RESIDENTIAL BUILDERS ASSOCIATION OF SAN FRANCISCO v. San Francisco Campaign for Accountable Planning, Intervener and Respondent. (1989)

The charter for the City and County of San Francisco (charter) empowers any four members of the board of supervisors (one-third of its 11 members) to propose an ordinance for submission to city voters at the next general election.  (Charter, § 9.108(a), ¶ 3.)   Proposition M, a growth-control measure passed by voters in the November 1986 general election, reached the ballot in that manner after an identical measure failed, for formal deficiencies, to qualify as a voter-proposed initiative.

In this post-election action for writ of mandate, we examine the charter provision and conclude that supervisors may propose ordinances under it without first complying with the California Environmental Quality Act (CEQA) (Pub. Resources Code, § 21000 et seq.), the open-meeting provisions of the Ralph M. Brown Act (Gov.Code, § 54950 et seq.) or a charter provision requiring that zoning matters be referred to the city planning commission (charter, § 7.501).   We also conclude that parts of Proposition M which attempt to amend the city's master plan are invalid but severable and thus do not defeat the measure.

BACKGROUND

Proposition M, “The Accountable Planning Initiative,” began as a voter-sponsored initiative drafted and circulated by a coalition of voters calling itself the San Francisco Campaign for Accountable Planning (SFCAP).   Having gathered enough petition signatures to place the measure on the November 1986 ballot, SFCAP presented the petition to the city registrar of voters in late July 1986, on the last day for filing.   However, the registrar refused to certify it, citing the omission of a required “notice of intention” (see Elec. Code, § 4004).   SFCAP challenged the refusal in superior court, but the registrar's action was upheld.

With the deadline for filing a corrected petition past, an SFCAP executive committee member personally approached four members of the board of supervisors and persuaded them to propose the measure's inclusion on the ballot under the authority of the charter.   A letter signed by those supervisors (one of whom had to sign and mail in the letter from out of state) was accepted by the registrar on August 6, the last day for filing such proposals.   Only those supervisors supported the measure.   Six of the seven others, and then-Mayor Dianne Feinstein, openly opposed it.   The measure was approved as Proposition M on the November 4, 1986 general election ballot.

Proposition M has four parts.   Part 1 adds to the San Francisco Municipal Code (planning code) a section which establishes eight “Priority Policies,” requires that the planning commission amend the city's master plan to include them, and requires that certain zoning, development agreements, permit approvals and legislation be consistent with the master plan and the stated policies in the future.¹  Part 2 amends the planning code to limit total increased office development in any annual “approval period” to 950,000 square feet, downwardly adjusted by half each year until compensation has been made for certain development that exceeded that rate in years past (since inclusion of the “Downtown Plan” in the master plan in 1984).   Prior voter approval is required for all exemptions from the annual limit and any amendment or repeal of the limiting code sections.   Part 3 amends the planning code to mitigate previously found effects of new office development by encouraging, in various ways, the training and job placement of city residents.   The proposition also includes a severability clause (part 4).

Residential Builders Association of San Francisco (Residential Builders), an association of home builders, brought this action against the City and County of San Francisco, its board of supervisors and other agencies (hereafter collectively the city), urging that Proposition M is procedurally defective for failure to comply with CEQA, the Ralph M. Brown Act and the charter, and urging that the measure is substantively invalid under the charter for attempting to amend the city's master plan.   SFCAP intervened in the action by stipulation.   Residential Builders appeals from a judgment denying a peremptory writ of mandate.

APPEAL

I

All three procedural issues raised in this action hinge on a proper construction of the charter section allowing one-third of the supervisors to propose ballot measures.

San Francisco is a charter city and county with the right to implement for itself the reserved powers of initiative and referendum.  (Cal. Const., art. II, § 11, art. XI, § 6, subd. (a).)  The third paragraph of charter section 9.108(a) (hereafter § 9.108(a)) defines these ways, beyond voter-signed petitions, in which ordinances may be proposed to the voters:

“Any ordinance which the supervisors are empowered to pass may be submitted to the electors by a majority of the board at a general election or at a special election called for the purpose․   Any such ordinance may be proposed by one-third of the supervisors, or by the mayor, and when so proposed shall be submitted to the electors at the next succeeding general election․”  (Emphasis added.)

“Our primary aim in construing any law is to determine the legislative intent.  [Citation.]   In doing so we look first to the words of the [enactment], giving them their usual and ordinary meaning.  [Citations.]”  (Committee of Seven Thousand v. Superior Court (1988) 45 Cal.3d 491, 501, 247 Cal.Rptr. 362, 754 P.2d 708.)

Section 9.108(a) allows a measure to be proposed in three ways:  (1) by a majority vote of the board of supervisors (board);  (2) by one-third (at least four) of the supervisors or (3) by a mayor acting alone.   The first way implies a deliberative effort by the board as a whole, with the majority will prevailing.   The third way implies a decision by a mayor, independent of the board's action or will.

The issue that divides the parties in this case is how to characterize the second method.   Does it imply action by the board, with a lowered, minority-will vote requirement;  or, does it imply action by four supervisors acting ex officio but independently of the board?   The difference is important because the presence or absence of concerted action by the board affects the need for CEQA and Ralph M. Brown Act compliance, as well as the need for planning department referral under the charter (see parts II–V, infra).

The parties offer analogies.   Residential Builders urges that the charter creates a “voluntary referendum,” a term describing the process by which a legislative body, ordinarily by a majority or super-majority vote, submits a law directly to voters rather than first enacting the measure and then awaiting voter action.  (Comment, The Scope of the Initiative and Referendum in California (1966) 54 Cal.L.Rev. 1717, 1721.)   One author has used the term to describe the charter provision at issue here, although noting its novel, one-third-of-supervisors feature.  (Id., at p. 1721, fn. 26 [formerly charter § 179].)   To bolster that view, Residential Builders offers a conceptual distinction between initiatives and referenda based on sponsorship of the measure, arguing that initiatives originate with voters rather than some legislative body (or members of it) and that the charter therefore cannot be construed as an initiative.

The city and SFCAP, by contrast, view the charter as reserving a unique power of initiative by which four or more supervisors, acting ex officio but independently of the board, place the measure directly before the voters, without the more cumbersome process of signature gathering.

While the parties' analogies are interesting, it would not add anything to our analysis to label the charter device as “initiative” or “voluntary referendum.”   In a case where, for example, the issue was whether the board or voters exceeded constitutional authority as to subject matter by proceeding by referendum rather than by initiative, the distinction could be critical.   (Cal. Const., art. II, § 9, subd. (a) [exceptions to power of referendum];  see, e.g., Carlson v. Cory (1983) 139 Cal.App.3d 724, 730–731, 189 Cal.Rptr. 185.)   Here, however, it is not.   The only substantive challenge in this case turns on limitations imposed by the charter, and those limitations are unaffected by the differences between initiative and referendum (see part VI, infra).   It would be dictum to attach either label in this case, and we therefore decline to do so.   We note that this court, in a case concerning the same one-third-of-supervisors provision, characterized it as the board's “initiative” power (Clark v. Patterson (1977) 68 Cal.App.3d 329, 332–333 & fn. 3, 137 Cal.Rptr. 275), but the label was likewise unimportant in that case.   The pertinent question, we find, is whether the charter envisions concerted board action.

Looking to the words of section 9.108(a) itself (Committee of Seven Thousand v. Superior Court, supra, 45 Cal.3d 491, 501, 247 Cal.Rptr. 362, 754 P.2d 708), it is clear that something other than concerted board action is envisioned.   The section allows measures to be introduced both by a “majority of the board” and by “one-third of the supervisors.”  (§ 9.108(a).)   Residential Builders' position that the latter clause must be construed to mean concerted board action by a one-third, “special voting rule” cannot overcome three serious obstacles.

First, the construction would partially nullify the former clause.   There would be no need for a majority vote of six or more if a minority vote of four or more was enough.   Residential Builders notes that the majority vote would still have some independent vitality because it allows measures to be submitted at either a general or special election while the minority vote can be used only for general elections.   However, that still leaves the majority vote meaningless in general elections.   We must give effect to every phrase of the section if possible.  (Select Base Materials v. Board of Equal.  (1959) 51 Cal.2d 640, 645, 335 P.2d 672.)   Also, the distinction between special and general elections does not help much.   Whether or not concerted board action was envisioned, the drafters could have wanted the city to bear the expense of a special election only when a majority of supervisors felt it justified.

A second obstacle to finding concerted action is more basic.   If the charter envisioned concerted board action by a minority vote, then there would be situations where a majority opposed the measure but would be compelled to “act” according to the minority will.   Alternatively, it could mean that the board would deliberate and then “act” by proposing conflicting majority and minority measures.   A more sensible construction is that the minority acts independently of the board.

The third obstacle is that section 9.108(a) speaks of actions by “a majority of the board ” and actions by “one-third of the supervisors” (emphasis added).   Had concerted board action been envisioned, the drafters could have used parallel language—“one-third of the board.”  The use of different terms, so closely juxtaposed and on the same subject, supports the idea of action by four or more supervisors being distinct from board action.  (People v. Moore (1986) 178 Cal.App.3d 898, 903, 224 Cal.Rptr. 204.)   Changes in wording or phraseology are presumed to have been deliberately made.  (Estate of Simpson (1954) 43 Cal.2d 594, 600, 275 P.2d 467.)   This presumption is reinforced in this case because the same sentence which we are construing also provides that the mayor, without concurrence by any supervisors, may propose ballot measures.   This is a solid indication that the drafters intended measures to go on the ballot without board deliberation or action.   It is a phenomenal power to invest in one individual and makes the idea of investing that power in four individuals relatively unremarkable.   There is thus every reason to think that the drafters envisioned independent action when they wrote “one-third of the supervisors” as opposed to “a majority of the board.”   Nothing in Committee of Seven Thousand v. Superior Court, supra, 45 Cal.3d 491, 247 Cal.Rptr. 362, 754 P.2d 708, invoked by Residential Builders, persuades us otherwise.   There, the court noted that statutory use of the term “city council” or “board of supervisors,” as opposed to “legislative body,” strongly implies an intent to preclude an exercise of statutory powers by the electorate as opposed to the local legislative body.  (Id., at pp. 504–505, 247 Cal.Rptr. 362, 754 P.2d 708.)   Here, of course, the question is whether the term “supervisors” means collective rather than individual supervisor action—not whether it connotes a power of the electorate.   Also, as noted already, the term used here is “supervisors” rather than “board of supervisors” or “board,” thus suggesting a power independent of the full board.

Looking beyond section 9.108(a), to the other charter provisions of which it is a part (Select Base Materials v. Board of Equal., supra, 51 Cal.2d 640, 645, 335 P.2d 672), we see further signs that the supervisors act independently of the board.   Charter section 2.300 provides:  “Action by the board of supervisors shall be by ordinance or resolution in writing introduced by a member or by a committee of said board and passed or adopted by a majority of all the members of the board at each reading.   Every legislative act shall be by ordinance.”  (Emphasis added.)   Clearly, a proposal by only four supervisors does not meet this test of board “action” or, for that matter, “legislative act.”  (Cf. Clark v. Patterson, supra, 68 Cal.App.3d 329, 334–336, 137 Cal.Rptr. 275.) ²

We have not found any provision expressly authorizing board action by less than a majority, except for charter section 2.102, which directs that statewide general law, instead of the charter, applies should a “disaster ․ render[ ] unavailable a majority of [board] members․”   This indicates that board action under the charter (as opposed to general law) must be by majority or greater vote.   Section 9.108(a) does not appear to create a special, one-third “quorum rule.”

We believe that the only reasonable construction of the charter is that section 9.108(a)'s provision for a proposal by “one-third of the supervisors” contemplates action by four or more individual supervisors rather than concerted board action by minority vote.   Nevertheless, assuming for sake of argument that the charter is ambiguous and thus in need of construction by extrinsic means, we examine further arguments.

The provisions now contained in section 9.108(a) were originally added to the charter in 1912, and the parties agree that no legislative history is available.   However, because the provisions came into being at about the same time as the 1911 amendments adding initiative and referendum provisions to the state Constitution and because the state and local enactments were both fueled by the progressive movement of the early 1900's (Associated Home Builders etc., Inc. v. City of Livermore (1976) 18 Cal.3d 582, 591, 135 Cal.Rptr. 41, 557 P.2d 473), Residential Builders urges us to compare them.   While we find much of this comparison misdirected at trying to label the supervisors' power one of referendum rather than initiative, the comparison is still worthwhile.   We believe it shows that the charter was designed to surpass—not just mirror—constitutional provisions.

The state Constitution provides that all legislative power is vested in the state Legislature except that “the people reserve to themselves the powers of initiative and referendum” (art. IV, § 1).  “Initiative” is “the power of the electors to propose ․ and to adopt or reject” statutes or constitutional amendments (art. II, § 8, subd. (a));  “referendum” is “the power of the electors to approve or reject” certain statutes in whole or part (id., § 9, subd. (a)).   General law allows measures to be submitted either by direct, voter-circulated petitions (Elec.Code, § 4001) or by proposals from a city's legislative body (id., § 4017).   The charter allows both of those means but then goes further, providing for the exercise of power by one-third of the supervisors and by the mayor alone.   There is no like provision in state statutes, the Constitution or, as far as we know, any other current city charter.

The charter's departure from the general law model shows an intent to go further in the exercise of the powers of initiative or referendum, and its framers were constitutionally entitled to do so.   While initiative and referendum powers must generally be exercised “under procedures that the Legislature shall provide,” that constraint “does not affect a city having a charter” (Cal. Const., art. II, § 11).   A charter may provide for, among other things, the “conduct of city elections” (id., art. XI, § 5, subd. (b)(3)).   Such provisions become “the law of the State and have the force and effect of legislative enactments” (id., art. XI, § 3).  “[A]s between the provisions of the Constitution and the provisions of a city charter, those which reserve the greater or more extensive referendum [or initiative] power in the people will govern.”  (Hunt v. Mayor & Council of Riverside (1948) 31 Cal.2d 619, 623, 191 P.2d 426;  Hopping v. Council of City of Richmond (1915) 170 Cal. 605, 611, 150 P. 977.)   The initiative and referendum “are part of the progressive reforms of 1911 which provided the electorate with the tools of direct democracy thereby returning the government to the people” (Carlson v. Cory, supra, 139 Cal.App.3d 724, 731, 189 Cal.Rptr. 185);  courts must resolve any reasonable doubts “in favor of this precious reserved right of the people” (id., at pp. 731–732, 189 Cal.Rptr. 185;  Brosnahan v. Brown (1982) 32 Cal.3d 236, 241, 186 Cal.Rptr. 30, 651 P.2d 274;  Associated Home Builders etc., Inc. v. City of Livermore, supra, 13 Cal.3d 582, 591, 135 Cal.Rptr. 41, 557 P.2d 473).

 There is no room for serious doubt that the charter's provision for “one-third of the supervisors” to propose ballot measures (§ 9.108(a)) was intended to be exercised independently of the rest of the board rather than as a special, minority-vote rule for the board deliberating as a whole.³  Simply put, independent action, not board action, was contemplated by section 9.108(a).

Finally, as the city notes, the political climate in San Francisco at the time of the power's creation reinforces the idea that independent action was intended.   San Francisco had recently witnessed the protracted graft prosecutions of sitting Mayor Eugene Schmitz (People v. Schmitz (1908) 7 Cal.App. 330, 94 P. 407) and political boss Abraham Ruef (People v. Ruef (1910) 14 Cal.App. 576, 114 P. 54), prosecutions which exposed rampant corruption among the supervisors, 16 out of 17 of whom admitted bribe-taking under the Reuf–Schmitz machine.   The prosecutions were a major impetus for the statewide progressive movement.   Hiram W. Johnson, a special prosecutor in Ruef's trial, was elected governor in 1910.   Leading newly-elected progressives in the Legislature in 1911, he secured passage of the constitutional amendments that added the initiative, referendum and recall.⁴  (Crouch & McHenry, California Government, Politics and Administration (1945) pp. 81–82;  see generally Bean, Boss Ruef's San Francisco (1981 ed.).)

Against this background, in 1912, the charter was amended to add what is now section 9.108(a).   Its provision for ballot measures to be proposed by one-third of the supervisors or by a mayor acting alone must be seen as insurance against future dominance of the city's legislative processes by another Ruef-Schmitz-type regime.  (See Crouch & McHenry, op. cit. supra, at p. 90 [invoking this section during disputes with the board, Mayor James Rolph occasionally “contrived to ‘let the People decide’ ”].)   This confirms our view that the provision envisions supervisor action independent of the board.

On the eve of oral argument, Residential Builders offered a new legal theory which is at odds with its “voluntary referendum” theory.   We are asked to construe section 9.108(a) as conferring on four supervisors the power to refer only those ordinances that have already been passed or adopted by the full board.   Residential Builders urges, consistent with the result reached in Whitmore v. Carr (1934) 2 Cal.App.2d 590, 38 P.2d 802 (Whitmore), that an unpassed ordinance like Proposition M, although proposed to and ultimately adopted by city voters, is of no effect.   We reject the argument.

Whitmore construed this provision (§ 179) from the 1919 version of the Oakland Charter:  “ ‘Any ordinance which the council is empowered to pass may be submitted by an affirmative vote of three (3) of its members at a general municipal election only, subject to the provisions of this article so far as applicable.’ ”  (2 Cal.App.2d 590, 593, 38 P.2d 802.)   The provision was like San Francisco's in that it allowed less than a majority on the governing body to place measures before the voters, but there were important differences that account for the Whitmore court's construction of it as a power of referendum.   Most notably, the provision was found in a part of the charter dealing exclusively with referendum (former Oakland Charter, art. XXII, §§ 177–186 [titled “THE REFERENDUM”] ), and the provision itself was titled “Referendum by Council.” ⁵  The construction was reinforced by the charter's consistent use of the term “ordinance” throughout its referendum provisions and fairly consistent use of the term “proposed ordinance” in its initiative provisions (id., art. XXI, §§ 162–176).

Section 9.108(a), by contrast, is titled “Initiative, Referendum, and Recall” and does not have subheadings.   Its only subdivisions are unnumbered paragraphs addressing initiative petitions by voters (¶ 1), “declaration[s] of policy” submitted for voter approval or rejection (¶ 2), the provisions at issue here (¶ 3), referendum petitions by voters (¶ 4), matters not subject to referendum (¶ 5), and recall (¶ 6).   While we need not draw an initiative/referendum distinction in this case, we note that the section uses the words “propose” and “proposed” when addressing initiative (¶ 1) and “submit” or “submitted” when addressing referendum (¶'s 3 [ordinances effective in 60 days unless referendum petition filed] & 4).   The language which we are construing authorizes ordinances to be “submitted” by a board majority yet “proposed” by one-third of the supervisors.   This militates against a conclusion that only referendum is involved when less than a majority act.   The fact that section 9.108(a) limits its scope to “ordinance[s] which the supervisors are empowered to pass” (emphasis added) does not, as Residential Builders argues, connote prior enactment by the full board.   The same language (plus language embracing matters entrusted by charter to other boards, commissions and officers) appears in the voter-initiative paragraph.   The limitation on board- or supervisor-sponsored ballot measures appears designed to ensure that their proposal powers do not cross into legislative matters which the charter, with its unique checks and balances, assigns exclusively to others.  (See, e.g., power to amend the master plan, as discussed in part VI, infra).   It does not connote pre-enactment.

Finally, even if section 9.108(a) could be said to confer a referendum power on individual supervisors, it would not follow that the power could not be exercised without prior enactment by the full board.  Whitmore, to the extent that it implies this as a universal principle rather than one divined from a particular charter provision, goes too far.   General law authorizes the legislative body of a county or city to submit proposals to voters for “the repeal, amendment, or enactment of any ordinance” (Elec.Code, §§ 3750, 4017, emphasis added), language which necessarily implies that there may not have been any prior enactment.   As Residential Builders itself notes, the device has long been recognized as the “voluntary referendum.”  (31 Ops.Cal.Atty.Gen. 100, 100–101 (1958) [measure not binding if couched in advisory language and not reduced to a definite proposed enactment].)   To extend the Whitmore holding beyond its facts would bring about the invalidation of countless ordinances throughout the state.   The Election Code's use of the term “ordinance” also supports our view that the same term, as used in section 9.108(a), does not imply pre-enactment.

II

The city took no steps to comply with the provisions of CEQA before Proposition M was put on the ballot.   Residential Builders contends that the measure is invalid for that failure.   We disagree.

CEQA requires an investigation into environmental impact for “discretionary projects proposed to be carried out or approved by public agencies” (Pub. Resources Code, § 21080, subd. (a)) and not to any such projects that are “[m]inisterial” (id., subd. (b)(1)).   We agree with the city that the proposal in this case constituted neither a “project” nor an “approval.”

“Project” is defined in relevant part as “[a]ctivities directly undertaken by any public agency” (Pub. Resources Code, § 21065, subd. (a)), and “public agency” includes “any city and county ․ or other political subdivision” (id., § 21063).   State CEQA Guidelines promulgated under the act (id., §§ 21083, 21084) further define “project” as “the whole of an action, which has a potential for resulting in a physical change in the environment, directly or ultimately,” which includes a public agency's “enactment and amendment of zoning ordinances [or] General Plans or elements thereof ․” (Cal.Code Regs., tit. 14 (hereafter Guidelines), § 15378, subds. (a) and (a)(1)).   The Guidelines specify, however, that “project” does not include “[t]he submittal of proposals to a vote of the people of the state or of a particular community” (id., subd. (b)(4), emphasis added), which is our case.⁶

Acknowledging that the Guideline just quoted has been relied on to hold that submitting a voter-sponsored initiative does not constitute a project (Stein v. City of Santa Monica (1980) 110 Cal.App.3d 458, 460–461, 168 Cal.Rptr. 39), Residential Builders urges that the exception hinges on the discretionary-ministerial distinction for public agency action (Pub. Resources Code, § 21080) and thus should be read narrowly to mean that a legislatively-sponsored measure does constitute a project since in this case a public agency made a discretionary decision to place a measure on the ballot whereas placing a voter-sponsored measure on the ballot is purely ministerial.

The short answer to this long argument is that there was no “public agency” action in this case (Pub. Resources Code, § 21063).   All of the cases on which Residential Builders relies are distinguishable, at least on that score.  (E.g., Fullerton Joint Union High School Dist. v. State Bd. of Education (1982) 32 Cal.3d 779, 796, 187 Cal.Rptr. 398, 654 P.2d 168 [school board's approval of school secession plan not exempt under the Guidelines exception merely because it had to be ratified by the voters];  Bozung v. Local Agency Formation Com. (1975) 13 Cal.3d 263, 278–279, 118 Cal.Rptr. 249, 529 P.2d 1017 [LAFCO approval of city annexation plan was a project, despite need for later voter approval];  People ex rel. Younger v. Local Agency Formation Com. (1978) 81 Cal.App.3d 464, 476–477, 146 Cal.Rptr. 400 [same regarding deannexation plan].)  “Public agency” is defined in section 15379 of the Guidelines as including “any state agency, board, or commission and any local or regional agency, as [further] defined in these Guidelines.”   We know of no further definition that would bring four individual supervisors, acting independently of the board under their charter power, within the scope of “public agency.”

The answer to the question of “approval” is similar.  “ ‘Approval’ means the decision by a public agency which commits the agency to a definite course of action in regard to a project intended to be carried out by any person․”  (Guidelines, § 15352, emphasis added.)   There was no decision or commitment by any “public agency.”   The board was not consulted, did not make a decision and was not committed to any course of action, definite or otherwise.

III

Residential Builders contends that Proposition M is invalid for failure of the proposing supervisors to hold an open meeting under the Ralph M. Brown Act (hereafter Brown Act), and under an open-meeting provision of the charter, before placing the measure on the ballot.   We hold that compliance with neither provision was needed.

The Brown Act mandates:  “All meetings of the legislative body of a local agency shall be open and public, and all persons shall be permitted to attend any meeting of the legislative body of a local agency, except as otherwise provided in this chapter.”  (Gov.Code, § 54953, subd. (a).)  “Local agency” includes a “city and county ․ or any board, commission or agency thereof” (id., § 54951) and concededly includes the city.   The question for us is whether four or more supervisors acting under their charter authority constitute a “legislative body.”

“Legislative body” includes “the governing board, commission, directors or body of a local agency, or any board or commission thereof, and shall include any board, commission, committee, or other body on which officers of a local agency serve in their official capacity as members and which is supported in whole or in part by funds provided by such agency․”  (Gov.Code, § 54952.)   It also includes any permanent board or agency of the local agency (id., § 54952.5), any advisory body created by formal action of the legislative body (id., § 54952.3) and any board, commission or other multimember body exercising “any authority of a legislative body of a local agency delegated to it by that legislative body” (id., § 54952.2).

We have concluded that there is no concerted board action when one-third of the supervisors propose ballot measures under the charter, and we have rejected the notion that they act as a board minority or special quorum.   It follows that they do not act as a “legislative body” or have to hold open meetings under the Brown Act.   Residential Builders does not argue that the supervisors act as a permanent city board or as an advisory body created by formal action of the board of supervisors.

We are asked to hold that the supervisors acted as a body exercising the “authority of a legislative body of a local agency delegated to it by that legislative body” (Gov.Code, § 54952.2)—whether or not their action constituted concerted board action—since the charter limits proposed measures to subjects on which the board would be competent to act.   The statute, however, requires that the authority be delegated by the legislative body—the board.   The “delegation” here (in fact a “reservation” of the electorate's powers of initiative or referendum;  see charter, § 2.101) is by the charter.   This authority is limited to matters on which the board would be “empowered to pass” (§ 9.108(a)), but a charter reservation of board-like powers does not amount to delegation by the board.

Residential Builders relies on charter section 3.500(f), an amendment passed by voters as Proposition E in November 1972 and, according to ballot materials included in the record, intended to surpass the Brown Act's protections by requiring open meetings in situations where a quorum of a board or commission is not present.  (Compare Gov.Code, § 54952.6, defining “action taken” as “a collective decision made by a majority of the members”.)   It is unclear whether the section is invoked as a separate ground for invalidating Proposition M or simply as evidence that voters have repudiated any “expedited” ballot procedure inherent in section 9.108(a).   Either way, however, we are not convinced.   First, the ballot materials show an intent to affect only meetings of appointed boards and commissions, a scope which appears consistent with the charter section as a whole (see fn. 2, ante).   Second, the provision's call for open meetings for “all committees, whether composed of more than or less than a majority of the parent board or commission ․” (charter, § 3.500(f)), clearly does not apply to supervisors acting under the one-third provision of section 9.108(a).   Nothing in the charter suggests that supervisors acting under that power are a board “committee.”

IV

 Residential Builders contends that section 9.108(a), if construed as allowing noncompliance with CEQA and the Brown Act, conflicts with and is preempted by them since they cover matters of statewide importance, namely, environmental impact and open meetings.   We agree, of course, that the acts involve matters of statewide concern but disagree that the charter is preempted.

Residential Builders relies on the concept, applied in People ex rel. Seal Beach Police Officers Assn. v. City of Seal Beach (1984) 36 Cal.3d 591, 205 Cal.Rptr. 794, 685 P.2d 1145 (Seal Beach ), and Committee of Seven Thousand v. Superior Court, supra, 45 Cal.3d 491, 247 Cal.Rptr. 362, 754 P.2d 708 (Committee of Seven Thousand ), that a charter city's exercise of home-rule and municipal-election powers may be limited by state legislation on matters of statewide concern.   In Seal Beach, council-proposed charter amendments passed by city voters (Cal. Const., art. XI, § 3, subd. (b)) effectively altered the terms and conditions of city employment without satisfying the meet-and-confer requirement (Gov.Code, § 3505) of the Meyers–Milias–Brown Act (MMBA).   The Supreme Court held the amendments invalid, concluding that the council's power to propose charter amendments and the MMBA requirement were compatible and that the council therefore had to meet and confer with public employee unions before proposing the amendments.  (Seal Beach, supra, 36 Cal.3d at pp. 596–602, 205 Cal.Rptr. 794, 685 P.2d 1145.)   There was no conflict because the meet-and-confer requirement left the council ultimately free to refuse an agreement and thus propose the charter amendment if no agreement were reached.  (Id., at p. 601, 205 Cal.Rptr. 794, 685 P.2d 1145.)

In Committee of Seven Thousand, a voter-proposed initiative measure requiring voter approval before a charter city's counsel could impose development fees or taxes to finance new road construction conflicted with a state statute (Gov.Code, § 66484.3) allowing the council to do so without voter approval.   Finding that the statute involved statewide concerns, rather than purely municipal affairs, the Supreme Court held the proposed initiative preempted and thus invalid.   The presence of statewide concerns meant that the statute did not violate the city's constitutionally granted home-rule or initiative and referendum powers (Cal. Const., art. XI, § 5, art. II, § 11).  (Committee of Seven Thousand, supra, 45 Cal.3d 491, 509–512, 247 Cal.Rptr. 362, 754 P.2d 708.)  “In matters of statewide concern, the state may if it chooses preempt the entire field to the exclusion of all local control.   If the state chooses instead to grant some measure of local control and autonomy, it has authority to impose procedural restrictions on the exercise of the power granted, including the authority to bar the exercise of the initiative and referendum.  [Citations.]”  (Id., at p. 511, 247 Cal.Rptr. 362, 754 P.2d 708.)

These cases do suggest that the Legislature may invoke statewide concerns to limit the exercise of charter powers like those in section 9.108(a), and the goals of CEQA and the Brown Act are certainly of statewide concern.   However, while CEQA and the Brown Act could be drafted to interfere with the charter, our discussions in parts II and III of this opinion show that the acts so far have not been drafted that way.   We have seen that there is no conflict.   Each of them exempts from their scope the kind of activity that concerns us here.   By contrast, the meet-and-confer requirement in Seal Beach did not exempt charter amendments, and the delegation of road-funding authority in Committee of Seven Thousand apparently did not accommodate prior voter approval.

Residential Builders' “preemption theory” thus amounts to an argument that the Legislature has overlooked the city's unique charter provision and would have provided against its exercise had its effect been understood.   However, the statutes can be amended any time if that is truly the case, and we are not convinced that it is.

V

 Residential Builders contends that Proposition M, which proposed amending the planning code, had to be submitted to the city's planning commission under charter section 7.501, under which proposed land-use ordinances, or proposed amendments to them, are referred to the planning commission for approval before adoption by the board.   Section 302 of the planning code implements these procedures.

It is clear that this process was not meant to apply to ballot measures proposed by “one-third of the supervisors” under section 9.108(a).   As previously discussed, supervisors acting under that provision act independently of the full board, not as a special minority.   Charter section 7.501 speaks of proposals “initiated by the board of supervisors,” referred to the planning commission and then adopted by the “board.”   Proposition M was not initiated or adopted by the board.   It was “initiated” (proposed) by four supervisors and “adopted” (enacted) by the voters.   Only if the full board had proposed the proposition could it be argued that compliance with charter section 7.501 was required.  (See Terminal Plaza Corp. v. City and County of San Francisco (1986) 177 Cal.App.3d 892, 900–902, 223 Cal.Rptr. 379.)

VI

Part 1 of Proposition M adds eight “Priority Policies” to the planning code, requires the planning commission to amend the city's master plan to include them, and requires certain zoning ordinances, development agreements and permits to be consistent with the planning code and master plan.  (See fn. 1, ante.)   Residential Builders contends that the attempt to amend the master plan is invalid under the charter, that the remainder of part 1 is not severable and that the whole part is therefore invalid.   We agree that the master plan amendments are invalid but find that the remainder is severable.

 Respondents concede the partial invalidity, which stems from a limitation in the charter.   A city's charter is its constitution, and an ordinance can no more change or limit the effect of the charter than a statute can modify or supersede a provision of the state Constitution (Brown v. City of Berkeley (1976) 57 Cal.App.3d 223, 231, 129 Cal.Rptr. 1).   Section 9.108(a) limits ballot proposals to ordinances “which the supervisors are empowered to pass․”  (See also City and County of San Francisco v. Patterson (1988) 202 Cal.App.3d 95, 101, 248 Cal.Rptr. 290.)   Charter section 2.101 vests all powers of the city in the board “except th[ose] powers reserved to the people or delegated to other officials, boards or commissions by this charter․”  (Emphasis added.)   The charter makes the adoption and any amendment of the master plan solely “the function and duty of the city planning commission․”  (Charter, § 3.524;  see also id., § 3.525.)   It follows that a ballot measure proposed under section 9.108(a) may not amend, or force the city planning commission to amend, the master plan.   Part 1 of Proposition M is invalid to the extent that it tries to do this.

 However, under settled doctrine, the invalid parts are severable.   (Santa Barbara Sch. Dist. v. Superior Court (1975) 13 Cal.3d 315, 332, fn. 7, 118 Cal.Rptr. 637, 530 P.2d 605.)   We start by noting that Proposition M contains a severability clause.⁷  While not conclusive, such a clause normally calls for sustaining valid parts of an enactment, especially when invalid ones are mechanically severable.  (Id., at p. 331, 118 Cal.Rptr. 637, 530 P.2d 605.)   Mechanical severance does not have to be of an entire part;  it may be by paragraph, sentence, clause, phrase or even single words.   (In re Blaney (1947) 30 Cal.2d 643, 655, 184 P.2d 892.)   Finally, severance depends on whether the remainder of the enactment is complete in itself and would have been enacted had the voters foreseen the partial invalidation—in other words, whether the invalid parts are so connected with the rest that they are inseparable.  (Santa Barbara Sch. Dist. v. Superior Court, supra, 13 Cal.3d at p. 331, 118 Cal.Rptr. 637, 530 P.2d 605.)

Proposition M would amend the master plan indirectly by adding a section to the planning code which in turn requires, among other things, that the master plan be amended.   No one disputes that the board may amend the planning code.   The central question, then, is whether city voters, had they known that they could not amend the master plan to incorporate the eight “Priority Policies” (policies) established by the new planning code section (hereafter § 101.1), nevertheless would have wanted to make them part of the planning code.

We believe that they would.   Section 101.1 requires inclusion of the policies in the master plan by January 1, 1988, as well as consistency with those policies in later permit approvals and other development actions.  (§§ 101.1(a), 101.1(d);  see fn. 1, ante.)   At the same time, however, section 101.1 independently establishes those policies in the planning code and requires conformance with them immediately after election day (Nov. 4, 1986), long before the contemplated amendment of the master plan.   Section 101.1(c) forbids the adoption of certain zoning ordinances or development agreements “authorized ․ after November 4, 1986, unless prior to that adoption [the city] has specifically found that the ordinance or development agreement is consistent with the Priority Policies established above.”  (Emphasis added.)   Section 101.1(e) similarly preconditions the issuance of certain permits and legislation upon a finding of consistency “with the Priority Policies established above.”  (Emphasis added.)   That language, by distinguishing between conformance with the new planning code section and with the amended master plan, shows an intent to give the planning code amendment independent force.

Arguing that voters were not aware of the planning code amendment, Residential Builders notes that the title of part 1 is “MASTER PLAN,” that section 101.1 is captioned “MASTER PLAN CONSISTENCY AND IMPLEMENTATION” and that ballot materials available to voters did not differentiate between amending the planning code and the master plan.   However, any voter who read even the first few paragraphs of the proposition itself would have known that the planning code was being amended.   We are unwilling to assume that city voters read only ballot summaries and arguments, ignoring the measure itself.  “[T]he provisions to be severed must be so presented to the electorate in the initiative that their significance may be seen and independently evaluated in the light of the assigned purposes of the enactment.   The test is whether it can be said with confidence that the electorate's attention was sufficiently focused upon the parts to be severed so that it would have separately considered and adopted them in the absence of the invalid portions.”   (People's Advocate, Inc. v. Superior Court (1986) 181 Cal.App.3d 316, 332–333, 226 Cal.Rptr. 640, emphasis added.)   The code amendment was clearly apparent in the initiative, and we are confident that this was enough to focus voter attention on it.

The offending provisions are mechanically severable.   Section 101.1(a) states:  “The Master Plan shall be an integrated, internally consistent and compatible statement of policies for San Francisco.   To fulfill this requirement, after extensive public participation and hearings, the City Planning Commission shall in one action amend the Master Plan by January 1, 1988.”  The emphasized language must be excised.   The first sentence may remain.   As SFCAP notes, it only parrots a general statement of legislative intent (Gov.Code, § 65300.5) which already binds the city, and it does not depend on the new policies.   Next, section 101.1(b) begins:  “The following Priority Policies are hereby established.   They shall be included in the preamble to the Master Plan and shall be the basis upon which inconsistencies in the Master Plan are resolved:  ․” The emphasized language is easily excised, leaving the intent to establish the policies and the eight policies themselves, which follow.   Section 101.1(c), which requires zoning ordinances and development agreements to be consistent with the policies as thus set out in the planning code, is completely valid.   Jumping ahead, the first sentence of section 101.1(e), which requires the same consistency with regard to certain permits or legislation, is valid as well.

The difficult question is what to do with section 101.1(d) and the second sentence of section 101.1(e), which require, for the same development actions, consistency with the master plan after January 1, 1988.   The fact that consistency is required only after the proposed amendment of the master plan strongly suggests that voters would not have wanted consistency with the plan if they knew that the plan could not be amended to include the eight new policies.   Further support for this idea comes from section 101.1(a), which calls for the plan to be internally consistent, and from section 101.1(b), which speaks of resolving “inconsistencies in the Master Plan” by applying the eight new policies.   The overall picture is that the current master plan was not internally consistent, that the new policies were needed to resolve those inconsistencies, that the planning commission would have 14 months (until January 1988) in which to incorporate the new policies and integrate the plan, and that development actions by the city would have to conform to the plan only after it was amended.   Voters might well have feared that, without the ability to amend the plan, requiring consistency with it was potentially against their interests and in conflict with the new planning code amendments, which also bound the city.   For these reasons, we hold that sections 101.1(d) and the second sentence of section 101.1(e) are too connected with the invalid provisions to survive severance.⁸

Invalidation has remained an issue in this case only because of its potential impact on those who voted in favor of Proposition M.   The master plan amendments were never enforced, and the city informs us that the planning commission has since amended the master plan, independent of the proposition, to include the same policies.   In light of this, we see no need to remand the case for further proceedings.

DISPOSITION

We hereby modify the judgment to reflect that the master-plan provisions are unenforceable to the limited extent noted in part VI of this opinion.   As so modified, the judgment is affirmed.   The parties shall bear their own costs on appeal.  (Cal.Rules of Court, rule 26.) ⁹

FOOTNOTES

1.   “PART 1—MASTER PLAN“Be it ordained by the people of the City and County of San Francisco that Part II, Chapter II, of the San Francisco Municipal Code (City Planning Code) is hereby amended by adding section 101.1 as follows:“SECTION 101.1.   MASTER PLAN CONSISTENCY AND IMPLEMENTATION.“(a) The Master Plan shall be an integrated, internally consistent and compatible statement of policies for San Francisco.   To fulfill this requirement, after extensive public participation and hearings, the City Planning Commission shall in one action amend the Master Plan by January 1, 1988.“(b) The following Priority Policies are hereby established.   They shall be included in the preamble to the Master Plan and shall be the basis upon which inconsistencies in the Master Plan are resolved:“1. That existing neighborhood-serving retail uses be preserved and enhanced and future opportunities for resident employment in and ownership of such businesses enhanced;“2. That existing housing and neighborhood character be conserved and protected in order to preserve the cultural and economic diversity of our neighborhoods;“3. That the City's supply of affordable housing be preserved and enhanced;“4. That commuter traffic not impede Muni transit service or overburden our streets or neighborhood parking;“5. That a diverse economic base be maintained by protecting our industrial and service sectors from displacement due to commercial office development, and that future opportunities for resident employment and ownership in these sectors be enhanced;“6. That the City achieve the greatest possible preparedness to protect against injury and loss of life in an earthquake;“7. That landmarks and historic buildings be preserved;  and,“8. That our parks and open space and their access to sunlight and vistas be protected from development.“(c) The City may not adopt any zoning ordinance or development agreement authorized pursuant to Government Code Section 65865 after November 4, 1986, unless prior to that adoption it has specifically found that the ordinance or development agreement is consistent with the Priority Policies established above.“(d) The City may not adopt any zoning ordinance or development agreement authorized pursuant to Government Code Section 65865 after January 1, 1988, unless prior to that adoption it has specifically found that the ordinance or development agreement is consistent with the City's Master Plan.“(e) Prior to issuing a permit for any project or adopting any legislation which requires an initial study under [CEQA], and prior to issuing a permit for any demolition, conversion or change of use, and prior to taking any action which requires a finding of consistency with the Master Plan, the City shall find that the proposed project or legislation is consistent with the Priority Policies established above.   For any such permit issued or legislation adopted after January 1, 1988 the City shall also find that the project is consistent with the City's Master Plan.”

2.   Residential Builders relies on charter section 3.500(i), a provision defining “quorum for the transaction of official business” as “a majority of all the members of each board or commission,” although “a smaller number may adjourn from time to time and compel the attendance of absent members․”   The section adds (emphasis ours), “A majority, two-thirds, three-fourths, or other vote specified by this charter for any board or commission shall mean a majority, two-thirds, three-fourths, or other vote of all the members of such board or commission.”   The “other vote specified by this charter,” however, appears to refer to some other super-majority vote rather than a minority vote.  (See, e.g., charter § 2.302, referring to any resolution acted upon “by unanimous consent” of the supervisors present on the date of its introduction.)   Moreover, the section purports to govern only the “powers and duties” of “[e]ach board and commission appointed by the mayor, or otherwise provided by this charter” (charter, § 3.500), and thus has doubtful application to the board of supervisors.

3.   As further support for this construction, the city points to an administrative construction of section 9.108(a) by its registrar under which four or more supervisors need only express their intentions in writing, collectively or individually and despite lack of a full board vote, in order to qualify measures for the ballot.   When a board majority submits a measure, by contrast, the registrar requires certification from the board's clerk that a majority voted in favor of the submission.We do not rely on that practice and find it only marginally relevant.   The provision at issue here was apparently not used by supervisors until the late 1960's, about 55 years after its enactment.   The registrar's construction is thus not a contemporaneous one that might shed light on legislative intent in 1912.  (Cf. Whitcomb Hotel, Inc. v. Cal. Emp. Com. (1944) 24 Cal.2d 753, 756–757, 151 P.2d 233.)   While the construction is now 20 years old and thus suggests longstanding reliance, that fact would be germane only if the validity of the practice itself were at issue (id., at p. 757, 151 P.2d 233;  cf. Brubaker v. Morton (9th Cir.1974) 500 F.2d 200, 202), and it is not.For similar reasons, although the city relies on a 1986 opinion from its city attorney responding to a question from the registrar about the charter device (see DeYoung v. City of San Diego (1983) 147 Cal.App.3d 11, 18, 194 Cal.Rptr. 722 [city attorney formal opinion on charter matter is entitled to weight] ), it is of little help in discerning 1912 legislative intent.

4.   Franklin Hichborn, a contemporary observer, wrote in 1914:  “[T]he graft prosecution did something infinitely more important than the sending of a few corruptionists to cell and stripes.   It awakened a State to its helplessness against a corrupt system.   The People arose in rebellion against the ‘System,’ and is laboring to throw the ‘System’ off.  [¶]  In 1910 and 1911 a political revolution was worked in California.  [¶'s]  Had there been no San Francisco graft prosecution, there would, in 1910, have been no successful political uprising in California.   Hiram W. Johnson would not have been a candidate for Governor.   The accomplished reforms which are the boast of the State, and the models which other States are adopting, would still be the unrealized dreams of ‘reformers.’   The ‘System’ would still be in the saddle.”   (Hichborn, “The System” (1969 ed.) p. 464.)

5.   The parties have furnished us copies of the 1919 version.   The current version provides that the initiative and referendum powers are governed by general law.  (Oakland Charter, § 1004.)

6.   The Supreme Court recently declined to resolve lower court dispute over whether the Guidelines are regulatory mandates or only interpretive aids but did note, “At a minimum, ․ courts should afford great weight to [them] except when a provision is clearly unauthorized or erroneous under CEQA.  [Citation.]”  (Laurel Heights Improvement Assn. v. Regents of University of California (1988) 47 Cal.3d 376, 391, fn. 2, mod. 47 Cal.3d 982a, 253 Cal.Rptr. 426, 764 P.2d 278.)   Residential Builders has not made any such showing of invalidity in this case.

7.   Part 4 of the proposition provides:  “If any part of this initiative is held invalid by a court of law, or the application thereof to any person or circumstance is held invalid, such invalidity shall not affect the other parts of the initiative or applications which can be given effect without the invalid part or application hereof and to this end the sections of this initiative are separable.”

8.   SFCAP suggests that voters would have wanted to require consistency with the unamended plan because the city had until then availed itself of statutes allowing a charter city to escape the general law mandate of compliance with master plans.  (Gov.Code, §§ 65803, 65860.)   We disagree.   First, the record does not show that voters knew about this consistency exemption.   Second, even if they had known, they would have suspected from the language of the proposition that the existing plan was potentially inconsistent with the eight new policies being placed in the planning code and that to simultaneously compel consistency with both sources of law was thus unwise.

9.   At the close of its brief, SFCAP asks for “an award of attorneys fees.”   We deem the request abandoned for failure provide any supporting authority or argument.  (Berger v. Godden (1985) 163 Cal.App.3d 1113, 1119, 210 Cal.Rptr. 109.)   Assuming that these fees are sought under the private-attorney-general statute (Code Civ.Proc., § 1021.5), SFCAP may request them in the superior court after this appeal.  (Maria P. v. Riles (1987) 43 Cal.3d 1281, 1289, 240 Cal.Rptr. 872, 743 P.2d 932;  Citizens Against Rent Control v. City of Berkeley (1986) 181 Cal.App.3d 213, 226–227, 226 Cal.Rptr. 265.)   We, of course, express no opinion on the propriety or amount of such an award.

SMITH, Acting Presiding Justice.

BENSON and PETERSON, JJ., concur.