I respectfully dissent. I believe our Supreme Court’s exegesis of the application of Proposition 13 to long-term city obligations in Carman v. Alvord (1982) 31 Cal.3d 318 [182 Cal.Rptr. 506, 644 P.2d 192] should inform the analysis of Huntington Beach’s retirement benefits in this case. Since I am unconvinced by my colleagues’ attempt to distinguish Carman, I have no choice but to dissent.
In Carman v. Alvord, supra, 31 Cal.3d 318 (Carmen), our Supreme Court held that a city’s future obligations under its pension plan constituted the type of “indebtedness approved by the voters” which qualifies as an exception to the property tax limitation of Proposition 13. In Carman, the voters of the City of San Gabriel had approved the city’s participation in PERS in 1948, and empowered the city to “ ‘levy and collect annually, as contemplated in [the statewide statute], a special tax sufficient to raise the amount estimated by [the City] council to be required to meet the obligations of said City to said retirement system.’ ” (Id. at p. 322.) In the wake of California’s passage of Proposition 13, a taxpayer brought a class action alleging that the city’s continued levy of an excess tax to fund the retirement system was in violation of article XIIIA of the California Constitution.
The court explained that “[c]ourts construe constitutional phrases liberally and practically; where possible they avoid a literalism that effects absurd, arbitrary, or unintended results.” (Carman, supra, 31 Cal.3d at p. 327.) It then based its conclusion the pension plan was unobjectionable on two bases: “Subdivision (b)’s focus on voter approval implies a concern that irrevocable, long-term obligations, solemnly approved by local electorates and entered on faith in taxing powers then available, not be frustrated by a revolutionary tax limitation imposed from outside the community. [Citations.] It also implies a recognition that failure to create a ‘prior debt’ exception might lead to problems under the federal contract clause.” (Id. at p. 328.)
However, the court made clear that the impairment of contract issue was not its paramount concern, and specifically held that the special tax levy could be applied even to benefits voluntarily offered to employees hired after the effective date of Proposition 13. “It might be argued that contract clause problems do not arise as to employees hired after the effective date of article XIII A, since they perhaps did not enter service in reliance on City’s power to levy the special tax. Yet we see no reason to carve an exception for such persons. We may not assume that subdivision (b) sought to force local governments to the complex calculations necessary to separate their obligations to pre- and post-1978 employees. Article XIII A exempts ‘interest and redemption charges on any indebtedness previously approved by the electors.’ San Gabriel’s voters in 1948 obviously understood that subsequently hired employees too would be covered.” (Carman, supra, 31 Cal.3d at p. 333, fn. 11.)
*1390The court also expressly rejected the suggestion that Proposition 13’s exception applied only to indebtedness which was “fixed and certain” at the time of voter approval, noting that “the subdivision imposes no such restriction. It speaks only of the time of approval, not the time an indebtedness is incurred or accrues.” (Carman, supra, 31 Cal.3d at p. 326, fn. 6.)
In this case, the issue is not whether the excess tax may be levied to cover benefits offered to subsequently hired employees (an issue which respondents understandably concede in light of Carman), but whether it may be levied to cover the subsequently added benefits. Under the circumstances of this case, I conclude the Supreme Court’s reasoning in Carman establishes that it may.
First, I cannot accept respondents’ suggestion that the inclusion of the retirement provisions within the noncontroversial Proposition D portion of the proposed 1978 charter concludes the analysis. I am not convinced by respondents’ insistence that inclusion somehow indicates that neither the city council nor the voters could have intended to give the council authority to impose taxes for additional retirement benefits negotiated in the future. As the city points out, there is no substantial evidence to support that conclusion. In fact, the city’s authority to levy a special tax to fund its retirement system was already a part of its 1968 charter, and although the city was restricted by that charter to participation in PERS (absent approval of the voters to terminate), the city had some ability to make alterations to its benefits even within PERS system—and the record here demonstrates it had done so.
Moreover, while Huntington Beach’s voters approved two measures in 1976 which restricted the power of the city council to take actions which had the effect of raising taxes—requiring approval of a supermajority of the council to do so—they expressly excluded the retirement tax from those restrictions. Under these circumstances, I cannot conclude the voters would have considered the possibility of increased taxes to fund future retirement benefits to be particularly controversial.
Indeed, the most substantial retirement-related change made in the 1978 charter was the provision lifting the restriction to participation in PERS, and expressly allowing the city council discretion to set “reasonable” and “appropriate” benefits for the city’s employees. But again, I see no basis to conclude that would have been controversial. The new charter does not require any change in the retirement system, and I see no reason why the voters would not appreciate the benefits of allowing the city to “shop around.”
In any event, I agree with the city that even assuming the ballot proposition was mislabeled as noncontroversial, that does not mean its provisions should not be enforced according to their language. The city attorney’s impartial *1391analysis specifically exhorted the voters to give the charter amendments a “close reading,” and I assume the voters did so prior to casting their votes. Indeed, the fact that 40 percent of the voters opposed Proposition D suggests that they did form their own conclusions about its provisions, rather than blindly accepting the “noncontroversial” label.
The provisions of the 1978 charter expressly granted the city council authority to set “reasonable compensation and fringe benefits as are appropriate” for city officials and employees. Obviously, such authority forecasts future changes in such compensation and benefits, and makes no exception for pension or retirement benefits. The charter also lifts the prior restriction limiting the city’s participation to the PERS system, and allows the city to participate in another system. Finally, the charter provides that excess taxes may be imposed “sufficient to meet all obligations of the City for the retirement system in which the City participates, due and unpaid or to become due during the ensuing fiscal year.” It does not say, “all existing obligations ...” or even “all obligations in the City’s existing retirement plan.”
These provisions, taken together, indicate the voters in Huntington Beach intended to grant their city council the discretion to select the best retirement program for the city’s employees, to offer benefits that are “reasonable” and “appropriate,”1 and in the absence of other provision therefor, to fund those benefits (as the city apparently always had) through an excess property tax.
Respondents do not really quarrel with the first two conclusions. They agree that the charter allows the city to participate in other retirement systems of its choosing, and that the electorate gave the council authority to increase the level of benefits offered above those provided in 1978; however, they assert that Proposition 13 prohibits the levy of an excess property tax for such benefits. In other words, respondents suggest the electorate anticipated and approved future changes in the city’s retirement system, but could not have approved any tax funding for benefits not already specified. Presumably then, respondents assume the electorate envisioned a bake sale of monumental proportions.
But as the Supreme Court explained in Carman, Proposition 13’s exemption for excess taxation approved by the voters may be applied to future obligations under a city’s retirement plan, even when those obligations are not yet entered into or known at the time of approval, as long as the future *1392obligations would have been anticipated by the voters. Thus, in Carman, the court concluded that because the voters of San Gabriel understood in 1948 that future city employees would be added to the city’s pension system, those unknown (and uncounted) employees’ benefits would be considered as having been approved by the 1948 voters.
Respondents argue that this case does not fall under the reasoning of Carman, because while the attrition of old employees and addition of new employees is to be naturally expected and is a “statistical fact of life,” the addition of new retirement benefits is distinctly voluntary. I cannot agree, because the issue in Carman was not the mere hiring of new employees, but the extension of pension benefits to them. And while it may be true that a city must naturally hire new employees as older ones leave their employment or retire, it doesn’t follow that the city must offer those new employees any retirement benefits. It is not obligated to do so, and certainly many people work without such benefits. Thus, the Supreme Court in Carman was recognizing not an imperative of governmental biology, but a likelihood, understood by the city’s voters, that the city would choose to extend retirement benefits to new employees in the future.
Unfortunately, the majority, like respondents, ignores this aspect of Car-man and chooses instead to build its analysis on the purported distinction between “obligations,” which can be funded by the tax override provision, and “benefits” voluntarily offered to city employees in the wake of Proposition 13. In my view, this analysis ignores the city’s charter provision expressly giving the council continuing discretion to establish reasonable and appropriate fringe benefits. When the council exercises that discretion, the benefits established are part of the city’s retirement system obligation.
In this case, Huntington Beach’s retirement benefits changed several times in the years immediately proceeding the adoption of its 1978 charter. Against that backdrop, I cannot but conclude that the voters’ approval of the new charter, expanding the council’s authority to make such changes, reflected an understanding that those benefits would continue to change into the future. Thus, the concurrent approval of an excess tax, “sufficient to meet all obligations of the City for the retirement system in which the City participates” (italics added), must be construed as a prior approval of payment for those “reasonable” and “appropriate” benefits which the council was authorized to negotiate in the future.
Moreover, the Supreme Court’s rejection of the suggestion that Proposition 13 could be construed as “forc[ing] local governments to the complex calculations necessary to separate their obligations to pre- and post-1978 employees” applies equally to the issue of pre- and post-1978 benefits. (Carman, supra, 31 Cal.3d at p. 333, fn. 11.) In *1393fact, the trial court’s statement of decision highlights that very problem. As appellant points out, the court expressly found that the voters did intend the 1978 charter to authorize the city’s participation in a retirement system other than PERS. If the city chose to do so, however, how could it possibly keep track of which portions of an entirely new retirement system would be analogous to the pension benefits offered under PERS in 1978? Would it just be a question of taxing the same amount as allowed by the retirement tax for the PERS pension in 1978? Respondents say “no.” They expressly reject the notion that the city is allowed to tax for any pension contribution not expressly owed in the current year for the level of benefits offered by PERS in 1978. Thus, under respondents’ view, if the city changed retirement systems entirely, the amount for which it could levy an excess tax in future years would be dependent upon enlisting the cooperation of PERS to determine what PERS would be charging for the 1978 benefits under a retirement system which no longer exists. That would seem to present a problem, if not an impossibility. I do not think that is what the voters of Huntington Beach intended.
Finally, because the voter-approved authority at issue here is confined to setting “reasonable” and “appropriate” compensation and fringe benefits for employees, it does not present the scenario of an “open-ended voter approval ... to incur any government expense deemed desirable from year to year and to tax accordingly” which was of concern to the Supreme Court in Carman. (Carman, supra, 31 Cal.3d at p. 326, fn. 6.) Significantly, respondents do not contend that any of the retirement benefits added by Huntington Beach after 1978 are unreasonable or unexpected.
Respondents contend, however, that appellate court cases after Carman establish that Proposition 13’s exemption for “voter approved indebtedness” can never be applied to authorize a tax levy on any pension system not actually in existence and specifically approved by a city’s electorate prior to July 1, 1978, and that a charter provision such as the one in this case, giving the city council “open-ended” authority to elect new retirement systems in the future is insufficient as a matter of law to constitute voter approval of the indebtedness. In support of that proposition, respondents rely upon Valentine v. City of Oakland (1983) 148 Cal.App.3d 139 [196 Cal.Rptr. 59]. However, Valentine says no such thing. In that case, while the city charter in question did authorize the City of Oakland to “join ... or continue as a contracting agency in, any retirement or pension system or systems existing or hereafter created under state or federal law” (id. at p. 142, fn. 2), the city expressly disclaimed any reliance on the provision as establishing voter approval of its subsequent participation in PERS. Thus, the Valentine court had no occasion to consider the issue. And as respondents specifically *1394recognize in another context, “ ‘cases are not authority for propositions not considered.’ ” (American Federation of Labor v. Unemployment Ins. Appeals Bd. (1996) 13 Cal.4th 1017, 1039 [56 Cal.Rptr.2d 109, 920 P.2d 1314].)
What Valentine does establish, however, is that despite the contrary suggestion in Carman, Proposition 13’s exemption applies to authorize an excess tax levy for any indebtedness approved by the voters, even if the voters did not expressly approve the levy itself: “Once the indebtedness is found to have had the voters’ prior approval, ad valorem taxes etc. to pay the obligations arising thereunder are exempt, and there is no express requirement that such taxes need also be voter approved.” (Valentine v. City of Oakland, supra, 148 Cal.App.3d at p. 149.) In this case, of course, the taxation itself was expressly approved. And that approval, while not required, does emphasize the Huntington Beach voters’ express understanding of the implications flowing from their approval of the indebtedness.
The other case relied upon by respondents is City of Fresno v. Superior Court (1984) 156 Cal.App.3d 1137 [202 Cal.Rptr. 313]. City of Fresno is of little analytical use, however, because while the charter language granting the city council authority to make future changes in the city’s pension does appear to be fairly broad, the appellate court construed it as conferring only a very narrow power.
In City of Fresno, the voters approved a charter in 1957 which included section 1100, giving the city council authority to “ ‘establish a fund or funds for the relief and pensioning of all employees of the City of Fresno ...; provided, however, that retirement benefits established by any ordinance existing at the effective date of this Charter shall not be reduced, decreased or diminished.’ ” (City of Fresno v. Superior Court, supra, 156 Cal.App.3d at pp. 1140-1141.) Although the proviso specifically restricts only the lowering of benefits (and hence seems to inferentially approve their increase), the appellate court nonetheless construed it as “limiting the retirement benefits to the benefits established by ordinance as of the date of adoption of section 1100.” (Id. at p. 1145.) The court further stated that in section 1100 “the voters’ approval was limited to ‘retirement benefits’ and did not include all ‘retirement system costs’ referred to [in a later ordinance.]” (Id. at p. 1143.) But that was also incorrect. Section 1100 authorized creation of a fund or funds for the rather broad purpose of “relief and pensioning of all employees” (id. at p. 1140), and used the specific phrase “retirement benefits” only in the proviso setting the minimum benefits allowed. Thus, I cannot agree with the City of Fresno court’s interpretation of the charter provision at issue. Nonetheless, that interpretation, concluding that voters had approved only a very narrow authority to make changes in the city’s pension and retirement system was the foundation for the court’s *1395conclusion that retirement benefits and costs approved by ordinance subsequent to the voters’ approval of the charter did not qualify for the exemption to Proposition 13. That analysis has no application here.
The record here establishes that as early as its 1966 charter, more than a decade prior to passage of Proposition 13, Huntington Beach had already imposed a general 1 percent limitation on property taxes, subject to specific voter-approved exceptions, including an exception for a tax sufficient to fund the city’s retirement obligations. Thus, the passage of Proposition 13, which imposed the same tax limitation on a statewide basis, but also, like Huntington Beach, provided for voter approved exceptions, would presumably have been understood by the Huntington Beach voters as having little direct impact on the manner in which taxation was effected in their city.
Thus, the passage of Huntington Beach’s 1978 charter, which expressly continued the practice of allowing a special tax levy for retirement benefits in excess of the basic 1 percent tax limitation, while expressly giving the city council authority to set reasonable and appropriate employee benefits in the future, can only be construed as implying voter approval of (and hence taxation for) those reasonable and appropriate benefits offered by the city in the future. Unless I were able to conclude that Proposition 13 operates as a complete prohibition on excess taxation for any pension obligation not specifically quantifiable or incurred prior to July 1, 1978 (and in light of Carman’s allowance of such taxation for an unknown—and unknowable— number of future city employees to whom benefits were voluntarily extended, I cannot) I must conclude that Huntington Beach’s 1978 charter was intended to constitute voter approval of all future indebtedness for reasonable and appropriate retirement benefits offered by the city to its employees, regardless of whether the specific benefits offered were known to the voters in 1978.
The majority fails to acknowledge the significance of the provision restricting the council to offering only benefits that are “reasonable” and “appropriate.” In my view, that restriction is sufficient to protect the city’s taxpayers from the danger that the council might include Ferraris or houses as part of the city’s retirement package.