. The trial court termed the housing costs throughout Bergen County as “astronomical.” Plaintiffs’ evidence established that 58.3 percent of Bergen County families and 70 percent of New Jersey families as a whole cannot afford to pay more than $50,000 for housing, but only 15 percent of the housing in Mahwah would sell below that price. Also, 38.8 percent of Bergen’s population and 53 percent of New Jersey’s population cannot afford to pay more than $40,000 for housing, but only 5.6 percent of Mahwah’s residential units would sell below that price.

2. Standing to Challenge Mahwah’s Ordinance

Before proceeding to a discussion of Mahwah’s fair share obligations, we address the Township’s claim that plaintiffs lack *337standing in this case due to the closing of Ford’s Mahwah plant. As mentioned previously, the Appellate Division upheld the standing of the individual non-resident employees at an earlier stage of litigation, and found it unnecessary to rule on the Urban League’s standing. The Township claims that the closing of the Ford plant undercuts the individual plaintiffs’ right to sue, and therefore makes dismissal appropriate, since they suggest that the Urban League lacks standing to continue the suit on its own. We disagree, and intend to clarify the standing doctrine as it applies to exclusionary zoning cases.

As this Court pointed out in Crescent Park Tenants Ass’n v. Realty Equities Corp. of N.Y., 58 N.J. 98 (1971), New Jersey cases have “historically taken a much more liberal approach on the issue of standing than have the federal cases.” Id. at 101. New Jersey courts, we noted, have never allowed “procedural frustration” to prevent determinations on the merits where the plaintiff can demonstrate a legitimate interest in the lawsuit. Id. at 107-08. See also State v. Alston, 88 N.J. 211, 225-230 (1981); N.J. Chamb. Commerce v. N.J. Elec. Law Enforce. Comm., 82 N.J. 57, 67-69 (1980); Home Builders League of So. Jersey, Inc. v. Twp. of Berlin, 81 N.J. 127, 131-32 (1979).

We believe that the need for a “liberal approach” to standing is especially important in Mount Laurel litigation. The people who have the greatest interest in ending exclusionary zoning, non-resident poor people and organizations such as the Urban League, -which represent the interests of such people, very often have little or no direct relationship with particular exclusionary municipalities. In fact, the whole problem is that exclusionary zoning prevents such relationships from developing. Thus, we hold that any individual demonstrating an interest in, or any organization that has the objective of, securing lower income housing opportunities in a municipality will have standing to sue such municipality on Mount Laurel grounds.

In Mahwah itself, we agree with the Appellate Division’s holding in Urban League of Essex Cty. v. Township of *338 Mahwah, 147 N.J.Super. 28, certif. den., 74 N.J. 278 (1977), that the individual plaintiffs had standing to sue. We find it irrelevant that the Mahwah Ford plant at which these plaintiffs worked is now closed, since the plaintiffs still express an interest in securing lower income housing opportunities in Mahwah. Further, we hold that the Urban League plainly has standing to sue in this case since, in effect, it represents those nonresident poor most in need of expanded lower income housing opportunities in suburbs such as Mahwah.

3. Determination of Mahwah’s Fair' Share

We turn to the trial court’s disposition of Mahwah’s Mount Laurel obligation. The trial court, after deciding that Mahwah was “developing,” held that the Township satisfied the resulting Mount Laurel obligation by providing “least cost” housing. It rejected plaintiffs’ claim that housing costs could be reduced in Mahwah so that housing could be constructed at prices affordable by lower income families, stating instead that the judge’s personal knowledge of the economic situation led him to conclude that the housing available reflected a bona-fide attempt at achieving “least cost.”70 We disagree with both the trial court’s approach and ultimate conclusion, and therefore reverse and remand.

On remand the trial court will immediately appoint an expert to assist the court in the fair share hearing, which shall be conducted in accordance with this opinion. Assuming the Township does not challenge the SDGP characterization of Mahwah as containing a growth area, and assuming the court finds that Mahwah’s present ordinance does not provide a realistic opportunity for the construction of its present and prospective fair share of the regional lower income housing need (and this seems certain from the record before us), it will enter a *339judgment invalidating the ordinance along with an order requiring Mahwah to revise its ordinance to achieve compliance. To the extent necessary, that revision should eliminate the cost-generating provisions that have made “least cost” housing impossible to construct. That order should require Mahwah fully to utilize the remedial devices set forth above, and should include appointment of a master to aid in the amendment of Mahwah’s ordinance. Mahwah, which contains a growth area under the SDGP,71 will have to document its efforts at providing lower income housing opportunities (efforts that must include use of affirmative devices where necessary), and then if, and only if, provision of such opportunities is shown to be impossible, it will be required to prove that the opportunities it does provide are in fact “least cost.”

We therefore reverse the trial court’s judgment and remand for further proceedings consistent with this opinion.

F. Urban League of Greater New Brunswick v. Borough of Carteret

This case provides a fitting conclusion to our opinion in these matters. The action was started in 1974. Plaintiffs proved beyond any question that there was a present actual need for low and moderate income housing in the 23 Middlesex County municipalities initially joined as defendants and that this need would become overwhelming in the future. They proved a pattern of exclusionary zoning that was clear. They portrayed a county exploding with growth, providing jobs for all, and promising even more in the future, including employment for low and moderate income families, a county where the opportunity for lower income housing shrank faster than its need grew.

*340Armed with substantial documentation of the need, the exclusionary practices, and the obvious ability of the municipalities to absorb any reasonably calculated fair share of the region’s need for lower income housing, the trial court conscientiously attempted to determine the precise regional need and its allocation among the municipalities. In the process it effectively settled the matter as to 11 of the municipalities that have apparently satisfactorily modified their ordinances to comply with Mount Laurel. As to the remaining 11 (one was dismissed as being fully developed), it entered an order that, while general, included substantially all of the arsenal of devices available as requirements imposed upon those municipalities to ensure compliance with their constitutional obligation.

As noted later, we disagree in certain important respects with the trial court’s decision, but it is otherwise a basically solid decision based on an opinion that rings with the true sound of the constitutional obligation. The recitation of the facts, 142 N.J.Super. at 20-35, convinces us of the essential correctness of the court’s determinations of the need for housing, the exclusionary practices that contributed to that need, and the remedies to satisfy it, as well as the necessity for judicial action in this area.

Before the appeal was disposed of, we rendered our decision in Madison. The impact of the holding and the spirit of Madison on Mount Laurel cases is nowhere better illustrated than in this case. The Appellate Division’s reversal, understandable in view of Madison, delivers a clear message to the trial bench in Mount Laurel cases, and the message is “hands off.” Taking Madison and some further restrictive language in Paseack as its guide, the Appellate Division read our language in Mount Laurel and Madison suggesting that a county was probably too small to comprise a region and held that a county could never be a region. The effect of this holding was not to increase the fair share of the municipalities (as might be the case with the expansion of this particular region) but to dismiss plaintiffs’ claims. The Appellate Division further held that our language *341in Madison relieving the trial court of the obligation to determine a precise numerical fair share for defendant-municipalities meant that the trial court was prohibited from making any such finding or imposing any numerical fair share in its judgment. And to complete the process, it further held that in any event, it was inappropriate to impose any requirements until the municipalities had had an opportunity to “act without judicial supervision.”

As far as the municipalities are concerned, the lesson of all of this litigation is that the Mount Laurel obligation is a matter between them and their conscience.

If, after eight years, the judiciary is powerless to do anything to encourage lower income housing in this protracted litigation because of the rules we have devised, then either those rules should be changed or enforcement of the obligation abandoned.

1. Procedural and Factual Setting

On July 23, 1974, plaintiffs (the Urban League of Greater New Brunswick and 7 individuals representing themselves and others similarly situated) brought this action against 23 of the 25 municipalities in Middlesex County.72 Plaintiffs alleged that the zoning ordinances of these municipalities failed to provide realistic opportunities for low and moderate income housing as required by Mount Laurel and were discriminatory against blacks in violation of the Thirteenth and Fourteenth Amendments to the United States Constitution. The latter federal claim was rejected by both courts below and it does not appear that it is being pressed before this Court.

*342The defendants responded with a four-pronged procedural challenge to the suit as brought, alleging that plaintiffs lacked standing to sue, were improperly joined as a class, and had failed to exhaust the administrative remedies available to them, and that the defendant municipalities were also joined improperly. The first three claims were rejected by the trial court at an early stage of the proceedings,73 and a pretrial hearing was conducted on the issue of severance.

After hearing testimony from plaintiffs, defendants and ami-cus Public Advocate, the trial court denied the motion for severance. It determined that since plaintiffs were alleging that defendants (except for Perth Amboy and New' Brunswick) were collectively responsible for county-wide exclusion of lower income residents, proof of such allegations required the joint presence of all defendants. The trial court also concluded that a total severance would unduly burden plaintiffs74 and might impair the court’s ability to design individual municipal remedies for a regional problem. However, the court also concluded that certain proofs required individual treatment, in particular, proofs regarding the specific provisions of each ordinance being attacked and the individual justifications each municipality *343might offer for its failure to provide lower income housing opportunities. In an effort to accommodate these and other competing considerations, the trial court ordered a bifurcated trial.

During the first part of the trial, plaintiffs submitted proofs on the appropriateness of Middlesex County as a lower income housing allocation region, on the inability of individual plaintiffs to find adequate lower income housing in Middlesex County, on the general need for lower income housing in Middlesex County, and on the exclusionary nature of zoning in the county as a whole outside of Perth Amboy and New Brunswick.75 Based on these proofs, the trial court concluded that there was an unmet need for lower income housing in the County, that the exclusionary zoning practices of the defendant municipalities played a large role in preventing this need from being met, and that *344Middlesex County is an appropriate “region” for lower income housing allocation purposes.76

At the second part of the trial, plaintiffs isolated five aspects of the municipalities’ ordinances that they claimed to be exclusionary:

a. Prohibitions on multi-family housing and mobile homes:

(1) Cranbury, East Brunswick, Helmettá, Monroe, North Brunswick, Piscataway, Sayreville, and South Plainfield prohibit mobile homes. Edison, Middlesex, Plainsboro, South Brunswick, and Spotswood have severe restrictions on mobile homes.

(2) Cranbury and South Plainfield prohibit any new multifamily housing. Monroe permits multi-family units only in its retirement community. Carteret, East Brunswick, Edison, Jamesburg, Middlesex, Milltown, North Brunswick, Old Bridge, Piscataway, Plainsboro, Sayreville, South Amboy, South Brunswick, South River, and Woodbridge place restrictions on multifamily housing.

b. Restrictions on development that are not necessary for health and safety or required by good planning practices: e.g., low densities per acre, excessive lot sizes, excessive frontages, and excessive square footage for buildings themselves. Plaintiffs claim all the municipalities now before the Court have numerous instances of such excessive standards.

c. Bedroom restrictions:

(1) North Brunswick and Plainsboro placed limitations on the number of two bedroom units allowed and prohibit larger units.

(2) Carteret, Highland Park, Middlesex, South Amboy, Spots-wood, and Woodbridge limit the number of rooms or bedrooms in multi-family housing.

(3) Old Bridge and Piscataway in effect limit apartments to efficiencies and one bedroom units.

*345(4) East Brunswick, Sayreville and South Brunswick have low-density requirements that limit the size of units.

(5) Edison limits the number of high rises.

d. Overzoning for industry and too much large-lot zoning:

(1) Cranbury, East Brunswick, Edison, Monroe, North Brunswick, Old Bridge, Plainsboro, Sayreville, South Amboy, South Brunswick and Spotswood are overzoned for industry up to as much as 700 percent over projected demand.

e. The discretionary authority vested in municipal authorities to discourage housing developments:

(1) Edison’s cluster options are subject to the discretion of municipal officials.

(2) East Brunswick’s Planning Board has the discretionary authority to assess improvement fees against builders.

On May 4, 1976, two years after the complaint was filed and after a trial that lasted two full months, Judge Furman handed down a two part decision. 142 N.J.Super. 11 (Ch.Div.1976). First, he divided the defendant municipalities into three categories:

a. All claims against the Township of Dunellen were unconditionally dismissed because the court found Dunellen to be a “developed” community outside the scope of Mount Laurel. Plaintiffs did not appeal this dismissal.

b. Claims against 11 communities (Carteret, Helmetta, Highland Park, Jamesburg, Metuchen, Middlesex, Milltown, South Amboy, South River, Spotswood and Woodbridge) were dismissed, conditioned upon the adoption of certain zoning amendments by these municipalities. The court noted that these municipalities were substantially built up, though not quite “developed” for Mount Laurel purposes. The plaintiffs agreed at trial to these conditional dismissals.

*346c. The zoning practices of the remaining 11 municipalities (Cranbury, East Brunswick, Old Bridge (formerly Madison), Monroe, North Brunswick, Plainsboro, South Brunswick, Sayre-ville, Edison, Piscataway and South Plainfield) were found to be in violation of Mount Laurel. In order to remedy these violations the court ordered each of the municipalities to provide a realistic opportunity for the construction of the fair share of the lower income housing allocation determined by the court.

The trial court determined numerical fair share allocations for the 11 municipalities found to be in violation of Mount Laurel. 142 N.J.Super. at 35-38. It accepted plaintiffs’ suggestion that Middlesex County was the appropriate housing region for remedial purposes. It concluded that each of the municipalities should accommodate low and moderate income people in proportions equal to that of the County as a whole — 15 percent low income and 19 percent moderate income. After finding the prospective low and moderate income housing need of the County through 1985 to be 18,697 units, the trial court allocated 4,030 of these units to remedy imbalances in the provision of low and moderate income housing among the 11 municipalities themselves as follows:

Cranbury 18

East Brunswick 1,316

Edison 1,292

Monroe 23

North Brunswick 180

Old Bridge 301

Piscataway -0-

Plainsboro -0-

Sayre ville 328

South Brunswick 156

South Plainfield 416

The remaining 14,667 units of the County’s lower income housing need were allocated equally among the 11 municipalities, i.e., 1,333 units per municipality. All units allocated to each municipality were to be apportioned as 45 percent low income housing and 55 percent moderate income housing. 142 N.J.Super. at 37.

*347The trial court emphasized that these 11 municipalities were required to do more than just refrain from zoning out their fair share allocation of lower income housing. Affirmative steps to encourage the construction of lower income housing, such as utilizing mandatory set-asides and density bonuses and pursuing federal and state housing subsidies, were required. All but 4 of these 11 municipalities (Old Bridge, North Brunswick, Edison and Sayreville), appealed.

On appeal, the municipalities claimed that the trial court erred by using Middlesex County as a region for fair share purposes. They maintained that this. Court specifically rejected the use of counties as regions in both Mount Laurel I and Madison. This claim was pressed most vigorously by the townships on the outskirts of the County, particularly Plainsboro and South Plainfield, who argued that their fair share allocations could not be properly arrived at without a consideration of the housing needs and obligations of municipalities in adjacent counties.

The following individual defenses were raised by the seven municipalities on appeal:

a. South Plainfield claimed it was “developed” and therefore not bound by Mount Laurel.

b. Cranbury, Monroe, and Plainsboro claimed they were “rural” communities and therefore not bound by Mount Laurel.

c. Plainsboro also claimed that it had made “bona fide” efforts to provide low and moderate income housing, and therefore had complied with Madison.

d. Piscataway claimed to have met its fair share obligation and noted that it had amended its zoning ordinance to comply with the trial court’s order.

*348e. South Brunswick claimed that it met its obligation under Mount Laurel through “non-exclusionary timed growth.”

f. East Brunswick claimed that it had satisfied its Mount Laurel obligations by providing for 1,750 units of lower income housing.

2. The Appellate Division Opinion

The Appellate Division, in an opinion reported at 170 N.J.Super. 461 (1979), reversed the trial court’s order allocating to the 11 municipalities a specific fair share quota of lower income housing. The Appellate Division based its decision on the “error” it found in the trial court’s decision using Middlesex County as the appropriate housing region. The court emphasized that this Court in both Mount Laurel I and Madison had criticized the use of counties as regions, and that the trial court in Madison had specifically refused to use Middlesex County as the housing region for Madison Township (now Old Bridge). Id. at 475. The court concluded that since the plaintiffs had failed to prove an appropriate housing region, they could not have established that the defendant municipalities’ zoning ordinances were impermis-sibly exclusionary. Id. at 476-77. The Appellate Division reversed the trial court’s order completely and dismissed the plaintiffs’ claims.77

We believe that the Appellate Division erred, both in its interpretation of the lower court’s opinion and in its failure to remand the case to that court for further proceedings. It is *349clear from a careful reading of Judge Furman’s detailed opinion that he did not intend to use Middlesex County as the region but simply took advantage of the simplicity of county boundaries in designing a fair share remedy in the case before him. Furthermore, while we noted in Madison that counties did not appear to be realistically suitable for use as regions, we did not preclude their use for that purpose, 72 N.J. at 537, making a summary reversal inappropriate. We therefore reverse the Appellate Division’s judgment and remand for a full fair share hearing involving the seven municipalities now before this Court.78

Our assumption is that Judge Furman, the author of a definition of region accepted by this Court that clearly would encompass more than Middlesex County, never intended to limit the measurement of housing needs by the boundaries of this particular County whose housing market is so clearly part of a larger region. Unfortunately it is impossible from the decision to determine precisely, or even generally, how the trial court arrived at its figure of 18,697 as the number of units of lower income housing needed by 1985. If that figure represented a portion of the larger region’s need, then it might be satisfactory; if, however, it was truly a calculation based solely upon the consideration of Middlesex County, then it might be subject to serious challenge. Again, we do not mean to exclude the possibility that a county might be suitable for characterization as a region under certain circumstances, and we would be reluctant to upset any trial court’s conscientious use of same unless convinced that the delay in meeting the need for lower income housing occasioned by such a reversal was clearly justified by the probable difference in figures that would result on remand. Here, however, there is no way of telling, other than *350the prior pronouncements of Judge Furman, whether Middlesex was viewed as the region, or simply, as suggested above, as part of a larger region, isolated for the purpose of devising a remedy.

Once the region’s need is determined, it is not advisable to allocate that need so as to equalize the proportion of lower income units in each municipality, nor to attempt to make that proportion the same as that found in the County. At one step in its provision of a remedy, the trial court did precisely that. The effect of such a remedy, of course, would be to make one municipality a demographic mirror image of another. That is not the purpose of Mount Laurel. Nor is there any justification for allocating a particular regional need equally among municipalities simply because they have enough land to accommodate such equal division. There may be factors that render such a determination defensible, but they would have to be strong factors, and certainly not the simple fact that there is enough land there. The issue in these cases is the overall group of factors that must be considered, all subsumed in the word “suitability.” Those factors have been described and need not be repeated here.

3. Procedure on Remand

In view of the uncertainty in the selection of regions and the calculation of regional need, as well as the error in allocating that need, we cannot accept the trial court’s resolution of this matter. As already indicated, however, we believe that the dismissal by the Appellate Division was inappropriate.

On remand there need be no trial concerning non-compliance with the Mount Laurel obligation (unless the municipality’s land use ordinance has been substantially amended), see supra at 199 n. 1, for that has already been amply demonstrated. All that is at issue is the determination of region, fair share and allocation and, thereafter, revision of the land use ordinances *351and adoption of affirmative measures to afford the realistic opportunity for the requisite lower income housing. Calculation of the prospective need will undoubtedly be projected beyond 1985, but the precise outside date will be left to the discretion of the trial court.

The trial court shall consider the existing record, and may also accept further proofs regarding changes in population, development, and land use regulation. It may appoint such experts as are required to assist it in determining region and the fair share allocation plan, and a master to aid the municipality in revising the ordinances and the court in passing on same. We leave the issue of continued joinder of the seven municipalities to the court’s discretion, noting only that there are obvious advantages and disadvantages to consider. See supra at 254-255.

In determining fair share, the trial court shall review the SDGP’s characterization of each of the municipalities before it. Our own examination of the Plan indicates that each includes growth areas and that some of them consist entirely of such an area. As previously stated, determination of fair share must take into consideration, where it is a fact, the inclusion within particular municipalities of non-growth areas where, according to the plan, growth is to be “discouraged.” See supra at 331 and 212. We believe that Plainsboro, Cranbury, South Brunswick, East Brunswick and Monroe all contain some non-growth as well as growth areas.

The judgment of the Appellate Division is reversed, and the matter is remanded for proceedings consistent with this opinion.

Conclusion

We have reexamined the Mount Laurel doctrine and we have found it correct. We have reaffirmed the judiciary’s commit*352ment to the enforcement of the constitutional right and its resulting remedy. We have found it necessary to rectify the ineffective administration of this doctrine in our courts. We have simplified the scope of litigation; the Mount Laurel obligation is to provide a realistic opportunity for housing, not litigation. We have substituted as a remedy the plans for growth in the State Development Guide Plan for the concept of developing municipalities, directed lower courts to dispose of the Mount Laurel litigation on a one-stop basis, and provided for the assignment of three judges to manage the cases statewide on a uniform basis. We have required municipalities to take affirmative action to comply with Mount Laurel and refocused the litigation on the question of whether low and moderate income housing will be built.

As we said at the outset, while we have always preferred legislative to judicial action in this field, we shall continue — until the Legislature acts — to do our best to uphold the constitutional obligation that underlies the Mount Laurel doctrine. That is our duty. We may not build houses, but we do enforce the Constitution.

The provision of decent housing for the poor is not a function of this Court. Our only role is to see to it that zoning does not prevent it, but rather provides a realistic opportunity for its construction as required by New Jersey’s Constitution. The actual construction of that housing will continue to depend, in a much larger degree, on the economy, on private enterprise, and on the actions of the other branches of government at the national, state and local level. We intend here only to make sure that if the poor remain locked into urban slums, it will not be because we failed to enforce the Constitution.

In Mount Laurel, we reverse the trial court and remand (but affirm as to the builder’s remedy); in both Chester and Franklin, we affirm the trial court in part but reverse and remand for *353limited further proceedings described herein; in Clinton, we reverse the Appellate Division and remand; in Mahwah, we reverse the trial court and remand; and in the Middlesex County cases, we reverse the judgment of the Appellate Division and remand. In all eases the remand is to the trial court for further proceedings consistent with this opinion.

For reversal and remandment in Nos. A-4, A-18, A-35/86 and A-37 — Chief Justice WILENTZ and Justices SULLIVAN, PASHMAN, CLIFFORD, SCHREIBER, HANDLER and POLLOCK — 7.

Opposed — N one.

For affirmance in part; reversal in part and remandment in Nos. AS and A-7/21 — Chief Justice WILENTZ and Justices SULLIVAN, PASHMAN, CLIFFORD, SCHREIBER, HANDLER and POLLOCK — 7.

Opposed — None.

For affirmance as modified in No. A-172 — Chief Justice WILENTZ and Justices SULLIVAN, PASHMAN, CLIFFORD, SCHREIBER, HANDLER and POLLOCK — 7.

Opposed — None.

*354 Appendix

*355

*356

*357

*358

*359

*360SCALE IN MILES

*361

*362

*363

*364

*365

*366

*367

*368

*369

*370

*371

*372

*373

*374