Opinion

I. INTRODUCTION

Southern California is a “semi-desert with a desert heart.” Visionary engineers and scientists have done a remarkable job of making our home habitable, and too many of us south of the Tehachapis never give a thought to its remarkable reclamation. In his brilliant — if opinionated — classic Cadillac Desert, the late Marc Reisner laments how little appreciation there is of “how difficult it will be just to hang on to the beachhead they have made.”

In this case we deal with parties who have an acute appreciation of how tenuous the beachhead is, and how desperately we all must fight to protect it. But they disagree about what steps are allowable — or required — to accomplish that task. We are called upon to determine not what is the right — or even the more reasonable — approach to the beachhead’s preservation, but what is the one chosen by the state’s voters.

We hope there are future scientists, engineers, and legislators with the wisdom to envision and enact water plans to keep our beloved Cadillac Desert habitable. But that is not the court’s mandate. Our job — and it is daunting enough — is solely to determine what water plans the voters and legislators of the past have put in place, and to determine whether the trial court’s rulings complied with those plans.

We conclude the trial court erred in holding that Proposition 218 does not allow public water agencies to pass on to their customers the capital costs of improvements to provide additional increments of water — such as building a recycling plant. Its findings were that future water provided by the improvement is not immediately available to customers. (See Cal. Const., art. XIII D, § 6, subd. (b)(4) [no fees “may be imposed for a service unless that service is actually used by, or immediately available to, the owner of the property in question”].) But, as applied to water delivery, the phrase “a service” cannot be read to differentiate between recycled water and traditional, potable water. Water service is already “immediately available” to all customers, and continued water service is assured by such capital improvements as water recycling plants. That satisfies the constitutional and statutory requirements.

However, the trial court did not err in ruling that Proposition 218 requires public water agencies to calculate the actual costs of providing water at various levels of usage. Article XIII D, section 6, subdivision (b)(3) of the California Constitution, as interpreted by our Supreme Court in Bighorn-Desert View Water Agency v. Verjil (2006) 39 Cal.4th 205, 221 [ 46 Cal.Rptr.3d 73 , 138 P.3d 220 ] (Bighorn) provides that water rates must reflect the “ ‘cost of the service attributable’ ” to a given parcel. While tiered, or inclined rates that go up progressively in relation to usage are perfectly consonant with article XIII D, section 6, subdivision (b)(3) and Bighorn, the tiers must still correspond to the actual cost of providing service at a given level of usage. The water agency here did not try to calculate the cost of actually providing water at its various tier levels. It merely allocated all its costs among the price tier levels, based not on costs, but on predetermined usage budgets. Accordingly, the trial court correctly determined the agency had failed to carry the burden imposed on it by another part of Proposition 218 (art. XIII D, § 6, subd. (b)(5)) of showing it had complied with the requirement water fees not exceed the cost of service attributable to a parcel at least without a vote of the electorate. That part of the judgment must be affirmed.

II. FACTS

Sometimes cities are themselves customers of a water district, the best example in the case law being the City of Palmdale, which successfully invoked Proposition 218 to challenge the rates it was paying to a water district. (See City of Palmdale v. Palmdale Water Dist. (2011) 198 Cal.App.4th 926 [ 131 Cal.Rptr.3d 373 ] (Palmdale).) And sometimes cities are, as in the present case, their own water district. As amicus curiae Association of California Water Agencies (ACWA) points out, government water suppliers in California are a diverse lot that includes municipal water districts, irrigation districts, county water districts, and, in some cases, cities themselves. To focus on its specific role in this case as a municipal water supplier — as distinct from its role as the provider of municipal services which consume water such as parks, city landscaping or public golf courses — we will refer to appellant City of San Juan Capistrano as “City Water.”

In February 2010, City Water adopted a new water rate structure recommended by a consulting firm. The way City Water calculated the new rate structure is well described in City Water’s supplemental brief of November 25, 2014. City Water followed a pattern generally recommended by a manual used by public water agencies throughout the western United States known as the “M-l” manual. It first ascertained its total costs, including things like debt service on previous infrastructural improvements. It then identified components of its costs, such as the cost of billing and the cost of water treatment. Next it identified classes of customers, differentiating, for example, between “regular lot” residential customers and “large lot” residential customers, and between construction customers and agricultural customers. Then, in regard to each class, City Water calculated four possible budgets of water usage, based on historical data for usage patterns: low, reasonable, excessive and very excessive.

The four budgets were then used as the basis for four distinct “tiers” of pricing. For residential customers, tier 1, the low budget, was assumed to be exclusively indoor usage, based on World Health Organization guidelines concerning the “minimum quantity of water required for survival,” with adjustments for things like “low-flush toilets and other high-efficiency appliances.” Tier 2, the reasonable budget, included an outdoor allocation based on “typical landscapes,” and assumed “use of native plants and drought-tolerant plants.” The final two tiers were based on budgets of what City Water considers excessive usages of water or overuse volumes. Using these four budgets of consumption levels, City Water allocated its total costs in such a way that the anticipated revenues from all four tiers would equal its total costs, and thus the four-tier system would be, taken as a whole, revenue neutral, and City Water would not make a profit on its pricing structure. City Water did not try to calculate the incremental cost of providing water at the level of use represented by each tier, and in fact, at oral argument in this court, admitted it effectively used revenues from the top tiers to subsidize below-cost rates for the bottom tier.

Here is the rate structure adopted, as applied to residential customers:

Tier Usage Price

1 Up to 6 ccf $2.47 per ccf

2 7 to 17 ccf $3.29 per ccf

Tier Usage Price

3 18 to 34 ccf $4.94 per ccf

4 Over 34 ccf $9.05 per ccf

City Water obtains water from five separate sources: a municipal groundwater recovery plant, the Metropolitan Water District, five local groundwater wells, recycled water wells, and the nearby Moulton Niguel Water District. With the exception of water obtained from the Metropolitan Water District, City Water admits in its briefing that the record does not contain any breakdown as to the relative cost of each source of supply.

The breakdown of cost from percentage terms: each of its various sources of water is, in

Source Percent of Supply Cost to Supply

Groundwater Recovery Plant 51.95% Not ascertained

Metropolitan Water District 28.54% $1,007 per-acre foot

Local Wells 7.79% Not ascertained

Recycled Wells 6.11% Not ascertained

Moulton Niguel Water District 5.61% Not ascertained

Various percentages of City Water’s overhead — or fixed costs in the record — were allocated in percentages to some of the sources of water, so the price per tier reflected a percentage of fixed costs and costs of some sources.

This chart reflects those allocations:

Tier Price Percentage Allocation

1 $2.47 $1.78 to fixed costs, $0.62 to wells

2 $3.29 $1.78 to fixed, $1.46 to wells

3 $4.94 $1.53 to fixed, $0.69 to wells, $0.17 to the Metropolitan Water District, and $2.50 to the groundwater recovery plant

4 $9.05 0 to fixed, 0 to wells, $0.53 to groundwater recovery plant, $2.53 to recycled, $3.32 to the Metropolitan Water District, and $2.64 to penalty set-aside

There is no issue in this case as to the process of the adoption of the new rates, such as whether they should have been voted on first under the article Xni C part of Proposition 218. For purposes of this appeal it is enough to say City Water adopted them.

In August 2012, the Capistrano Taxpayers Association, Inc. (CTA), filed this action, challenging City Water’s new rates as violative of Proposition 218, specifically article XIII D, section 6, subdivision (b)(3)’s limit on fees to the “cost of the service attributable to the parcel.” After a review of the administrative record and hearing, the trial court found the rates were not compliant with article XIII D, noting it “could not find any specific financial cost data in the [record] to support the substantial rate increases” in the progressively more expensive tiers. In particular the trial judge found a lack of support for the inequality between the tiers.

The statement of decision also concluded that the imposition of charges for recycling within the rate structure violated the “immediately available” provision in article XIII D, section 6, subdivision (b)(4), because recycled water is not used by residential parcels. (City Water concedes that when the recycling plant comes on line, it will supply water to some, but not all, of its customers. Residences, for example, are not typically plumbed to receive nonpotable recycled water.) City Water has timely appealed from the declaratory judgment, challenging both determinations.

III. DISCUSSION

A. Capital Costs and Proposition 218

We first review the constitutional text. Article XIIID, section 6, subdivision (b)(4) provides: “No fee or charge may be imposed for a service unless that service is actually used by, or immediately available to, the owner of the property in question. Fees or charges based on potential or future use of a service are not permitted. Standby charges, whether characterized as charges or assessments, shall be classified as assessments and shall not be imposed without compliance with Section 4.”

The trial court ruled City Water had violated this provision by “charging certain ratepayers for recycled water that they do not actually use and that is not immediately available to them.” The trial judge specifically found, in his statement of decision, that “City [Water] imposed a fee on all ratepayers for recycled water services and delivery of recycled water services, despite the fact that not all ratepayers used recycled water or have it immediately available to them or would ever be able to use it.”

But the trial court assumed that providing recycled water is a fundamentally different kind of service from providing traditional potable water. We think not. When each kind of water is provided by a single local agency that provides water to different kinds of users, some of whom can make use of recycled water (for example, cities irrigating park land) while others, such as private residences, can only make use of traditional potable water, providing each kind of water is providing the same service. Both are getting water that meets their needs. Nonpotable water for some customers frees up potable water for others. And since water service is already immediately available to all customers of City Water, there is no contravention of subdivision (b)(4) in including charges to construct and provide recycled water to some customers.

On this point, Griffith v. Pajaro Valley Water Management Agency (2013) 220 Cal.App.4th 586 [ 163 Cal.Rptr.3d 243 ] (Griffith) is instructive. Griffith involved an augmentation fee on parcels that had their own wells. An objection to the augmentation fee by the well owners was that the fee included a charge for delivered water, even though some of the properties were outside the area and not actually receiving delivered water. The Griffith court said that even if some parcel owners were not receiving delivered water, revenues from the augmentation fee still benefited those parcels, since they funded “activities required to prepare or implement the groundwater management program for the common benefit of all water users.” (Id. at p. 602.) In Griffith the augmentation fee was thus intended to fund aggressive capital investments to increase the general supply of water, including some customers receiving delivered water when other customers did not. It was undeniable that by funding delivered water to some customers, water was freed up for all customers. (See Griffith, supra, 220 Cal.App.4th at p. 602 ; accord, Paland v. Brooktrails Township Community Services Dist. Bd. of Directors (2009) 179 Cal.App.4th 1358 [ 102 Cal.Rptr.3d 270 ] [customer in rural area who periodically went inactive still had water immediately available to him].)

In the present case, there is a Government Code definition of water, which shows water to be part of a holistic distribution system that does not distinguish between potable and nonpotable water: “ ‘Water’ means any system of public improvements intended to provide for the production, storage, supply, treatment, or distribution of water from any source.” (Gov. Code, § 53750, subd. (m).)

A recycling plant, like other capital improvements to increase water supply, obviously entails a longer timeframe than a residential customer’s normal one-month billing cycle. As shown in Morgan v. Imperial Irrigation Dist. (2014) 223 Cal.App.4th 892 [ 167 Cal.Rptr.3d 687 ], the timeframe for the calculation of the true cost of water can be, given capital improvements, quite long. (See id. at p. 900 [costs amortized over a six-year period].) And, as pointed out by amicus curiae Howard Jarvis Taxpayers Association, Government Code section 53756 contemplates timeframes for water rates that can be as long as five years. There is no need, then, to conclude that rates to pay for a recycling plant have to be figured on a month-to-month basis.

The upshot is that within a five-year period, a water agency might develop a capital-intensive means of production of what is effectively new water, such as recycling or desalinization, and pass on the costs of developing that new water to those customers whose marginal or incremental extra usage requires such new water to be produced. As amicus curiae Mesa Water District points out, Water Code section 31020 gives local water agencies the power to do acts to “furnish sufficient water in the district for any present or future beneficial use.” (Wat. Code, § 31020, italics added.) The trial court thus erred in concluding the inclusion of charges to fund a recycling operation was, by itself, a violation of subdivision (b)(4).

However, the record is insufficient to allow us to determine at this level whether residential ratepayers who only use six ccf or less — what City Water considers the superconservers — are being required to pay for recycling facilities that would not be necessary but for above-average consumption. Proposition 218 protects lower-than-average users from having to pay rates that are higher than the cost of service for them because those rates cover capital investments their levels of consumption do not make necessary. We note, in this regard, that in Palmdale, one of the reasons the court there found the tiered pricing structure to violate subdivision (b)(3) was the perverse effect of affirmatively penalizing conservation by some users. (See Palmdale, supra, 198 Cal.App.4th at pp. 937-938; accord, Brydon, supra, 24 Cal.App.4th at p. 202 [“To the extent that certain customers overutilize the resource, they contribute disproportionately to the necessity for conservation, and the requirement that the District acquire new sources for the supply of domestic water.”].)

There is a case with an analogous lacuna, the Supreme Court case of California Farm Bureau Federation v. State Water Resources Control Bd. (2011) 51 Cal.4th 421 [ 121 Cal.Rptr.3d 37 , 247 P.3d 112 ] (Farm Bureau). In Farm Bureau, the record was also unclear as to the issue of apportionment between a regulatory activity’s fees and its costs. (Id. at p. 428.) Accordingly, the high court directed the matter to be remanded to the trial court for such necessary findings.

That seems to us the appropriate way to complete the record in our case. Following the example of Farm Bureau, we remand the matter for further findings on whether charges to develop City Water’s nascent recycling operation have been improperly allocated to users whose levels of consumption are so low that they cannot be said to be responsible for the need for that recycling.

B. Tiered Pricing and Cost of Service

1. Basic Analysis

We begin, as we did with the capital cost issue, with the text of the Constitution. In addition to subdivision (b)(3), the main provision at issue in this case, we also quote subdivision (b)(1), because it throws light on subdivision (b)(3). Subdivision (b) describes “Requirements for Existing, New or Increased Fees and Charges,” and provides that “A fee or charge shall not be extended, imposed, or increased by any agency unless it meets all of the following requirements: [¶] (1) Revenues derived from the fee or charge shall not exceed the funds required to provide the property related service. [¶] . . . [¶] (3) The amount of a fee or charge imposed upon any parcel or person as an incident of property ownership shall not exceed the proportional cost of the service attributable to the parcel.” (Italics added.)

In addition to these two substantive limits on fees, article XIII D, section 6, subdivision (b)(5) puts an important procedural limit on a court’s analysis in regard to the burden of proof: “In any legal action contesting the validity of a fee or charge, the burden shall be on the agency to demonstrate compliance with this article.” The trial court found City Water had failed to carry its burden of proof under subdivision (b)(5) of showing its 2010 tiered water fees were proportional to the cost of service attributable to each customer’s parcel as required by subdivision (b)(3).

As respondent CTA quickly ascertained, the difference between tier 1 and tier 2 is a tidy one-third extra, the difference between tier 2 and 3 is a similarly exact one-half extra, and the difference between tier 3 and tier 4 is precisely five-sixths extra. This fractional precision suggested to us that City Water did not attempt to correlate its rates with cost of service. Such mathematical tidiness is rare in multidecimal-point calculations. This conclusion was confirmed at oral argument in this court, when City Water acknowledged it had not tried to correlate the incremental cost of providing service at the various incremental tier levels to the prices of water at those levels.

In voluminous briefing by City Water and its amici curiae allies, two somewhat overlapping core thoughts emerge: First, they contend that when it comes to water, local agencies do not have to — or should not have to— calculate the cost of water service at various incremental levels of usage because the task is simply too complex and thus not required by our Constitution. The second core thought is that even if agencies are required to calculate the actual costs of water service at various tiered levels of usage, such a calculation is necessarily, as City Water’s briefing contends, a legislative or quasi-legislative, discretionary matter, largely insulated from judicial review. We cannot agree with either assertion.

The appropriate way of examining the text of Proposition 218 has already been spelled out by the Supreme Court in Silicon Valley Taxpayers’ Assn., Inc. v. Santa Clara County Open Space Authority (2008) 44 Cal.4th 431, 448 [ 79 Cal.Rptr.3d 312 , 187 P.3d 37 ] (Silicon Valley): “We ‘ “ ‘must enforce the provisions of our Constitution and “may not lightly disregard or blink at ... a clear constitutional mandate.” ’ ” ’ [Citation.] In so doing, we are obligated to construe constitutional amendments in a manner that effectuates the voters’ purpose in adopting the law. [Citation.] [¶] Proposition 218 specifically states that ‘[t]he provisions of this act shall be liberally construed to effectuate its purposes of limiting local government revenue and enhancing taxpayer consent.’ (Ballot Pamp., [Gen. Elec. (Nov. 5, 1996)] text of Prop. 218, § 5, p. 109; see Historical Notes, [2A West’s Ann. Const. (2008 supp.) foil. Cal. Const., art. XIII C.] at p. 85.) Also, as discussed above, the ballot materials explained to the voters that Proposition 218 was designed to: constrain local governments’ ability to impose assessments; place extensive requirements on local governments charging assessments; shift the burden of demonstrating assessments’ legality to local government; make it easier for taxpayers to win lawsuits; and limit the methods by which local governments exact revenue from taxpayers without their consent.” (Silicon Valley, supra, 44 Cal.4th at p. 448 , italics added.)

If the phrase “proportional cost of the service attributable to the parcel” (italics added) is to mean anything, it has to be that article XIII D, section 6, subdivision (b)(3) assumes that there really is an ascertainable cost of service that can be attributed to a specific — hence that little word “the” — parcel. Otherwise, the cost of the service language would be meaningless. Why use the phrase “cost of the service to the parcel” if a local agency doesn’t actually have to ascertain a cost of service to that particular parcel?

The presence of subdivision (b)(1) of section 6, article XIII D, just a few lines above subdivision (b)(3), confirms our conclusion. Constitutional provisions, particularly when enacted in the same measure, should be construed together and read as a whole. (Bighorn, supra, 39 Cal.4th at p. 228 .) The “proportional cost of the service” language from subdivision (b)(3) is part of a general subdivision (b), and there is an additional reference to costs in subdivision (b)(1). Subdivision (b)(1) provides that the total revenue from fees “shall not exceed the funds required to provide the property related service.” (Italics added.)

It seems to us that to comply with the Constitution, City Water had to do more than merely balance its total costs of service with its total revenues^ — that is already covered in subdivision (b)(1). To comply with subdivision (b)(3), City Water also had to correlate its tiered prices with the actual cost of providing water at those tiered levels. Since City Water didn’t try to calculate the actual costs of service for the various tiers, the trial court’s ruling on tiered pricing must be upheld simply on the basis of the constitutional text.

We find precedent for our conclusion in the Palmdale case. There, a water district obtained its water from two basic sources: 60 percent from a reservoir and the state water project, and the 40 percent balance from the district’s own area groundwater wells. Most (about 72 percent) of the water went to single-family residences, with irrigation users accounting for 5 percent of the distribution. (Palmdale, supra, 198 Cal.App.4th at p. 928 .) For the previous five years, the district had spent considerable money to upgrade its water treatment plant ($56 million) but revenues suffered from a “ ‘decline in water sales,’ ” so its reserves were depleted. The district wanted to issue more debt for “future capital projects.” (Id. at pp. 928-929.) Relying on consultants, the water district adopted a new, five-tiered rate structure, which progressively increased rates (for the top four tiers) for three basic categories of customers: residences, businesses, and irrigation projects. The tiered budgets for irrigation users were more stringent than for residential and commercial customers. (Id. at p. 930.) The way the tiers operated, all three classes of customers got a tier 1 budget, but irrigation customers had less leeway to increase usage without progressing to another tier. Thus, for example, the tier 2 rates for residential customers did not kick in until 125 percent of the budget, but tier 2 rates for irrigation customers kicked in at 110 percent of the budget. The tiered rate structure was itself based on a monthly allocated water budget. (Ibid.)

Two irrigation users — the city itself and its redevelopment agency — sought to invalidate the new rates. The trial court had the advantage of the newly decided Supreme Court opinion in Silicon Valley, which had clarified the standard of review for Proposition 218 cases. There, the high court made it clear that in Proposition 218 challenges to agency action, the agency had to bear the burden of proof of demonstrating compliance with Proposition 218, and both trial and reviewing courts are to apply an independent review standard, not the traditional, deferential standards usually applicable in challenges to governmental action. (Silicon Valley, supra, 44 Cal.4th at p. 448 .) More directly, said Silicon Valley, it is not enough that the agency have substantial evidence to support its action. That substantial evidence must itself be able to withstand independent review. (See id. at pp. 441, 448-449 [explaining why substantial evidence to support the agency action standard was too deferential in light of Prop. 218’s liberal construction in favor of taxpayer feature].)

With this in mind, the Palmdale court held the district had failed to carry its burden of showing compliance with Proposition 218. ( Palmdale, supra, 198 Cal.App.4th at pp. 937-938.) The core of the Palmdale court’s reasoning was twofold. First, there was discrimination against irrigation-only customers, giving an unfair price advantage to those customers in other classes who were inclined to inefficiently use — or, for that matter, waste — outdoor water. (The opinion noted the perfect exemplar of water waste; hosing off a parking lot.) Thus an irrigation user, such as a city providing playing fields, playgrounds and parks, was disproportionately impacted by the inequality in classes of users. (Palmdale, supra, 198 Cal.App.4th at p. 937 .) Second, the discrimination was gratuitous. The district’s own consultants had proposed a “cost of service” option that they considered Proposition 218 compliant, but the district did not choose it because it preferred a “fixed” option providing better “ ‘rate stability.’ ” ( Palmdale, supra, 198 Cal.App.4th atpp. 937, 929.) In fact the choice had the perverse effect of entailing a “ ‘weaker signal for water conservation’ ” for “ ‘small customers who conserve water.’ ” (Id. at pp. 929, 937, some italics added.)

We recognize that Palmdale was primarily focused on inequality between classes of users, as distinct from classes of water rate tiers. But, just as in Palmdale where the district never attempted to justify the inequality “in the cost of providing water” to its various classes of customers at each tiered level (Palmdale, supra, 198 Cal.App.4th at p. 937 ), so City Water has never attempted to justify its price points as based on costs of service for those tiers. Rather, City Water merely used what it thought was its legislative, discretionary power to attribute percentages of total costs to the various tiers. While an interesting conversation might be had about whether this was reasonable or wise, we can find no room for arguing its constitutionality. It does not comply with the mandate of the voters as we understand it.

2. City Water’s Arguments

a. Article X, Section 2

In supplemental briefing prior to oral argument, this court pitched a batting practice fastball question to City Water, intended to give the agency its best chance of showing that the prices for its various usage tiers, particularly the higher tiers (e.g., $4.94 for all usage over 17 ccf to 34 ccf, and $9.05 for usage over 34 ccf) corresponded with its actual costs of delivering water in those increments. We were hoping that, maybe, we had missed something in the record that would demonstrate the actual cost of delivering water for usage over 34 ccf per month really is $9.05 per ccf, and City Water would hit our question into the upper deck.

What we got back was a rejection of the very idea behind the question. As would later be confirmed at oral argument, City Water’s answer was that there does not have to be a correlation between tiered water prices and the cost of service. Its position is that the “cost-of-service principle of Proposition 218” must be “balance[dj” against “the conservation mandate of article X, section 2.” In short, City Water justifies the lack of a correlation between the marginal amounts of water usage represented by its various tiers and the actual cost of supplying that water by saying the lack of correlation is excused by the subsidy for low usage represented by tier 1, on the theory that subsidized tier 1 rates are somehow required by article X, section 2. While we agree that low-cost water rates do not, in and of themselves, offend subdivision (b)(3) (see Morgan v. Imperial Irrigation Dist., supra, 223 Cal.App.4th at p. 899 ), we cannot adopt City Water’s constitutional extrapolation of that point.

We quote the complete text of article X, section 2 in the footnote. Article X, section 2 was enacted in 1928 in reaction to a specific Supreme Court case decided two years earlier, Herminghaus v. South. California Edison Co. (1926) 200 Cal. 81 [ 252 P. 607 ] (Herminghaus). The Herminghaus decision, as Justice Shenk wrote in his dissent there, allowed downstream riparian landowners — basically farmers owning land adjacent to a river — to claim 99 percent of the flow of the San Joaquin River even though they were actually using less than 1 percent of that flow. To compound that anomaly, the downstream riparian landowners’ claims came at the expense of the efforts of an electric utility company to generate electricity for general, beneficial use by building reservoirs at various points upstream on the river. (See Herminghaus , at p. 109.) In the process of upholding the downstream landowners’ “riparian rights” over the rights of the electric company to use the water to make electricity, the Herminghaus majority invalidated legislation aimed at preserving water in the state for a reasonable beneficial use, thereby countenancing what Justice Shenk perceived to be a plain waste of good water. (Herminghaus, supra, 200 Cal. at p. 123 (dis. opn. of Shenk, J.).) As our Supreme Court would describe Herminghaus about half a century later: “we held not only that riparian rights took priority over appropriations authorized by the Water Board, a point which had always been clear, but that as between the riparian and the appropriator, the former’s use of water was not limited by the doctrine of reasonable use.” (National Audubon Society v. Superior Court (1983) 33 Cal.3d 419, 442 [ 189 Cal.Rptr. 346 , 658 P.2d 709 ].)

The voters overturned Herminghaus in the 1928 election by adopting article X, section 2, then denoted article XIV, section 3. (See Gin S. Chow v. City of Santa Barbara (1933) 217 Cal. 673, 699 [ 22 P.2d 5 ] (Gin Chow).) In the 1976 constitutional revision, old article XIV, section 3, was recodified verbatim as article X, section 2. (See Gray, “In Search of Bigfoot”: The Common Law Origins of Article X, Section 2 of the California Constitution (1989) 17 Hastings Const. L.Q. 225 (hereinafter Origins of Article X, Section 2).

The purpose of article X, section 2 was described in Gin Chow, the first case to reach the Supreme Court in the wake of the adoption of what is now article X, section 2, in 1928. Justice Shenlc, having been vindicated by the voters on the point of a perceived need to prevent the waste of water by letting it flow to the sea, summarized the new amendment in terms emphasizing beneficial use: “The purpose of the amendment was stated to be ‘to prevent the waste of waters of the state resulting from an interpretation of our law which permits them to flow unused, unrestrained and undiminished to the sea’, and is an effort ‘on the part of the state, in the interest of the people of the state, to conserve our waters’ without interference with the beneficial uses to which such waters may be put by the owners of water rights, including riparian owners. That such purpose is reflected in the language of the amendment is beyond question. Its language is plain and unambiguous. In the main it is an endeavor on the part of the people of the state, through its fundamental law, to conserve a great natural resource, and thereby render available for beneficial use that portion of the waters of our rivers and streams which, under the old riparian doctrine, was of no substantial benefit to the riparian owner and the conservation of which will result in no material injury to his riparian right, and without which conservation such waters would be wasted and forever lost.” (Gin Chow, supra, 217 Cal. at p. 700 .)

The emphasis in the actual language of article X, section 2 is thus on a policy that favors the beneficial use of water as against the waste of water for nonbeneficial uses. That is what one would expect, consistent with both Justice Shenk’s dissent in Herminghaus and his majority opinion in Gin Chow. (See Origins of Article X, Section 2, supra, 17 Hastings Const. L.Q. at p. 263 [noting emphasis in text on beneficial use].) The word “conservation” is used in the introductory sentence of the provision in the context of promoting beneficial uses: “ ‘the conservation of such waters is to be exercised with a view to the reasonable and beneficial use thereof in the interest of the people and for the public welfare.’ ” (Origins of Article X, Section 2, at p. 225, italics added.)

But nothing in article X, section 2, requires water rates to exceed the true cost of supplying that water, and in fact pricing water at its true cost is compatible with the article’s theme of conservation with a view toward reasonable and beneficial use. (See Palmdale, supra, 198 Cal.App.4th at pp. 936-937 [reconciling art. X, § 2 with Prop. 218]; accord, Brydon, supra, 24 Cal.App.4th at p. 197 [noting that incremental rate structures create an incentive to reduce water use].) Thus it is hard for us to see how article X, section 2, can be read to trump subdivision (b)(3). We would note here that in times of drought — which looks increasingly like the foreseeable future — • providing water can become very pricey indeed. And, we emphasize, there is nothing at all in subdivision (b)(3) or elsewhere in Proposition 218 that prevents water agencies from passing on the incrementally higher costs of expensive water to incrementally higher users. That would seem like a good idea. But subdivision (b)(3) does require they figure out the true cost of water, not simply draw lines based on water budgets. Thus in Palmdale, the appellate court perceived no conflict between Proposition 218 and article X, section 2, so long as article X, section 2 is not read to allow water rates that exceed the cost of service. Said Palmdale: “California Constitution, article X, section 2 is not at odds with article XIII D so long as, for example, conservation is attained in a manner that ‘shall not exceed the proportional cost of the service attributable to the parcel.’ (Art. XIII D, § 6, subd. (b), par. (3).)” ( Palmdale, supra, 198 Cal.App.4th at pp. 936-937, italics added.) And as its history, and the demonstrated concern of the voters in 1928 demonstrates, article X, section 2 certainly does not require above-cost water rates.

In fact, if push came to shove and article X, section 2 really were in irreconcilable conflict with article XIII D, section 6, subdivision (b)(3), we might have to read article XIII D, section 6, subdivision (b)(3) to have carved out an exception to article X, section 2, since Proposition 218 is both more recent and more specific. (Greene v. Marin County Flood Control & Water Conservation Dist. (2010) 49 Cal.4th 277, 290 [ 109 Cal.Rptr.3d 620 , 231 P.3d 350 ] [“ ‘As a means of avoiding conflict, a recent, specific provision is deemed to carve out an exception to and thereby limit an older, general provision.’ ”]; Izazaga v. Superior Court (1991) 54 Cal.3d 356, 371 [ 285 Cal.Rptr. 231 , 815 P.2d 304 ] [same].)

Fortunately, that problem has not arisen. We perceive article X, section 2 and article XIIID, section 6, subdivision (b)(3) to work together to promote increased supplies of water — after all, the main reason article X, section 2 was enacted in the first place was to ensure the capture and beneficial use of water and prevent its wasteful draining into the ocean. As a pre-Proposition 218 case, Brydon, supra, 24 Cal.App.4th 178 observed, one of the benefits of tiered rates is that it is reasonable to assume people will not waste water as its price goes up. (See id. at p. 197 [noting that incremental rate structures create an incentive to reduce water use].) Our courts have made it clear they interpret the Constitution to allow tiered pricing; but the voters have made it clear they want it done in a particular way.

b. Brydon and Griffith

We believe the precedent most on point is Palmdale, and we read Palmdale to support the trial court’s conclusion City Water did not comply with the subdivision (b)(3) requirement that rates be proportional to cost of service. The two cases City Water relies on primarily for its opposite conclusion, Brydon and Griffith, do not support a different result.

Brydon was a pre-Proposition 218 case upholding a tiered water rate structure as against challenges based on 1978’s Proposition 13 rational basis and equal protection challenges. Similar to the case at hand, the water district promulgated an “inclining block rate structure.” (Brydon, supra, 24 Cal.App.4th at p. 182 ; see p. 184 [details of four-tier structure].) Proposition 218 had not yet been enacted, so the opponents of the block rate structure did not have the “proportional cost of the service attributable to the parcel” language in subdivision (b)(3) to use to challenge the rate structure. They relied, rather, on the theory that Proposition 13 made the rate structure a “ ‘special tax,’ ” requiring a vote. (Brydon, at p. 182.) As backup they made traditional rational basis and equal protection arguments. They claimed the rate structure was “arbitrary, capricious and not rationally related to any legitimate legislative or administrative objective” and, further, that the structure unreasonably discriminated against customers in the hotter areas of the district. (Brydon, supra, at p. 182 .) The Brydon court rejected both the Proposition 13 and rational basis/equal protection arguments.

But Brydon — though it might still be read as evidence that tiered pricing not otherwise connected to cost of service would survive a rational basis or equal protection challenge — simply has no application to post-Proposition 218 cases. In fact, the construction of Proposition 13 applied by Brydon was based on cases Proposition 218 was designed to overturn. The best example of such reliance was Brydon" s declining to follow Beaumont Investors v. Beaumont-Cherry Valley Water Dist. (1985) 165 Cal.App.3d 227 [ 211 Cal.Rptr. 567 ] (Beaumont) on the issue of the burden of proof. Beaumont had held it was the agency that had the burden of proof to show compliance with Proposition 13. Brydon , however, said the burden was on the taxpayers to show lack of compliance. In coming to its conclusion, Brydon invoked Knox v. City of Orland (1992) 4 Cal.4th 132 [ 14 Cal.Rptr.2d 159 , 841 P.2d 144 ], Knox , said Brydon , had “cast substantial doubt” on the “propriety of shifting the burden of proof to the agency.” ( Brydon, supra, 24 Cal.App.4th at p. 191.) But, more than a decade later, our Supreme Court in Silicon Valley recognized that Knox itself was one of the targets of Proposition 218. (See Silicon Valley, supra, 44 Cal.4th at p. 445 .) In the wake of Knox’s, fate (see in particular subd. (b)(5) [changing burden of proof]), it seems safe to say that Brydon itself was part of the general case law which the enactors of Proposition 218 wanted replaced with stricter controls on local government discretion.

As the Silicon Valley court observed, Proposition 218 effected a paradigm shift. Proposition 218 was passed by the voters in order to curtail discretionary models of local agency fee determination. (See Silicon Valley, supra, 44 Cal.4th at p. 446 [“As further evidence that the voters sought to curtail local agency discretion in raising funds . . . .”].) Allocation of water rates might indeed have been a purely discretionary, legislative task when Brydon was decided, but not after passage of Proposition 218.

The other key case in City Water’s analysis of this point is Griffith. There, the fee itself varied according to the location of the property, e.g., whether the parcels with wells were coastal and metered, noncoastal and metered, or residential and nonmetered. Objectors to the fee asserted certain tiers in the fee, based on the geographic differences in the parcels covered by the fee, were not proportional to the cost they were paying. One objector in particular complained the fee was improperly established by working backwards from the overall amount of the project, subtracting other revenues, the balance being the augmentation charge, which was then apportioned among the users. (Griffith, supra, 220 Cal.App.4th at p. 600 .) This objector argued that the proportional cost of service had to be calculated prior to setting the rate for the charge.

The court noted the M-l industry manual recommends such a workbaclcwards-from-total-cost methodology in setting rates, and held that the objectors did not attempt to explain why such an approach “offends Proposition 218 proportionality.” (Griffith, supra, 220 Cal.App.4th at p. 600 .) The best the objectors could do was to point to what Silicon Valley had said about assessments, namely, agencies cannot start with “ ‘an amount taxpayers are likely to pay’ ” and then determine their annual spending budget from that. (Ibid., quoting Silicon Valley, supra, 44 Cal.4th at p. 457 .) The Griffith court distinguished the language from Silicon Valley, however, by saying the case before it did not entail any what-the-market-will-bear methodology. (Griffith, supra, 220 Cal.App.4th at p. 600.)

The objectors had also relied on Palmdale for the proposition that “. . . Proposition 218 proportionality compels a parcel-by-parcel proportionality analysis.” (Griffith, supra, 220 Cal.App.4th at p. 601.) The Griffith court rejected that point by stating, “Apportionment is not a determination that lends itself to precise calculation,” for which it cited a pre-Proposition 13, pre-Proposition 218 case, White v. County of San Diego (1980) 26 Cal.3d 897, 903 [ 163 Cal.Rptr. 640 , 608 P.2d 728 ], without any explanation. (Griffith, supra, 220 Cal.App.4th at p. 601.)

When read in context, Griffith does not excuse water agencies from ascertaining the true costs of supplying water to various tiers of usage. Its comments on proportionality necessarily relate only to variations in property location, such as what side of a water basin a parcel might fall into. That explains its citation to White , which itself was not only pre-Proposition 218, but pre-Proposition 13. Moreover, while the Griffith court may have noted that the M-l manual generally recommends a work-backwards approach, we certainly do not read Griffith for the proposition that a mere manual used by utilities throughout the western United States can trump the plain language of the California state Constitution. The M-l manual might show working backwards is reasonable, but it cannot excuse utilities from ascertaining cost of service now that the voters and the Constitution have chosen cost of service.

To the extent Griffith does apply to this case, which is on the (b)(4) issue, we find it helpful and have followed it. But trying to apply it to the (b)(1) and (b)(3) issues is fatally flawed.

c. Penalty Rates

A final justification City Water gives for not tying tier prices to cost of service is to say it doesn’t make any difference because the higher tiers can be justified as penalties not within the purview of Proposition 218 at all. (In the context of art. X, § 2, City Water euphemistically refers to its higher tiered rates as conservation rates as if such a designation would bring them within art. X, § 2 and exempt them from subd. (b)(3), but as we have explained, art. X, § 2, does not require what art. XIII D, § 6, subd. (b)(3) forbids) and designating something a “conservation rate” is no more determinative than calling it an “apple pie” or “motherhood” rate.

City Water’s theory of penalty rates relies on article XIII C, section 1, subdivision (e)(5). This subdivision defines the word “tax” to exclude fines “imposed by” a local government “as a result of a violation of law.” That is hardly a revelation, of course. We may take as a given that Proposition 218 was never meant to apply to parking tickets.

But City Water’s penalty rate theory is inconsistent with the Constitution. It would open up a loophole in article XIII D, section 6, subdivision (b)(3) so large it would virtually repeal it. All an agency supplying any service would need to do to circumvent article XIII D, section 6, subdivision (b)(3), would be to establish a low legal base use for that service, pass an ordinance to the effect that any usage above the base amount is illegal, and then decree that the penalty for such illegal usage equals the incrementally increased rate for that service. Such a methodology could easily yield rates that have no relation at all to the actual cost of providing the service at the penalty levels. And it would make a mockery of the Constitution.

IV. CONCLUSION

All of which leads us to the conclusion City Water’s pricing violates the constitutional requirement that fees “not exceed the proportional cost of the service attributable to the parcel.” (Art. XIII D, § 6, subd. (b)(3).) This is not to say City Water must calculate a rate for 225 Elm Street and then calculate another for- the house across the street at 226. Neither the voters nor the Constitution say anything we can find that would prohibit tiered pricing.

The way Proposition 218 operates, water rates that exceed the cost of service operate as a tax, similar to the way a “carbon tax” might be imposed on use of energy. But, we should emphasize: Just because such above-cost rates are a tax does not mean they cannot be imposed — they just have to be submitted to the relevant electorate and approved by the people in a vote. There is no reason, for example, why a water district or local government cannot, consistent with Proposition 218, seek the approval of the voters to impose a tax on water over a given level of usage — as we indicated earlier, that might be a good idea. However, if a local government body chooses to impose tiered rates unilaterally without a vote, those tiers must be based on cost of service for the incremental level of usage, not predetermined budgets. (For the moment, of course, we need not decide whether such a proposed tax would constitute a general tax or special tax.)

Having chosen to bypass the electorate, City Water’s article X, section 2 position kept it from explaining to us why it cannot anchor rates to cost of service. Nothing in our record tells us why, for example, they could not figure out the costs of given usage levels that require City Water to tap more expensive supplies, and then bill users in those tiers accordingly. Such computations would seem to satisfy Proposition 218, and City Water has not shown in this record it would be impossible to comply with the constitutional mandate in this way or some other. As the court pointed out in Howard Jarvis Taxpayers Assn. v. City of Fresno (2005) 127 Cal.App.4th 914, 923 [ 26 Cal.Rptr.3d 153 ], the calculations required by Proposition 218 may be “complex,” but “such a process is now required by the California Constitution.”

Water service rates to fund the costs of capital-intensive operations to produce more or new water, such as the recycling plant at issue in this case, do not contravene article XIII D, section 6, subdivision (b)(4). While that provision precludes fees for a service not immediately available, both recycled water and traditional potable water are part of the same service— water service. And water service most assuredly is immediately available to City Water’s customers now.

But, because the record is unclear whether low usage customers might be paying for a recycling operation made necessary only because of high usage customers, we must reverse the trial court’s judgment that the rates here are necessarily inconsistent with subdivision (b)(4), and remand the matter for further proceedings with a view to ascertaining the portion of the cost of funding the recycling operation attributable to those customers whose additional, incremental usage requires its development.

By the same token, we see nothing in article XIII D, section 6, subdivision (b)(3) that is incompatible with water agencies passing on the true, marginal cost of water to those consumers whose extra use of water forces water agencies to incur higher costs to supply that extra water. Precedent and common sense both support such an approach. However, we do hold that above-cost-of-service pricing for tiers of water service is not allowed by Proposition 218 and in this case, City Water did not carry its burden of proving its higher tiers reflected its costs of service. In fact it has practically admitted those tiers do not reflect cost of service, as shown by their tidy percentage increments and City Water’s refusal to defend the calculations. And so, on the subdivision (b)(3) issue, we affirm the trial court’s judgment.

Given the procedural posture the case now finds itself in, the issue of who is the prevailing party is premature. That question should be first dealt with by the trial court only after all proceedings as to City Water’s rate structure are final. Accordingly, we do not make an appellate cost order now, but reserve that matter for future adjudication in the trial court. (See Neufeld v. Balboa Ins. Co. (2000) 84 Cal.App.4th 759, 766 [ 101 Cal.Rptr.2d 151 ] [deferring question of appellate costs in case being remanded until litigation was final].)

On May 19, 2015, the opinion was modified to read as printed above.

Webb, The American West, Perpetual Mirage (May 1957) Harper’s Magazine. Reisner, Cadillac Desert: The American West and Its Disappearing Water (1986) page 6. Until Bighorn, there was a question as to whether Proposition 218 applied at all to water rates. In 2000, the appellate court in Howard Jarvis Taxpayers Assn. v. City of Los Angeles (2000) 85 Cal.App.4th 79, 83 [ 101 Cal.Rptr.2d 905 ] (Jarvis v. Los Angeles), held that a city’s water rates were not subject to Proposition 218, reasoning that water rates are mere commodity charges. Bighorn, however, formally disapproved Jarvis v. Los Angeles and held that water rates are subject to article XIII D of the California Constitution. (Bighorn, supra, 39 Cal.4th at p. 217, fn. 5 .) For reader convenience, we will occasionally refer in this opinion in shorthand to “subdivision (b)(1),” “subdivision (b)(3),” “subdivision (b)(4),” and “subdivision (b)(5),” and sometimes even just to “(b)(1)” “(b)(3),” “(b)(4)” or “(b)(5).” Each time those references refer to article XIII D, section 6, subdivision (b) of the California Constitution. Also, all references to any “article” are to the California Constitution. We requested supplemental briefing prior to oral argument to clarify the nature of the issues and precisely what was in, and not in, the administrative record. We are indebted to able counsel on all sides for giving us their best efforts to answer our questions. Such rate structures are sometimes called “inclining” as in the pre-Proposition 218 case Brydon v. East Bay Mun. Utility Dist. (1994) 24 Cal.App.4th 178, 184 [ 29 Cal.Rptr.2d 128 ] (Biydon). Amicus curiae ACWA estimates that over half its members now have some sort of tiered water rate system. As we will say numerous times in this opinion, tiered water rate structures and Proposition 218 are thoroughly compatible “so long as” — and that phrase is drawn directly from Palmdale — those rates reasonably reflect the cost of service attributable to each parcel. (Palmdale, supra, 198 Cal.App.4th at p. 936 .) Ccf stands for one hundred cubic feet, which translates to 748 gallons. (See Brydon, supra, 24 Cal.App.4th at p. 184 .) A precise figure for the usage is complicated by an attempt in the rate structure to distinguish indoor and outdoor use. Technically, tier 2 is tier 1+3 extra ccfs, plus an outdoor allocation that is supposed to average out to a total of 17 ccfs, i.e., 8 ccfs are allocated (on average) for outdoor use. Technically, tier 3 is defined as up to 200 percent of tiers 1 and 2, which, given City Water’s projected 17 ccf average, works out to be 34 ccf. While the consultants distinguished between regula and large lot residential customers, the final structure made no distinction between the two. In 2010, City Water was paying $719 per acre-foot for water from the Metropolitan Water District, and that cost was projected to increase incrementally each year until it reached $1,007 per acre-foot by 2014. One acre-foot equals 435.6 ccf. With a minor qualification that, given our disposition, it need not be addressed in too much detail. A minor issue in the briefing is whether City Water should have made its consultants’ report available for taxpayer scrutiny prior to the public hearing contemplated in article XIH D, section 6, subdivision (c). Since City Water is not able to show its price structure correlates with the actual cost of providing service at the various incremental levels even with the consultants’ report, we need not get bogged down in this issue. Government Code section 53756 provides in relevant part: “An agency providing water, wastewater, sewer, or refuse collection service may adopt a schedule of fees or charges authorizing automatic adjustments that pass through increases in wholesale charges for water, sewage treatment, or wastewater treatment or adjustments for inflation, if it complies with all of the following: “(a) It adopts the schedule of fees or charges for a property-related service for a period not to exceed five years pursuant to Section 53755.” (Italics added.) As described by the court, the fixed cost option was really a “fixed/variable” option, with fixed charges being 60 percent of total costs, the balance being variable. (Palmdale, supra, 198 Cal.App.4th at p. 929 , capitalization omitted.) “It is hereby declared that because of the conditions prevailing in this State the general welfare requires that the water resources of the State be put to beneficial use to the fullest extent of which they are capable, and that the waste or unreasonable use or unreasonable method of use of water be prevented, and that the conservation of such waters is to be exercised with a view to the reasonable and beneficial use thereof in the interest of the people and for the public welfare. The right to water or to the use or flow of water in or from any natural stream or water course in this State is and shall be limited to such water as shall be reasonably required for the beneficial use to be served, and such right does not and shall not extend to the waste or unreasonable use or unreasonable method of use or unreasonable method of diversion of water. Riparian rights in a stream or water course attach to, but to no more than so much of the flow thereof as may be required or used consistently with this section, for the purposes for which such lands are, or may be made adaptable, in view of such reasonable and beneficial uses; provided, however, that nothing herein contained shall be construed as depriving any riparian owner of the reasonable use of water of the stream to which the owner’s land is riparian under reasonable methods of diversion and use, or as depriving any appropriator of water to which the appropriator is lawfully entitled. This section shall be self-executing, and the Legislature may also enact laws in the furtherance of the policy in this section contained.” “In order to have the beneficial use of less than one per cent of the maximum flow of the San Joaquin River on their riparian lands the plaintiffs are contending for the right to use the balance in such a way that, so far as they are concerned, over ninety-nine per cent of that flow is wasted. This is a highly unreasonable use or method of the use of water.” (Herminghaus, supra, 200 Cal. at p. 123 (dis. opn. of Shenk, J.).) Professor Gray’s article is an exceptionally valuable source on the origins of article X, section 2. It was recently noted that Santa Barbara is dusting off a desalinization plant built in the 1990’s to provide additional water for the city in the current drought. (See Covarrubias, Santa Barbara Working to Reactivate Mothballed Desalinization Plant, L.A. Times (Mar. 3, 2015) (as of Apr. 20, 2015) [noting, among other things, that desalinization can be expensive].) Two examples of early, post-Proposition 13 cases that took a strict constructionist view of the provision are Los Angeles County Transportation Com. v. Richmond (1982) 31 Cal.3d 197, 199 [ 182 Cal.Rptr. 324 , 643 P.2d 941 ] (Los Angeles County v. Richmond) (strictly construing Prop. 13’s voting requirements to avoid finding a transportation commission was a “ ‘special district’ ”) and; City and County of San Francisco v. Farrell (1982) 32 Cal.3d 47, 54 [ 184 Cal.Rptr. 713 , 648 P.2d 935 ] (strictly construing words “special tax[]” used in § 4 of Prop. 13 as ambiguous to avoid finding municipal payroll and gross receipts tax was a “special tax”). Brydon expressly relied on Los Angeles County v. Richmond. (See Brydon, supra, 24 Cal.App.4th at p. 190 .) Proposition 218 effectively reversed these cases with a liberal construction provision. (See Silicon Valley, supra, 44 Cal.4th at p. 445 .) Here is the relevant passage from Silicon Valley. “As the dissent below points out, a provision in Proposition 218 shifting the burden of demonstration was included in reaction to our opinion in Knox . The drafters of Proposition 218 were clearly aware of Knox and the deferential standard it applied based on Dawson [v. Town of Los Altos Hills (1976)] 16 Cal.3d 676 [ 129 Cal.Rptr. 97 , 547 P.2d 1377 ].” (Silicon Valley, supra, 44 Cal.4th at p. 445 .) Here and there in City Water’s briefing there are references to a discretionary, legislative power in regard to local municipal water agencies conferred by article XI, section 9, which was a 1970 amendment to the Constitution, though one can trace it back to the Constitution of 1879. Basically, article XI, section 9 gives cities the right to go into the water supply business. We quote its text, unamended since 1970: “(a) A municipal corporation may establish, purchase, and operate public works to furnish its inhabitants with light, water, power, heat, transportation, or means of communication. It may furnish those services outside its boundaries, except within another municipal corporation which furnishes the same service and does not consent. [¶] (b) Persons or corporations may establish and operate works for supplying those services upon conditions and under regulations that the city may prescribe under its organic law.” Article XI, section 9 obviously does not require municipal corporations to establish fees in excess of their costs, so there is no incompatibility between it and the later enacted Proposition 218. The relevant text from article XIII C, section 1, subdivision (e)(5) is: “(e) As used in this article, ‘tax’ means any levy, charge, or exaction of any kind imposed by a local government, except the following: [¶] . . . [¶] (5) A fine, penalty, or other monetary charge imposed by the judicial branch of government or a local government, as a result of a violation of law.”