(dissenting). Because I disagree with the majority’s conclusions relative to both the “retro-activity” of the Supreme Court’s decision and the scope of available relief in this case, I respectfully dissent.1 The majority would dispose of the “rain tax” issue by declaring Bolt v Lansing, 459 Mich 152; 587 NW2d 264 (1998), such an extraordinary case that the Supreme Court’s holding in such case should only be applied prospectively; that is, the only relief any taxpayer can receive with regard to the unconstitutional “rain tax” is future declaratory relief. I find no reason to invoke the exceptional rule of prospectivity in this case. Rather, the Supreme Court’s decision, in my judgment, should be given complete retroactive application. Further, although it is unfortunate that the Headlee Amendment, Const 1963, art 9, §§ 25-32, was not more explicit concerning the precise scope of available relief, I believe that the Headlee Amendment mandates that city of Lansing taxpayers receive full refunds of their “rain taxes” in this case in order *63to effectively “enforce” the amendment. The majority’s finding that a general refund is not allowed, although it is dictum, is a prescription for eroding the Headlee Amendment by precluding its effective enforcement. Thus, I write separately in order to discuss these issues and to express my deep reservations regarding the implications of the majority’s opinion. Ultimately, I would grant the relief of a general “rain tax” refund to Lansing taxpayers in addition to the declaratory relief already granted by the Supreme Court. I join the majority’s opinion only with regard to the attorney fees issue.2
To begin with, I do not agree with the majority’s analysis and holding that the Supreme Court’s opinion in Bolt should have prospective effect only. I agree with the majority’s general statement of the law of retroactivity, but respectfully disagree with its conclusions regarding this issue. By holding that the Supreme Court’s opinion has prospective effect only, the majority essentially finds that the Supreme Court’s determination that the “rain tax” is unconstitutional has sharply limited practical effect. Although the city of Lansing would not be able to continue to collect the “rain tax” in the future, there could be absolutely no remedy for any past collection. This is an extraordinary outcome and effectively creates an incentive for municipalities to attempt to circumvent the Headlee Amendment, knowing that they will have free and clear use of the money that they collect *64through any unconstitutional tax, whatever may subsequently happen in the courts. In my judgment, this extraordinary outcome is not necessary in this case.
The clear general rule is that judicial decisions are to be given complete retroactive effect. Syntex Laboratories v Dep’t of Treasury, 233 Mich App 286, 292; 590 NW2d 612 (1988). This rule is so often invoked that there is generally no discussion whether a particular holding will be applied retroactively. James B Beam Distilling Co v Georgia, 501 US 529, 534; 111 S Ct 2439; 115 L Ed 2d 481 (1991). Even where there is some question about the complete retroactive application of a holding, limited retroactive application is available. Lincoln v General Motors Corp, 231 Mich App 262, 311; 586 NW2d 241 (1998) (WHITBECK, J., concurring). Exclusively prospective application is reserved for truly extraordinary cases that indisputably overrule clear and uncontradicted case law, Hyde v Univ of Michigan Bd of Regents, 426 Mich 223, 240; 393 NW2d 847 (1986), i.e., for cases in which an outcome is truly “ ‘unexpected’ and ‘indefensible’ in light of the law existing at the time of the underlying facts,” Lincoln, supra at 311 (Whitbeck, J., concurring), quoting People v Doyle, 451 Mich 93, 104; 545 NW2d 627 (1996). In Michigan Educational Employees Mut Ins Co [MEEMIC] v Morris, 460 Mich 180, 189; 596 NW2d 142 (1999), the Michigan Supreme Court quoted the United States Supreme Court’s restatement of the criteria for determining retroactivity in Chevron Oil Co v Huson, 404 US 97, 106-107; 92 S Ct 349; 30 L Ed 2d 296 (1971):
“First, the decision to be applied nonretroactively must establish a new principle of law, either by overruling clear past precedent on which litigants may have relied ... or by *65deciding an issue of first impression whose resolution was not clearly foreshadowed .... Second, it has been stressed that ‘we must . . . weigh the merits and demerits in each case by looking to the prior history of the rule in question, its purpose and effect, and whether retrospective operation will further or retard its operation.’ . . . Finally, we have weighed the inequity imposed by retroactive application, for ‘[w]here a decision of this Court could produce substantial inequitable results if applied retroactively, there is ample basis in our cases for avoiding the “injustice or hardship” by a holding of non-retroactivity.’ [Citations omitted.]”
Application of the full three-part test is necessary, however, only if the threshold question whether the decision in question clearly establishes a new principle of law is answered in the affirmative. Lincoln, supra at 312-313 (Whitbeck, J., concurring). “If the decision does not announce a new principle of law, then full retroactivity is favored.” MEEMIC, supra at 190. Further, the Supreme Court has observed:
The fact that a decision may involve an issue of first impression does not in and of itself justify giving it prospective application where the decision does not announce a new rule of law or change existing law, but merely gives an interpretation that has not previously been the subject of an appellate court decision. [Lindsey v Harper Hosp, 455 Mich 56, 68-69; 564 NW2d 861 (1997), quoting Jahner v Dep’t of Corrections, 197 Mich App 111, 114; 495 NW2d 168 (1992).]
In order to give the Supreme Court’s Bolt holding complete prospective application, the decision must have overruled clear and uncontradicted case law. Hyde, supra at 240. What such law was overruled here? It appears that the majority views the fee/tax issue addressed by the Supreme Court in Bolt as one that depends solely on the substantive nature of the project involved rather than on the totality of factors *66assessed case by case according to the Court’s standards. The majority explicitly states that, before Bolt, “Michigan municipalities could have reasonably concluded that charges such as the storm water service charge were user fees.”3 Ante at 47. Yet the majority also appears to look to a single factor—whether the charge is proportionate to and pays only for the use of the service—to determine whether the charge is a tax or a fee. It is not completely clear how this single factor and the nature of the project interact, according to the majority, but it appears that the majority believes that virtually any charge to fund a water/sewage treatment program would be proportionate—the majority specifically contends that “under previous case law, the storm water service charge would have been considered proportionate to plaintiff’s use.”4 Ante at 46. However, this is not the *67correct way to characterize this issue. The Supreme Court specifically stated that “[t]here is no bright-line test for distinguishing between a valid user fee and a tax.” Bolt, supra at 160. Instead, distinguishing between the two “involves consideration of several factors.” Id. at 161.
In making such a statement, the Supreme Court did not devise a wholly new test to distinguish between a fee and a tax, as the majority argues, but rather the Court merely clarified and interpreted existing law on the subject. The Supreme Court’s decision in Bolt did not overrule established precedent, and the distinctions between a tax and a fee relied on by the Court were not matters of first impression. Instead, they were matters of long-standing Michigan law. See id. at 161-169. In reaching its conclusion that the storm water service charge constituted a tax rather than a fee, the Bolt Court examined established case law to address the traditional criteria to be considered when distinguishing between the two, then applied those criteria to the facts of this case. Id. It is true that past cases distinguishing between fees and taxes did not explicitly set out a three-factor test, as in Bolt, or make reference to each of the factors discussed in Bolt. The Supreme Court acknowledged that “ ‘[t]he distinction between a fee and a tax is one that is not always observed with nicety in judicial decisions.’ ” Bolt, supra at 166, quoting 71 Am Jur 2d, State and Local Taxation, § 15, p 352. Previous opinions tended to mention only the specific factors that were most relevant to the case at hand. For example, in Rip*68perger v Grand Rapids, 338 Mich 682, 686; 62 NW2d 585 (1954), on which the majority in the previous Court of Appeals decision in this matter relied to find that the “rain tax” was a fee, 221 Mich App 79, 87; 561 NW2d 423 (1997), the Supreme Court noted that “ ‘no one can be compelled to take water unless he chooses’ ” and that the amount paid reflected the price of the service received. Although not set out explicitly as two factors of a fee/tax test, these statements nevertheless clearly articulate the second and third factors of the Bolt test. In Merrelli v St Clair Shores, 355 Mich 575, 583; 96 NW2d 144 (1959), the Court quoted Vernor v Secretary of State, 179 Mich 157, 167; 146 NW 338 (1914), for the proposition that a fee must be for regulation, not as a means primarily of producing revenue, and that it must not be disproportionate to the cost of the service. These are the first two factors of the Bolt test. Simply reading the Supreme Court’s articulation of the three main factors of the fee/tax distinction indicates that the Court did not invent new law in Bolt: The Court explicitly stated that “this Court has articulated three primary criteria to be considered when distinguishing between a fee and a tax.” Bolt, supra at 161 (emphasis added). The Court then proceeded to list these criteria, citing to the past Michigan Supreme Court cases in which the criteria were used to distinguish between a fee and a tax. Id. at 161-162, citing Bray v Dep’t of State, 418 Mich 149; 341 NW2d 92 (1983); Merrelli, supra; Ripperger, supra; Vernor, supra. The Supreme Court in Bolt merely collected these relevant factors together and set them forth in a more organized and coherent manner. Clarifying a standard is not the equivalent of establishing a new standard, much less *69one that is truly “unexpected.” Even less is such clarification the equivalent of replacing a clear previous standard. Chevron, supra at 106-107.5
The Supreme Court’s decision in Bolt, therefore, was not a decision of first impression that might support prospective rather than retroactive application. While the word “tax,” which the Court was charged with defining and distinguishing from “fee,” was from the language of the Headlee Amendment, neither term was defined in the amendment. The Court merely had to discern the common understanding of these terms. Bolt, supra at 160. Thus, the Court did not concoct a new definition or inteipretation of either the Headlee Amendment or the terms “tax” or “fee,” the Court did not declare the amendment ambiguous and invest it with a meaning that could not have been foreseen, and the Court did not formulate new or technical definitions for the words at issue. Instead, as discussed above, the Court merely relied on the common usage of these words and applied them to the language of the amendment. As with Profit v Citizens Ins Co of America, 444 Mich 281; 506 NW2d 514 (1993), as discussed in MEEMIC, supra at 192, the Supreme Court “simply reaffirmed the existing law, which was misinterpreted by the Court of Appeals.” The Supreme Court in this case made this point very clear, stating, “rather than standing for the proposition that sewage charges are always user fees, as the Court of Appeals majority contended, 221 Mich App 86-87, Ripperger *70actually articulated relevant criteria for determining whether a charge is a fee or a tax.” Bolt, supra, 459 Mich 163, n 12. Although the majority here is apparently convinced that the Supreme Court in fact announced a new rule of law in Bolt, this interpretation is directly at odds with the express language of the Supreme Court’s decision itself. For purposes of deciding this issue, we must accept the Supreme Court’s own determination that it was merely collecting and organizing the existing law of previous Supreme Court decisions regarding fees and taxes—a determination that, in my judgment, is altogether accurate.
Thus, the Bolt decision neither announced a new rule of law nor altered any clear existing law. It neither employed unforeshadowed interpretative tools nor reached any unforeseeable conclusion. The Supreme Court merely applied already identified factors to the new circumstances of the case. The fact that the defendant could make a reasonable argument in response to plaintiff’s prevailing argument does not make the ultimate holding surprising or unanticipated. Although the majority states that Headlee cases will not always be applied prospectively, following the majority’s basic reasoning here another Headlee suit on a different subject would have to be given prospective effect also, because there is often a reasonable argument to be made on both sides. However, this is not the test for prospectivity. Quite simply, there is no basis here for departing from the normal presumption of retroactivity.6
*71In particular, to allow such a departure in the context of the Headlee Amendment tax/fee dispute would be to dilute considerably the effect of the amendment, because, in virtually all such disputes, the application of a ‘totality of factors’ test will inevitably entail some uncertainty in terms of the ultimate result, while also affording both sides some measure of reasonable argument. Such circumstances, however, are not “extraordinary” and should not be viewed as sufficient to alter the normal rule of retroactivity.
Accordingly, having determined that the Supreme Court’s holding in this case must actually be applied to this case, I turn to the main issue: the proper scope of relief. The majority addresses this issue, although given the majority’s decision to apply the Supreme Court’s holding only prospectively, it is merely dictum. Under the majority’s holding, plaintiff here can receive no more than a declaratoiy judgment that the “rain tax” is unconstitutional and that he personally does not have to pay the tax in the future, regardless of the scope of relief he requested in his complaint or the scope of relief provided for in the Headlee Amendment. The majority concludes that accepting plaintiff’s substantive position regarding his proposed remedy would effectively result in an “ex post facto” “constitutional class action” that, is unnecessary, because a plaintiff could have brought a traditional *72class action in this type of case. I respectfully but strongly disagree.
We review de novo constitutional issues and constructions. Kuhn v Secretary of State, 228 Mich App 319, 324; 579 NW2d 101 (1998). A primary rule in interpreting a constitutional provision is the rule of “common understanding”:
“A constitution is made for the people and by the people. The interpretation that should be given it is that which reasonable minds, the great mass of the people themselves, would give it. ‘For as the Constitution does not derive its force from the convention which framed, but from the people who ratified it, the intent to be arrived at is that of the people, and it is not to be supposed that they have looked for any dark or abstruse meaning in the words employed, but rather that they have accepted them in the sense most obvious to the common understanding, and ratified that instrument in the belief that that was the sense designed to be conveyed.’ (Cooley’s Const Lim 81) ” (Emphasis added.) [Traverse City School Dist v Attorney General, 384 Mich 390, 405; 185 NW2d 9 (1971).]
A second rule is that courts may also consider the circumstances leading to the adoption of the constitutional provision and the purpose behind the provision in order to clarify the meaning of an amendment’s language. Macomb Co Taxpayers Ass’n v L’anse Creuse Public Schools, 455 Mich 1, 7; 564 NW2d 457 (1997). This Court has stated, in the context of determining whether money damages were recoverable in a suit brought pursuant to § 29 of the Headlee Amendment, that the words of the amendment “ ‘ “are to be applied to the subject matter and to the general scope of the provision, and they are to be considered in light of the general purpose sought to be accom*73plished or the evil sought to be remedied by the constitution.” ’ ” Wayne Co Chief Executive v Governor, 230 Mich App 258, 264; 583 NW2d 512 (1998) (citations omitted). The voters for the Headlee Amendment “ ‘were . . . concerned with ensuring control of local funding and taxation by the people most affected, the local taxpayers. The Headlee Amendment is the voters’ effort to link funding, taxes, and control.’ ” Macomb Co Taxpayers, supra at 7, quoting Durant v State Bd of Ed, 424 Mich 364, 383; 381 NW2d 662 (1985). “The ultimate purpose [of the Headlee Amendment] was to place public spending under direct popular control.” Waterford School Dist v State Bd of Ed, 98 Mich App 658, 663; 296 NW2d 328 (1980). The amendment “grew out of the spirit of ‘tax revolt’ and was designed to place specific limitations on state and local revenues.” Id.
Any discussion of the remedy for a violation of the Headlee Amendment must, of course, begin by reviewing the language of the amendment. Durant v State Bd of Ed, supra at 378. Although the Supreme Court held that the “rain tax” violated Const 1963, art 9, § 31, which “prohibits units of local government from levying any new tax or increasing any existing tax above authorized rates without the approval of the unit’s electorate,” Durant v Michigan, 456 Mich 175, 183; 566 NW2d 272 (1997), it is Const 1963, art 9, § 32 specifically that gives any taxpayer standing to enforce §§ 25 through 31. Section 32 states:
Any taxpayer of the state shall have standing to bring suit in the Michigan State Court of Appeals to enforce the provisions of Sections 25 through 31, inclusive, of this Article and, if the suit is sustained, shall receive from the applica*74ble unit of government Ms costs incurred in maintaining such smt.
Several factors must be analyzed to determined the specific nature of the remedy that will effectively “enforce” the Headlee Amendment in this case. First, I address what the “great mass of the people” intended in directing the Court of Appeals “to enforce” the amendment. Durant v Michigan, supra at 204. There is no explicit statement in the amendment regarding the type of enforcement intended in any particular situation. However, traditionally, a private citizen had no standing to enforce a public right where the citizen was not injured in any manner differently than the general public, and thus a taxpayer had no standing to challenge expenditures of public funds where the threatened injury to the taxpayer was no different than that to taxpayers generally. Waterford, supra at 662. This Court held that the plain language of § 32 altered the status quo in that it indicates an intent to provide broad standing to taxpayers to enforce the substantive provisions of the amendment.7 Id. This section facilitates control by the public by allowing a single taxpayer to bring suit to enforce the constitution directly in the Court of Appeals. Id. at 663.
Although § 32 does not explicitly address the scope of the remedy that can be pursued by an individual taxpayer, it does do so by implication, in my judgment. A person looking at the Headlee Amendment’s broad grant of standing to any single taxpayer “to *75enforce” the amendment would presume that such taxpayer had the power to effectively enforce the amendment. Certainly if a single taxpayer’s lawsuit is able to result in the declaration of a law enacted by a governmental body as unconstitutional and therefore void in the future with regard to all taxpayers (as in the instant case), then the persons comprising the “great mass of the people” would reasonably believe that any additional remedy derived by the suing taxpayer in “enforcement” of the Headlee Amendment would also apply to all taxpayers, not merely to the individual suing party. I do not believe that they would understand that plaintiff Bolt’s “enforcement” of the amendment, although generally applicable to everyone in terms of its future relief, i.e., the unconstitutionality of the tax, was only specifically applicable to Bolt in terms of its past relief, i.e., restoration of the status quo and the refund of wrongly collected taxes. This broad scope of relief is not normally available to a party-—generally a person must be a party to a lawsuit in order to obtain a remedy from it. Yet just as the Headlee Amendment altered the status quo regarding standing, the citizenry who voted for the Headlee Amendment would reasonably understand it also to have changed the status quo regarding the scope of relief allowed by the amendment.8 The common understanding would be that the Headlee *76Amendment allows whatever relief is necessary to effectively “enforce” the amendment. “[T]he voters have left it to this Court to delineate the proper remedy,” depending on the specific violation and facts of each case. Durant v Michigan, supra at 214, n 45.
Second, I address the scope of relief required for “enforcement” in this case. The Supreme Court has held that the Headlee Amendment’s grant of enforcement powers “was intended as a general directive, giving this Court the duty and authority to enforce [the Headlee Amendment, §§ 25 through 31] in the way that would most effectuate the balances struck by the people in the Headlee Amendment.” Durant v Michigan, supra at 205. Therefore, any remedy must be broad enough to actually enforce the amendment and its purpose. Such obligation to actually enforce the amendment is the equivalent of the concept of the ‘effective’ enforcement of the amendment that seems implicit, as noted above, in the common understanding of its language. To “enforce” a measure means to correct sufficiently the violation of a measure so that obedience is compelled and the law is made strong. See id. at 208; Random House Webster’s College Dictionary (2d ed, 1997), p 432. “Simply put, if declaratory relief is not sufficient to compel obedience to the constitutional mandate . . . then the electorate has authorized that additional relief be granted.” Durant v Michigan, supra at 208.9
In this case, the Supreme Court determined that the “rain tax” violated § 31 of the Headlee Amendment, by establishing a tax on the citizens of the city of Lan*77sing without their prior approval by vote. It is the entire Lansing Ordinance 925, which imposed the “rain tax,” that is unconstitutional and therefore null and void here. Thus, the city must not just discontinue the collection of this illegal tax, but we must also address the fact that it has already illegally collected this tax for several years. The tax did not become unconstitutional as of the time of the Supreme Court’s decision; rather it was unconstitutional ab initio and therefore illegal for the city to collect. Thus, declaratory relief alone is insufficient to remedy the violation here because it does not address this issue in plenary fashion. The majority states that plaintiff did not seek a monetary refund, instead seeking only declaratory and injunctive relief, arguing that he cannot receive any type of refund because he did not include this in his demands for relief. However, the declaratory judgment rule, MCR 2.605(F), states that “[further necessary or proper relief based on a declaratory judgment may be granted after reasonable notice and hearing, against a party whose rights have been determined by the declaratory judgment.” Thus, it was only necessary that plaintiff here request declaratory action, and any additional remedies could then be determined by the Court. Because declaratory relief is not “sufficient to compel obedience to the constitutional mandate” in the present case, the “electorate has authorized that additional relief be granted.” Durant v Michigan, supra at 208. In this case, the additional relief required is a general refund that would restore the status quo. Indeed, the Supreme Court itself has commented on the “obvious merit” of the refund remedy for a violation of § 31:
*78[Section] 32 does not mention a refund for taxes unconstitutionally collected under § 31. Yet the remedy of a refund is so obvious that the ordinary person would say that, without such a remedy, the court would not be enforcing § 31. We continue to be of the belief that the command of § 32 is general and empowering and contemplates a money judgment where necessary. [Durant v Michigan, supra at 216.]
Although this case requires a broader refund than that in Taxpayers Allied for Constitutional Taxation [TACT] v Wayne County, 450 Mich 119; 537 NW2d 596 (1995), which the Durant Court was discussing above, the general refund here is also “obvious” and “necessary” to enforce § 31.10 The majority claims that any reliance on Durant is misplaced because it did not “mandate refunds to all affected taxpayers every time a plaintiff successfully seeks injunctive relief from a Headlee violation.” Ante at 55. I agree that refunds are not mandated in every successful Headlee Amendment suit, where, for example, there are unusual financial or administrative burdens imposed on either a governmental body or this Court. The relief that would effectively remedy the situation in each individual case would have to be determined case by case. Thus, in contrast to the majority’s claim, there is no “dangerous precedent” set here. However, a general refund is certainly one possible remedy where a tax has been unconstitutionally levied, in light of both *79the language and the purpose of the Headlee Amendment. Indeed, in the instant case, I believe it to be an obvious remedy.
Third, I address the claim that effective “enforcement” of the Headlee Amendment could be accomplished through methods available in the existing law, such as individual suits by each taxpayer to receive a refund; thus no “constitutional class action” is necessary or allowed. I agree that if an array of equally effective remedies is available, this Court should choose those less onerous, in order to “most effectuate the balances struck by the people in the Headlee Amendment.” Durant v Michigan, supra at 205, 225 (Brickley, J., dissenting). I also agree that each taxpayer in the city could have individually filed a lawsuit within one year of the collection of the “rain tax” in order to receive full relief for the violation of the Headlee Amendment. However, I cannot conceive of a procedure better designed to undermine the purpose of the Headlee Amendment, and to erode its functionality as an effective check on government, than to require that every taxpayer must file an individual lawsuit. As the instant case adequately demonstrates, the great majority of taxpayers will never file such a lawsuit because it would impose far too much burden and financial risk on them. Instead, the principal role of the taxpayer in ensuring compliance with the Headlee Amendment is to vote on proposed tax increases, a relatively minor imposition on their personal resources, as opposed to the much higher imposition of requiring each taxpayer to file a lawsuit against his government to enforce compliance even where the law in question has already been found unconstitutional. Under this proposition, because *80most people are unlikely ever to file suit, a governmental body apparently faces few practical consequences of imposing an unconstitutional tax on its citizens. In my judgment, to remove this element of enforcement from the Headlee Amendment, the deterrent element inherent in a governmental body being made aware that it will not be able to gain financial advantage from an unconstitutional tax, would be to ineffectively “enforce” the amendment. As stated in Durant v Michigan, supra at 206, this “would authorize the [government] to violate constitutional mandates with little or no consequence.” Governmental bodies and their officials react to incentives and disincentives in the same manner as private actors; to introduce into the Headlee Amendment a substantially weaker incentive for compliance on the part of governmental bodies would be to sharply erode the ability of the individual taxpayer to “enforce” the provisions of the amendment.
Fourth, I address the contention that plaintiff should be forced to abide by class action rules in order to obtain a remedy that would benefit all taxpayers. In contrast to the majority, I am not convinced that a class action is required under the Headlee Amendment. The majority claims that plaintiff should have complied with the class action rule, MCR 3.501, if he wanted broader relief than an individual remedy. There are several factors relevant to the class action issue:
(a) Allowing or requiring a class action to enforce the Headlee Amendment would be to graft additional requirements into the amendment that axe nowhere mentioned (or probably even contemplated) in the text of the amendment itself. The plain language of *81§ 32 logically contemplates a class action. The amendment simply and plainly states that “[a]ny taxpayer . . . shall have standing ... to enforce the provisions of Sections 25 through 31.” The provision clearly refers to more than standing because it directs taxpayers to “enforce” the amendment. Yet there is no mention of a class action requirement in a place where this would be obviously mentioned if it was contemplated. There is not even a mention of a “group of taxpayers” enforcing the amendment, but merely “any taxpayer.” The most logical inference is that one individual can obtain whatever remedy is necessary to “enforce” the amendment.
(b) Class actions beget a host of complex issues regarding certification, representation, notice, jurisdiction, and several other matters. Although the majority states that plaintiff could have “easily” joined a representative taxpayer as a party to represent a class, and refers to a class certification proceeding as “simple,” these requirements are not mere procedural hurdles. Rather, they are substantive burdens that are designed to be difficult to satisfy. 5 Martin, Dean & Webster, Michigan Court Rules Practice (3d ed), p 18. Such burdens are not reasonably contemplated in the amendment, which seems instead to attempt to designate a relatively straightforward individual suit that will fully “enforce” the entire amendment.
(c) The Headlee Amendment is strikingly comprehensive and detailed, in contrast to the succinctness of most other constitutional amendments, particularly those concerned with the relationship between the individual and the government. Under such a constitutional scheme, it is logical to infer that if the framers had intended to require class actions in order for *82a taxpayer to effect a remedy for all, they would have so clearly stated.
(d) The Court of Appeals is not well equipped to handle class actions. The same requirements that make class actions extremely onerous for representative parties also weigh heavily on the judiciary. Even if MCR 3.501 does apply to the Court of Appeals,11 this Court was not designed to effectively manage class actions. Unlike the circuit courts, which have traditionally had responsibility for managing class actions, the Court of Appeals acts exclusively on the basis of three-person panels, undoubtedly a far more cumbersome structure for administering the decisions involved in certifying and managing a class. Although the Court perhaps could take on the additional responsibilities of fact finding, had class actions been truly intended by the Headlee Amendment I would have expected, given its wealth of specificity in other regards, that the amendment would have specifically identified this highly uncustomary responsibility.
Thus, the absence of any reference to class actions in the Headlee Amendment in the face of the heavy burdens that such actions inevitably impose, both on parties and the courts, strongly suggests that class actions are not intended as a precondition for relief for all in a suit brought to enforce the Headlee *83Amendment. A class action requirement is not only rendered inapposite by an absence of reference to such actions in the amendment, but by the obvious barriers that such a requirement would impose to the effective “enforcement” of the amendment.
Fifth, requiring abidance with class action requirements is not required under the Headlee Amendment because the Headlee Amendment clearly contemplates the possibility of the enforcement of the amendment before there is a designated class that could bring a class action. TACT, supra at 124, n 7, stated:
In fact, the only type of Headlee claim that would accrue at the time the resolution is passed is a claim brought merely on behalf of the public, as opposed to a claim brought by a taxpayer who has been or is about to be subject to the tax.
Thus, it is clear that the Supreme Court has recognized that the Headlee Amendment allows an individual suit on behalf of the public before the time that a class action is ripe. Yet it would be illogical to allow the individual suit, then disallow any general remedy. This serves little apparent purpose and would essentially preclude the type of Headlee claim specifically allowed by the Supreme Court in the quote above. The Supreme Court’s interpretation in TACT supports plaintiff’s argument that there are several different kinds of suits that can be filed under Headlee. As in this case, there is the declaratory judgment action brought by one taxpayer on behalf of the entire public, which is in contrast to the more typical private suit brought to recover individual damages. Where a constitutional provision conflicts with a general court *84rale regarding class actions, it is clear that the constitutional provision providing for an alternative to the class action must prevail. Regardless, therefore, whether a class action is or is not available to a Headlee plaintiff, a plaintiff in Alexander Bolt’s position is obliged only to follow the plain language of the Headlee Amendment. Grafting the complex and intricate requirements of a class action onto the Headlee Amendment is not something that this Court should undertake without considerably more evidence than presented here that this was intended by the framers and voters of the amendment.
In my judgment, it is clear that the Headlee Amendment allows a single taxpayer to “enforce” the amendment for the general public by whatever remedy is reasonably necessary. Even if a class action were available, plaintiff was not required to wait until he could fulfill all the requirements, of a class action because there was no ambiguity about the type of action taken by plaintiff here. Suit here was brought before the time that a class action for refunds was ripe and it was brought to “enforce” the Headlee Amendment on behalf of the general public by preventing the imposition of a tax and obtaining a declaratory ruling of its unconstitutionality. This is precisely the type of suit described in TACT, supra at 124, n 7. Plaintiff’s suit here appears to be a bona fide effort to enforce the Headlee Amendment on behalf of the general public in Lansing, rather than a suit principally for personal gain.
In addition, for the benefit of all the parties involved, we do not want to discourage future plaintiffs from bringing Headlee actions as quickly as possible by forcing them to invoke complex and intricate *85class action procedures before filing suit. To require a class action as a precondition for broad relief is virtually to ensure that future Headlee plaintiffs will act in a considerably less expeditious manner in filing suit. Of course, some unknown number of potential Headlee plaintiffs will be discouraged altogether from filing suits as a result of the burdens of these procedures, as well as because of the greater difficulty in finding counsel prepared to provide the necessary representation. Additionally, given the additional length of the class action certification process, where a Headlee class action ultimately prevails, the governmental body almost certainly will have wrongfully obtained larger amounts of revenue than in the case of an individual enforcement action. Whether such additional revenues are ultimately refunded to the taxpayers, or retained by the governmental bodies, more disruption will inevitably arise in the fiscal process. It is extremely hard to fathom how any of these circumstances are gauged to enhance the effective “enforcement” of the Headlee Amendment.
Sixth, effective “enforcement” should not hinge on the judicial system’s own inherent lack of expedition in deciding cases. Had the judicial process been able to immediately resolve this matter when it was filed—before the “rain tax” actually became due—and afford plaintiff the relief for which he asked, there would be no question or need of further remedy. However, the judicial process is often slow-moving, and, while waiting for a declaration of constitutionality, most taxpayers in Lansing paid the tax for several years. That the ultimate court decision issued after the tax had been paid by most taxpayers does not affect the type of suit filed and the original purpose *86of such suit, in my judgment. It does not affect the fact that, when plaintiff filed his suit, the relief sought would have ensured that no new taxes were imposed on any taxpayer. At this juncture, plaintiff still represents the public at large, but the prolonged nature of this suit has now necessitated a fuller remedy than that originally envisioned. Where a lawsuit has been brought early, as here, and declaratory judgment for the general public is the goal, a general remedy for all should be an option for the court. We should not penalize the public for the inherent deficiencies of the judicial branch.
Seventh, the majority’s holding would result in an inequality among similarly situated taxpayers that would be highly unfair and thus would undermine the integrity of the amendment as an effective tool for these same taxpayers in controlling spending and taxing. Should only those who refused to pay taxes benefit from the Supreme Court’s decision in Bolt? This creates an inequitable outcome of a sort similar to that in Armco Steel Corp v Dep’t of Treasury, 419 Mich 582; 358 NW2d 839 (1984). In Armco, following its unauthorized audits of certain taxpayers the Department of Treasury canceled deficiency assessments for those taxpayers who had refused to pay those assessments, but refused to grant refunds to those taxpayers who had paid their assessments and then later sought repayment. The Supreme Court found this to be unfairly disparate treatment. Id. at 595. The Court stated:
When faced, as in this case, with a choice between securing that which is due under the law and upholding the constitutional requirements of uniformity and equality, hie lat*87ter is to be preferred “as the just and ultimate purpose of the law.” [Id. at 594-595.]
In this case, the “choice” is even easier than in Armco because here there is nothing rightfully due under the law—the “rain tax” was declared void as of the moment of its passage—and equality requires that all similarly situated taxpayers be treated equally. See Horrigan v Klock, 27 Mich App 107; 183 NW2d 386 (1970). Thus, defendant should not be allowed to keep the “rain taxes” paid by some Lansing residents while those who did not pay benefit exclusively. This would leave those who have not complied with the “rain tax” better off than those citizens who did comply with the “rain tax.” I cannot think of a less prudent public policy nor one better designed to undermine respect for the law generally and for the constitutional language of the Headlee Amendment specifically.12
Eighth, the upshot of this inequality among taxpayers countenanced by the majority’s decision is the creation of several perverse incentives inimical to the orderly processes of the law. First, some number of taxpayers will be encouraged to withhold their taxes if there is any question regarding their validity under the Headlee Amendment. Where such withholding is *88the only way to ensure that their payments are not forfeited even where a tax is later declared invalid, some taxpayers will risk the penalties for nonpayment. Withholding money is a much less burdensome course of action than having to file a lawsuit seeking to recover invalid taxes; inevitably, some taxpayers will prefer the former “option.” Second, if a governmental body can keep all the money it collected from an invalid tax, there will be concomitantly less interest among governmental bodies in conscientiously ascertaining the validity of tax measures before enacting them. Laws have consequences—they impose incentives and disincentives for undertaking various actions. In part, enactment of the Headlee Amendment appears to reflect a judgment by the citizenry that additional constraints needed to be placed on the taxing propensities of their elected representatives, perhaps because the incentives acting on such officials to spend increasing amounts of public monies (and the resultant need to raise revenues for such spending) outweighed the disincentives in this regard. The purpose of the amendment is clearly to circumscribe the actions of public officials who are subject to this aggregation of incentives and disincentives. I fear that the majority’s opinion would effectively undermine this judgment in unpredictable ways by eliminating a critical disincentive for new taxes contained in the Amendment—that there will be utterly no benefit to a governmental body from enacting what is ultimately determined to be an unconstitutional tax. By eroding this disincentive and enabling governmental bodies to keep at least some measure of their wrongfully obtained funds, the following questions come to mind:
*89(1) Will governments be able to more easily justify higher levels of future taxes to the people on the grounds that wrongfully obtained funds will go to waste unless additional funds are obtained by a tax approved by the voters?
(2) Will governments be able to use the wrongfully obtained funds for any public purpose as opposed only to the purpose for which the funds were ostensibly raised?
(3) Will governments use the wrongfully obtained funds for existing programs and thereby justify higher levels of future spending (and therefore taxes) on the ground that new levels of current services need to be maintained?
(4) Will governments use the wrongfully obtained funds for new programs, in turn creating new spending constituencies, and thereby more easily justify higher levels of future spending (and therefore taxes) on the ground that such new programs need to be maintained? There is no basis in the language of the Headlee Amendment for experimenting with the array of disincentives for new taxes established by the amendment. Ensuring that the Headlee Amendment is effectively enforced is no less important than ensuring the effective enforcement of any other part of the Michigan Constitution. Headlee is no mere statute—it is a part of the constitution, entitled to no less regard than any other part. Because it deals with a nontraditional matter of constitutional focus makes it no less important—the voters of Michigan have decreed this fact—and we are obligated to uphold this judgment.
Ninth, I question why the Supreme Court remanded this case in the first place when plaintiffs personal rights were fully vindicated by its decision. Plaintiff *90did not pay the “rain tax” and thus does not need a refund for himself. What then is the reason for the remand? Although one could argue that perhaps the Supreme Court remanded on the basis of the attorney fees issue, this seems a rather trivial reason to send the entire case back to this Court without explanation to that effect. In my judgment, it would seem that the Supreme Court had some reason rather more substantial in mind when it chose to remand, in particular, the crafting of a remedy additional to that already granted above.
Tenth, I question why the drafters of the Headlee Amendment would have provided under § 32 for the public reimbursement of plaintiffs legal costs if he were merely vindicating a personal right. In my judgment, the award of costs to a taxpayer is intended to encourage exactly the type of suit brought by plaintiff in this case—a suit on behalf of the general public. There would be no need of the rather extraordinary award of costs where a taxpayer was merely vindicating his personal rights. However, where he has served the public good, the award of costs is easily understood. Further, I note that it is in the governmental body’s own best financial interest to enable a single lawsuit by a single taxpayer in order to determine the constitutional validity of a fee or tax for the entire public rather than a series of such suits in order to vindicate merely personal rights. Undoubtedly, the same could be said with regard to a class action. The attorney’s fees due plaintiff here would, almost certainly, have been far higher had this been a class action than a mere individual “enforcement” suit.
In my judgment, the majority’s holding, that a general refund is unavailable under the Headlee Amend-*91merit, would result in the difficulties laid out above and thus would undermine the rules of enforcement in the Headlee Amendment that we are attempting to uphold in Bolt. However, the majority states that it does not hold or imply that “relief in Headlee cases should or will always be prospective only,” ante at 49, and that the instant case provides potential future governmental defendants clear notice of the rules for fees and taxes under the amendment. Therefore, knowing the law as a result of Bolt, future taxing jurisdictions apparently will apply the correct constitutional test and the question of refunds will not arise. Respectfully, I do not believe that the answer is quite this simple. Adthough the Supreme Court in Bolt acted conscientiously in summarizing the standards for determining whether charges are fees or taxes, the standards set forth in Bolt are still not designed to lead to a purely mechanical determination of constitutionality in future cases. Contrary to the majority’s interpretation, Bolt did not set forth a fee/tax distinction based solely on the substantive nature of the project involved or on the answer to a single inquiry, e.g. the proportionality of the fee or tax. We cannot now assume that any charge for a water/sewer project will be deemed a tax simply because the charge for the water/sewer project in Bolt was determined to be a tax. Instead, there must be an application case by case of the “totality of factors” test, which will inevitably entail some uncertainty in terms of the ultimate result. It is almost certainly true that Bolt will assist future taxing jurisdictions in understanding and applying the Headlee Amendment in their own unique situations, such that the number of charges found unconstitutional will likely be reduced. That, of *92course, is one of the puiposes of all sound legal precedents—the minimization of future legal uncertainties and a resultant diminution in the need for litigation. However, the fact that the number of future legal controversies may be minimized by the Supreme Court’s decision in Bolt is no warrant for failing to offer relief in the instant case that, in my judgment, is required by the Michigan Constitution.
In conclusion, I believe that a broad remedy, according relief to all taxpayers in the city, is not only allowed under the Headlee Amendment, but is necessary in this case in order to uphold the amendment. While there may be instances in which such relief may be limited because of administrative or financial burdens, defendant here has not purported to make such a showing. In any event, plaintiff here acted in as expeditious a manner as possible in bringing his lawsuit, and defendant was apprised as early as possible that its levy was subject to some constitutional shadow.13 There has been no bad faith or ‘gamesman*93ship’ on the part of plaintiff and, as a result, no lengthier than necessary period for which the tax refunds should be made. Further, it appears quite clear who should receive refunds and in what amounts. Therefore, in view of the absence of unusual circumstances that would complicate refunding of the tax, and in recognition of the fact that unconstitutional statutes are void ab initio, Horrigan, supra at 108-109, all taxpayers who paid the “rain tax” in any years should receive a full refund.14
The scope of relief available here derives from the constitutional grant of standing to individual taxpayers to effectively “enforce” the Headlee Amendment. The relief required to effectively “enforce” the amendment will vary depending on the section violated and the facts of each case, but the amendment contemplates broad relief. Here, such a broad remedy is required. Mere declaratory relief will not sufficiently enforce the amendment, but rather will weaken its effectiveness. Declaring that a governmental body acted contrary to the constitution in taxing the people, but then allowing such body to retain much or all of the money that it wrongfully collected, will only dilute the authority of the constitution by eroding incentives for compliance on the part of government. “The intent of the people in enacting art 9, § 32 of the Michigan Constitution was not to enact a consti*94tutional provision that could not be effectively enforced.” Durant v Michigan, supra at 206-207.
In considering the “retroactive” application of the Supreme Court’s decision in Bolt, I would emphasize at the outset that this Court is not assessing the effect of the decision on another legal case or controversy. Rather, we are assessing the decision’s effect on the same case or controversy in which the decision was in fact rendered. To describe this as a “retroactive” application of Bolt is at least a debatable use of the term. But see Washtenaw Co v State Tax Comm, 422 Mich 346, 378; 373 NW2d 697 (1985); Placek v Sterling Heights, 405 Mich 638, 662; 275 NW2d 511 (1979).
In rejecting the notion that only the average hourly fee should be considered a “reasonable” fee for purposes of reimbursement under the Headlee Amendment, I do not read the majority opinion as suggesting that any hourly fee will be deemed “reasonable,” merely that under the particular circumstances of this case, plaintiff’s attorney’s fee is reasonable.
The majority further defines as the issue of “first impression” in this case: “whether a charge assessed to property owners to fund a federally mandated project is a user fee, or a tax.” Ante at 45. Later, the majority asserts, “Given the Ripperger/MereUi test and this pattern of case law, Michigan municipalities could have reasonably concluded that charges such as the storm water service charge were user fees, and not subject to Headlee Amendment requirements.” Ante at 47. Finally, it asserts that as of the date of the Supreme Court’s decision in Bolt, “all municipalities were put on notice that charges such as the storm water service charge are taxes, subject to the requirements of the Headlee Amendment.” Ante at 49. I respectfully believe that the majority has missed the larger point of Bolt.
The majority cites Detroit Water & Sewage Dep’t v Michigan, 803 F2d 1411 (CA 6, 1986), stating that the court there concluded that the charge for the treatment of runoff storm water “was not a tax, but a fee that the city charged for a service rendered.” Ante at 47. However, the federal court did not specifically address the issue whether the charge was a fee or a tax. The issue was whether there was any benefit to the county from the drainage and treatment of storm water from the county’s roads; thus the court assumed the charge was a fee and addressed whether the county was required to pay it. The case did not hold that any charge for water or sewage treatment was a fee. Further, the case has no precedential effect on this court because it was not decided by the Michigan *67Supreme Court or the Court of Appeals. Indeed, the federal court did not even address any Michigan law regarding the difference between a fee and a tax.
As the majority itself observes, “As we recognized in our initial decision . . . resolution of the issue was inordinately difficult because the Headlee Amendment does not define either the term ‘fee’ or the term • ‘tax.’ ” Ante at n 3, 45. Even given this perspective, it is hard to see what “clear, past precedent” has been overturned by the Supreme Court in Bolt.
Having determined, under the first part of the Chevron test, that Bolt did not establish a new principle of law, there is no need to address the second and third parts of the Chevron test. In any event, I do not find the *71majority’s analysis of these two factors to be persuasive. With regard to the second consideration—whether retrospective application will further or retard the operation of the “rule in question”—any interruption in the application of the long-standing criteria, as I argue elsewhere, would necessarily undermine the effectiveness of the Headlee Amendment’s enforcement. With regard to the third criteria, the majority makes no showing of any unusual financial or administrative burden on defendant.
WMle Michigan statutes have relaxed the common-law bar on taxpayer suits somewhat, it is clear that “§ 32 was intended to further ease the limitations on taxpayer suits.” Waterford, supra at 663.
The standing provision of the Headlee Amendment enables “any taxpayer ... to enforce” the amendment. This provision is not limited to the issue of standing but also encompasses the scope of available relief. The grant of power to taxpayers “to enforce” the amendment is made in the exact same sentence as that which transforms the status quo regarding standing. Although obviously less explicit in its treatment of the scope of relief than it is with regard to standing, the language and proximity of the enforcement language to that of the standing language arguably suggest an intention to alter the status quo regarding the scope of relief also.
The scope of enforcement available under § 32 depends on which section of the Headlee Amendment is violated. See Durant v Michigan, supra at 216.
I also note that this remedy is consistent with the one-year statute of limitation provision of MCL 600.308a(3); MSA 27A.308(1)(3). This is not a remedial invocation of a “constitutional class action,” but rather the application of a broad remedy to the public represented by plaintiff in his suit. Because plaintiff filed suit well within one year of the time of the “rain tax” application, a refund is not barred in any way by the statute of limitation.
It is not my contention that MCR 3.501 is simply “inapplicable” to this Court, as the majority implies. Instead, I believe that the fact that Headlee suits may be instituted in the Court of Appeals when the Court of Appeals is so clearly not structured to manage a class action is one factor among several that show that the framers and voters of the Headlee Amendment intended to institute a new system of taxpayer suits under Headlee, without the requirement of a class action. Regardless whether a class action can technically take place within the Court of Appeals, I do not believe that the Headlee Amendment requires one in this case.
The majority asserts that “[t]o the extent that there is any differential treatment of Lansing taxpayers, this difference is based on those who sought relief and those who did not.” Ante at 58. This is manifestly not so. Those Lansing taxpayers who withheld their “rain tax” payments, and who therefore had no need to seek relief, are in precisely the same position as those who “sought relief” (except that the former have not had to incur legal costs). The differential treatment here is instead between those who paid their taxes notwithstanding the pendency of a Headlee Amendment lawsuit and those who did not.
The majority asserts that the monies collected have already “been committed in good faith to the project.” Ante at 50. While I have no doubt that defendant acted in good faith in this regard, it was nevertheless aware of the pendency of the lawsuit that called the propriety of its actions into question. Had defendant acted in bad faith—and I emphasize again that I do not believe this to be the case—it is difficult to see how it would have acted any differently. I do not believe that defendant acted in bad faith because I agree with the majority’s statement in note 6 that “Defendant had the right to continue collecting the tax until a court ordered it to desist. . . . [W]e cannot countenance plaintiff’s suggestion that a municipality must impose a restraining order on itself whenever a taxpayer files a Headlee action.” Ante at 50. Good faith notwithstanding, however, defendant must accept the consequences of its conduct in spending the “rain tax” where it has been made fully aware of the constitutional controversy surrounding its actions. I am also perplexed by the majority’s statement that if defendant was required to refund the wrongfully collected tax, “in one way or another, the refund payments would ultimately have to be covered by the taxpayers.” Ante at 51. How a reimbursement to taxpayers of wrongfully collected tax revenues can be char*93acterized as a burden on taxpayers is hard to fathom, just as is the concept implicit in that assertion that taxpayers are a fungible entity and that it does not much matter which taxpayers have paid the taxes and which have received the refunds.
Although defendant apparently refunded all the money collected for the last tax year voluntarily, even this step on the city’s part would not be required under the majority’s holding that the Supreme Court’s decision should be applied only prospectively.