¶1 Our state constitution limits municipal indebtedness to protect taxpayers from legislative and voter improvidence. We must decide whether the city of Wenatchee (City) would exceed its debt limit by entering into a “contingent loan agreement” (CLA) with appellant Greater Wenatchee Regional Events Center Public Facilities District (District) to help the District finance a regional events center. The District argues that the CLA is not subject to the City’s debt limit because it creates a “contingent” liability, triggered only if the District is unable to make payments on the District’s bonds. We reject this argument because the City is unconditionally obligated to service the District’s debt if the District cannot and because the risk of loss falls upon the City and its taxpayers. We conclude that this case implicates the very concerns that prompted our framers to enact limits on municipal debt in the first place. We hold that because the City’s obligation *792under the CLA is essentially a guaranty, it would create indebtedness within the meaning of our constitution. Accordingly, we affirm the trial court.
¶2 The City could enter into the CLA if approved by a vote of the people, but not without a popular vote. Total municipal debt incurred without a public vote is limited to one and one-half percent of the total assessed value of all taxable property within the City, while debt approved by 60 percent of the voters can be 5 percent of the total assessed value. Const, art. VIII, § 6. Our decision accordingly places the approval of the CLA in the hands of the voters.
FACTS
¶3 The District is a municipal corporation organized under chapter 35.57 RCW. The District was formed in June 2006 by an interlocal agreement among the City; Chelan and Douglas Counties; the cities of East Wenatchee, Cashmere, Chelan, Rock Island, and Entiat; and the town of Waterville. The purpose of the District is to finance, construct, and operate the Greater Wenatchee Regional Events Center (Regional Center), a 167,531 square foot facility that hosts concerts, trade shows, family shows, sporting events, rodeos, and other gatherings. Construction of the Regional Center began in September 2006 and was completed in November 2008.
¶4 The interlocal agreement creating the District provided mechanisms for financing the Regional Center, giving the District authority to impose various taxes. Further, in September 2006, the District and the City agreed to make CLAs in the future requiring the City to loan money to the District as needed to meet the District’s debt service obligations.
¶5 The District planned to finance the Regional Center using bonds and, in November 2008, issued short-term bond anticipation notes (2008 Notes) worth $41,770,000 to purchase the Regional Center. The 2008 Notes were intended *793as a temporary funding mechanism because of an unfavorable bond market in 2008. Payments on the notes were interest only and came due on December 1, 2011. The City entered into a CLA with the District, obligating the City to loan money to the District to make interest payments on the 2008 Notes in the event the District could not. The 2008 CLA is not at issue in this case, and no one contends it would violate the City’s debt limit.
¶6 In 2011, in anticipation of the 2008 Notes maturing, the District took steps to issue long-term bonds to retire the 2008 Notes. To support the issuance of these bonds, the District proposed another CLA to the City. It is this 2011 CLA that is at issue in this appeal.
¶7 The proposed 2011 CLA requires the City to loan money to the District if and when the District cannot make its semiannual debt service payments. The 2011 CLA also includes several other important provisions: that the District will repay all such loans, with interest, from the District tax and facility revenues; that the City’s commitment to make loans is absolute and unconditional; that all debts of the District are the District’s alone; that holders of the District’s bonds will have no recourse against the City, its assets, or its tax revenues; and that the City has no obligation to impose new taxes or enter into its own debt obligations to fund the loans to the District. The agreement also grants third-party beneficiary status to bondholders (in theory allowing them to compel the City to make loans) and contains no limitations on the amount of money the City could be required to loan the District (this represents a change from earlier agreements, which limited the City’s loan obligation to its debt capacity).
¶8 The City passed a resolution approving the 2011 CLA conditioned on the City’s obtaining a judicial declaration that it has the right and authority to do so.1 Accordingly, the *794City filed a complaint in Chelan County Superior Court seeking a declaratory judgment whether execution of the 2011 CLA would cause the City to exceed its debt limits. The District intervened, and the court appointed a taxpayer representative to represent the taxpayer’s interests.
¶9 The superior court granted summary judgment to the City, ruling that the 2011 CLA is “indebtedness” within the meaning of article VIII, section 6 of our constitution and therefore subject to constitutional and statutory debt limits. The court also ruled that the agreement would cause the City to exceed its nonvoted debt limit.2 Finally, the court held that the amount of indebtedness incurred by the City equaled the entire amount that could possibly be loaned to the District to meet all of its debt service obligations, including both principal and interest over the life of the bonds. Without the 2011 CLA to support the issuance of new long-term bonds, the District was unable to refinance the 2008 Notes before they became due and defaulted on the notes on December 1, 2011. We granted the District’s request for direct review.
ANALYSIS
I. The municipal debt limit under article VIII of our constitution
A. The text and purpose of article VIII, section 6
¶10 Article VIII, section 6 of our state’s constitution forbids municipalities from becoming “indebted in any manner” beyond one and a half percent of taxable property within their boundaries:
No county, city, town, school district, or other municipal corporation shall for any purpose become indebted in any *795manner to an amount exceeding one and one-half per centum of the taxable property in such county, city, town, school district, or other municipal corporation, without the assent of three-fifths of the voters therein voting at an election to be held for that purpose, nor in cases requiring such assent shall the total indebtedness at any time exceed five per centum on the value of the taxable property therein. . . .[3]
This provision complements article VIII, section 1, which limits state debt.4
¶11 Our framers enacted debt limitations to remedy a particular historical evil. In the 19th century, state governments financed or guaranteed an increasing number of private and public capital and infrastructure projects, most notably railroads. See Robert S. Amdursky & Clayton P. Gillette, Municipal Debt Finance Law; Theory and Practice § 4.1.1, at 162 (1992). Many of these projects failed, leaving taxpayers liable to pay for them while receiving little or nothing in return. Id. In response, states around the country enacted debt limitations preventing legislative bodies from saddling current and future taxpayers with an unmanageable tax burden to support unsuccessful railroads and other unwise ventures. Id.; see also Dep’t of Ecology v. State Fin. Comm., 116 Wn.2d 246, 257, 804 P.2d 1241 (1991) (“Consti*796tutional debt limitations were enacted to protect future taxpayers from the kind of improvidence that led to state and local government bankruptcies in the 19th century.”).
¶12 At Washington’s constitutional convention, our framers were appropriately concerned with the effects unlimited indebtedness would have on future prosperity, see The Journal of the Washington State Constitutional Convention 1889, at 667 (Beverly Paulik Rosenow ed., 1962), and enacted debt limits to cure these ills by building an “ ‘impassible barrier’ ” around the public treasury. State ex rel. Jones v. McGraw, 12 Wash. 541, 543, 41 P. 893 (1895); State ex rel. Potter v. King County, 45 Wash. 519, 528, 88 P. 935 (1907) (debt limits “are intended for the protection of minorities, for the protection of posterity, and to protect majorities against their own improvidence ...”).
¶13 The role of our judiciary in this scheme is self-evident: We must enforce the constitution. Potter, 45 Wash. at 528 (stating that enforcing debt limits is the “duty of the courts”). Constitutional debt limits are premised on the belief that political accountability does not sufficiently check runaway debt. See Amdursky & Gillette, supra, § 4.1.1, at 160-61. Thus, we must not assume legislative bodies will police themselves; instead, it is our duty to ensure that public entities do not make promises that they have no constitutional authority to honor.
B. The “risk of loss” concept
¶14 In carrying out our constitutional duty under article VIII, we have created a wide vocabulary of principles, concepts, and exceptions. For example, we have articulated the “special fund doctrine,”5 the concept of “borrowed money,”6 the “contingency” exception,7 and the concept of *797“full faith and credit,”8 among many others. As discussed later in this opinion, some of these principles are contradictory, leading to opposite conclusions. But the apparent contradictions can be resolved because a close examination of these concepts reveals a discernible uniformity.
¶15 Nearly all of our public debt doctrines and decisions can be explained by determining who bears the risk of loss in the underlying obligation. Nearly every time we have determined that “debt” exists, the obligation in question places the risk of project failure on the taxpayer (independent of the consideration received) rather than the creditor or bondholder. We have found debt to exist where, if the project fails, the general fund is exposed and the taxpayers are saddled with the repayment burden. See, e.g., State ex rel. Wash. State Fin. Comm. v. Martin, 62 Wn.2d 645, 663-64, 384 P.2d 833 (1963). Conversely, where the risk of project failure lies not with the taxpayer but with the creditor or bondholder, we have found that there is no debt. See, e.g., Dep’t of Ecology, 116 Wn.2d at 257-58.
¶16 The “special fund” cases demonstrate this principle well. In those cases, bonds are repaid from a special fund replenished with project revenues or other funding sources having some nexus to the project. For example, in Winston, the city of Spokane issued “obligations” to pay for a waterworks system, with the obligations to be repaid through a percentage of waterworks revenues. Winston v. City of Spokane, 12 Wash. 524, 525-26, 41 P. 888 (1895). No obligations were to be repaid from the general fund, so the investors bore the debt risk: if the project failed to produce enough revenue to service debt payments, the taxpayers were not liable for any shortfall. We held that this obligation did not create “indebtedness” within the meaning of article VIII, section 6. Winston, 12 Wash. at 527-28. On the other hand, in Martin, although similar facts existed (bonds were issued for the construction of public buildings), the *798bonds were to be repaid from an excise tax on the sale of cigarettes. 62 Wn.2d at 646-47. If the project failed, the bonds would be repaid through increased excise taxes, placing the risk of project failure on the taxpayers. We held that the arrangement created debt within the meaning of article VIII, section 1. Martin, 62 Wn.2d at 663-64.
¶17 Nearly every case stretching back to statehood is consistent with this “risk of loss” principle.9 See Amdursky & Gillette, supra, § 4.1.2, at 164-70 (surveying Washington case law in detail and concluding that our debt limits are triggered where the risk of project failure falls on the taxpayers/general fund independent of the consideration received for the bonds).
¶18 In our most recent cases, we have begun explicitly relying on the risk of loss concept as a basis for our decisions. For example, in Department of Ecology, the Department of Ecology entered into a complex lease arrangement that required payment only so long as the legislature appropriated money. 116 Wn.2d at 258. We concluded there was no debt because “[t]he ultimate risk of loss is not on the State’s future taxpayers. Instead, the risk of loss is on the [investors], who will have entered into the transaction with full knowledge that they alone bear that risk.” Id. at 254-55.
¶19 The dissent misses the point of the risk of loss analysis, labeling it “a matter left to elected representatives,” and accusing the lead opinion of “second-guessing.” Dissent at 819, 820. The risk of loss analysis does not attempt to evaluate the likelihood or the amount of risk. To the contrary, the question is simply this: on whom does the *799risk of loss fall, the investors or the public? This is not a speculative inquiry because it is evident on the face of the operative documents. In this case, the risk of loss is on the City and its taxpayers. If the revenues of the District are inadequate to repay the bondholders, the City must make loans to the District to permit the payments to be made. The City, not the bondholders, is at risk.
¶20 The risk of loss concept can guide our decisions in this and future cases. Nevertheless, we are mindful of the fact that existing case law addresses many of the special problems that arise in the context of public debt, and we must turn to that case law first and foremost.
C. The contingency cases and the guaranty cases
¶21 This case sits at the intersection of two conflicting lines of case authority — the “contingency” cases and the “guaranty” cases. To resolve this case, we must decide which of these lines of authority is correct.
¶22 In most states, it is widely accepted that so-called “contingent liabilities” are not debt. See 15 Eugene McQuillin, The Law of Municipal Corporations § 41:22, at 480-81 (3d rev. ed. 2005). As Professor McQuillin explains, one example of a contingent liability is a contract to pay for goods that is contingent on the goods actually being furnished. Id. For example, in Walla Walla City v. Walla Walla Water Co., the United States Supreme Court found that there was a contingent liability, not debt, where the city of Walla Walla contracted to pay for waterworks contingent on the waterworks being built and water being available. 172 U.S. 1, 19 S. Ct. 77, 43 L. Ed. 341 (1898).
¶23 In this “pay-as-you-go” situation, it makes sense to find that there is no debt. Taxpayers are simply not at risk of being saddled with debt while receiving little or nothing in return, since payment hinges on consideration received. And indeed, most other jurisdictions that apply the contin*800gency doctrine do so in similar pay-as-you-go scenarios.10 We agree that this kind of contingent liability is not debt.
¶24 But we took the contingency doctrine one step beyond the pay-as-you-go situation in Comfort v. City of Tacoma, 142 Wash. 249, 255-56, 252 P. 929 (1927). The city of Tacoma created a special local improvement guaranty fund to secure repayment of local improvement bonds. The primary sources of funding were the local improvement taxes secured by the bonds and interest on any bond sale proceeds not yet expended for the improvement. If the fund was insufficient to make scheduled bond repayments, the fund issued warrants to the bondholders. But the warrants were limited to five percent of the outstanding bond obligations secured by the fund. Id. at 255. In other words, the city did not guarantee individual bonds, but instead made payments to the fund, subject to the five percent limitation. We held that this was only a contingent liability and therefore not debt, citing City of Walla Walla.11 Comfort, 142 Wash. at 255-56. But we continued, stating in dicta, “If A is indebted to B and C promises that, if A does not pay B, then he (C) will, no one would contend that C had an outstanding debt.” Id. This dictum overlooks the fact that unlike the hypothetical C’s unlimited obligation, Tacoma’s obligation was capped at five percent of the outstanding debt.
¶25 Comfort's expansive interpretation of contingent debt is flatly contradicted by another line of authority, the *801“guaranty” cases.12 In State Capitol Commission v. State Board of Finance, we held that a state guaranty of bonds was debt. 74 Wash. 15, 26-27, 132 P. 861 (1913). The State guaranteed bonds issued by the State Capitol Commission to finance the construction of buildings on our state capitol grounds. The bonds were to be repaid from the sale of valuable land set aside for that purpose, so it was highly unlikely there would ever be a shortfall requiring the State to pay. Nevertheless, we held that the guaranty was debt because it violated “the spirit and the letter” of our constitutional debt limits, rejecting the idea that whether a liability is debt depends on how likely it is to come due. Id. at 27. Years later, in Martin, we reaffirmed this proposition, stating that a “mere guaranty of the principal and interest” of bonds is debt, citing State Capitol Commission. Martin, 62 Wn.2d at 654-55. Thus, State Capitol Commission and Martin say a guaranty of bonds is debt, while Comfort, in dicta, says it is not.
¶26 The risk of loss principle discussed above resolves this conflict of authority. Even if the municipality’s liability is contingent upon the failure of payment by an intervening agency such as the District, such a contingent liability is subject to the debt limit if the ultimate risk of loss falls upon the municipality. We hold that the guaranty cases are correct, and the dicta in Comfort misstates the law.13 If a municipality could guarantee debt with no debt limit consequences, even a small town could back an almost limitless number of third-party projects by pledging its credit in the *802event of default. In fact, we noted in Comfort that the legislature established the five percent cap in part to remedy the “the lack of necessary restrictions to prevent pyramiding assessments ...” 142 Wash. at 251. Some of those guaranties would eventually come due, requiring the municipality to resort to taxes to pay for failed projects, the very evil against which our debt limits protect. These guaranties would transparently evade our constitutional debt limits and would frustrate not only the risk of loss concept, but also the very purpose of having debt limits in the first place. To ignore this would be to abdicate our solemn responsibility under the constitution.
¶27 When a municipality makes an absolute guaranty of another entity’s debt, the resulting obligation is indebtedness within the meaning of article VIII, section 6, and cannot properly be called a contingent liability. Comfort may well have been correctly decided on its facts, but State Capitol Commission correctly sets forth the law of our state.
¶28 The dissent relies heavily on Comfort, arguing that the CLA “merely creates a contingent liability.” Dissent at 816 (emphasis omitted). The dissent overlooks that unlike Comfort, the City’s liability to the District is not capped at five percent of the outstanding amount of the bonds. Equally importantly, the dissent entirely ignores the guaranty cases, such as State Capitol Commission, which clearly say that a guaranty of future bond payments is a debt within the constitutional limits.14
*803II. The 2011 CLA would create “indebtedness” if executed
¶29 We turn now to the main subject of this appeal: whether the 2011 CLA between the City and the District would create indebtedness triggering the City’s debt limit under article VIII, section 6. In examining the 2011 CLA, we must look beyond the labels used by the parties and analyze the substance of the agreement.
¶30 Under the risk of loss approach, the 2011 CLA is plainly debt because the City’s taxpayers bear the risk of project failure. If the Regional Center were to fail (i.e., shut down and cease producing revenues), the City would be required to make loans to service the District’s bonds with no foreseeable means for the loans ever to be repaid. The obligation to make loans is absolute, so the City would have to come up with money either from the general fund or increased taxes, potentially endangering basic city services (e.g., police, fire, and sewer). On the other hand, the bondholders carry no risk of loss. Assuming the City fulfills its obligation, all debt service payments will be timely made. There can be no doubt that, under the risk of loss approach, the City’s obligation under the 2011 CLA constitutes debt for purposes of article VIII, section 6. This conclusion is confirmed by the case law.
A. The 2011 CLA is a guaranty of the District’s bonds
¶31 In substance, the 2011 CLA is a guaranty, pledging the City’s taxing power to service the District’s debt. The essence of the 2011 CLA is that if the District is unable to make debt service payments, the City will provide money for those payments. The City nominally makes “loans,” but this is just a label. In fact, the whole point of the CLA is to make the bonds more marketable by pledging the City’s full faith and credit to ensure timely repayment. This is no different from the situation in State Capitol Commission.
¶32 We can also look to the common law of guaranty contracts, which defines a “guaranty” as an
*804“undertaking or promise on the part of one person which is collateral to a primary or principal obligation on the part of another, and which binds the obligor to performance in the event of nonperformance by such other, the latter being bound to perform primarily.”
Robey v. Walton Lumber Co., 17 Wn.2d 242, 255, 135 P.2d 95 (1943) (quoting 24 Am. Jur. § 2, at 873-74). At common law, a guaranty can be either absolute or conditional. An “absolute guaranty” is “ ‘an unconditional undertaking on the part of the guarantor that the debtor will pay the debt or perform the obligation.’ ” Id. at 255-56 (quoting 24 Am. Jur. § 16, at 885). In contrast, a conditional guaranty involves a “ ‘condition to liability on the part of the guarantor,’ ” which is the “ ‘happening of some contingent event other than the default of the principal debtor or the performance of some act on the part of the obligee.’ ” Id. at 256 (quoting 24 Am. Jur. § 16, at 885).
¶33 Under this test, the 2011 CLA is unquestionably an absolute guaranty. The 2011 CLA obligates the City to perform (by making loans) in the event of nonperformance by the District. In substance, the City takes over the District’s obligation to meet immediate debt service payments if the District cannot. Moreover, the only contingency on which the City’s obligation depends is default or nonperformance by the District. Under this test or any other, the 2011 CLA is a guaranty of the District’s debt.
B. The District’s attempts to characterize the obligation otherwise are unpersuasive
¶34 The District argues that this is not like a guaranty — and that it is not debt at all — because the City is obligated only to loan money to the District, not to actually make the District’s payments. We flatly rejected this logic a century ago in State ex rel. State Capitol Commission v. Lister, 91 Wash. 9, 156 P. 858 (1916), and we reject it again now. In Lister, we held that loans were no different from payments for debt limit purposes. Id. at 16-17. The state *805made loans from a fund to the State Capitol Commission to pay interest on construction bonds. The loans were virtually certain to be repaid because the Commission had set aside land for sale, the proceeds of which would repay the loans. Nevertheless, we held that there was debt, rejecting the notion that labeling an obligation as a “loan” changes its character. Id. Years later, we reaffirmed this proposition in Martin, stating that “even a loan of the interest from general taxes, with a guaranty of repayment” is debt. Martin, 62 Wn.2d at 657. Here, there is not even a guaranty of repayment. In fact, from the record, it appears possible that the City’s “loans” might never be repaid. The District’s argument ignores the reality of the situation. An obligation to make a loan can constitute debt, particularly where it appears unlikely it will be repaid.
¶35 The District also argues that the 2011 CLA does not create debt because the City can meet its obligation through current-year tax revenues, citing State ex rel. Troy v. Yelle, 36 Wn.2d 192, 217 P.2d 337 (1950). This argument is a red herring. Under the District’s theory, no bonds would count as debt as long as the municipality could meet its yearly debt service obligations through current-year taxes. This cannot be, for it would exempt virtually all existing public debt from constitutional limits. Rather, the current-year taxes exception applies where the entire obligation can be discharged through current-year taxes, as in the case of warrants covering current-year expenses in Yelle. There is no chance of that happening here. This is not a short-term obligation the City can discharge with present funds, but a long-term obligation that will persist for years to come. The facts of Yelle bear this out. Yelle dealt principally with “general fund warrants,” a short-term financing mechanism to pay for current-biennium appropriations when the state treasury contained insufficient funds. 36 Wn.2d at 193. Current-biennium appropriations are a far cry from the obligation at issue here, which is a long-term obligation that could require yearly payments. Wittier is equally un*806helpful; there, the alleged debt was essentially an appropriation that was paid for through current-year taxes. State ex rel. Wittler v. Yelle, 65 Wn.2d 660, 661-62, 399 P.2d 319 (1965). Long-term obligations are different. The “current year expenses” exception cannot be construed to cover long-term obligations like the City’s obligation under the CLA. The dissent claims incorrectly that our application of the risk of loss principle “essentially converts the debts of the District into debts of the City,” arguing that this is inconsistent with a footnote in one of our prior cases stating that “the debts of one municipal corporation are not to be considered the debts of a separate municipal corporation.” Dissent at 815 (citing Pierce County v. State, 159 Wn.2d 16, 43 n.14, 148 P.3d 1002 (2006)). The dissent overlooks the dispositive difference between our prior footnote and this case: in Pierce County, the counties never agreed to provide money to remedy any shortfall if Sound Transit was unable to repay its bonds. 159 Wn.2d at 25 (“The Sound Transit bonds are payable from and secured solely by the pledge of Sound Transit’s [motor vehicle excise tax] and sales tax.”).
¶36 Our footnote in Pierce County, although inapplicable here, helps to demonstrate why the City has incurred indebtedness within the meaning of our constitution. Our Pierce County footnote responded to the argument that the creation of Sound Transit unconstitutionally “expanded the debt limit of the counties.” Id. at 43 n.14. As noted above, the argument failed because the counties did not undertake to assist Sound Transit in repaying the bonds. But if a city were permitted to enter into CLAs with multiple municipal corporations, the city’s potential liability could quickly exceed its debt limit, creating virtually unlimited liability. Our constitution’s framers never intended that cities would be allowed such an evasion, endangering the fiscal health of the city at the risk of its citizens.
¶37 Finally, the District argues that the CLA cannot be debt because it is not “borrowed money.” In essence, the District argues that in order for there to be debt, the City *807must actually be the issuer of bonds. This contradicts not only our case law, but also the plain language of article VIII, section 6. Both State Capitol Commission and Lister find state debt even where an entity other than the state issues bonds. State Capitol Comm’n, 74 Wash. at 26-27; Lister, 91 Wash. at 17. Moreover, while the District’s argument might make some logical sense under the state debt limit in article VIII, section 1, it makes none when we are dealing with a municipality under article VIII, section 6. Our municipal debt limit prohibits municipalities from becoming “indebted in any manner.”15 Wash. Const. art. VIII, § 6. This language is broader than article VIII, section 1, and by its terms encompasses more than the classic case of a municipality issuing its own bonds to finance a public project.
C. The 2011 CLA is indebtedness
¶38 Having rejected the District’s arguments, we hold that the 2011 CLA, if effective, would create debt. The arrangement is in substance a guaranty, and the risk of loss falls squarely on the City’s taxpayers: if the Regional Center fails, they, not the District’s bondholders, wind up holding the bag. We affirm the trial court.
III. The total amount of indebtedness would include the principal of the debt plus any accrued interest
¶39 The parties dispute the amount of indebtedness the City would incur by executing the 2011 CLA. Under our case law, the correct amount of debt is equal to the total amount of bonds the City would guarantee. State Capitol Comm’n, 74 Wash. at 24, 27. The trial court in this case followed this rule, finding that the entire amount of principal on the District’s bonds would be City debt. However, the trial court erred by including the total amount of *808interest over the life of the bonds in its calculation. Only accrued interest is included in a municipality’s debt limit, not interest still to be accrued. Lister, 91 Wash. at 15. This being said, if the principal debt, plus any other debt subject to the limit, exceeds the constitutional debt limit, the interest issue may be moot. But in the event that either party believes further proceedings on this issue are necessary to resolve the case, the trial court can address those issues on remand consistently with this opinion.
IV. The trial court did not commit prejudicial error in considering the City’s declarations and the exhibits attached to them
¶40 Finally, the District raises several evidentiary objections to declarations and exhibits brought before the trial court. A trial court’s ruling on an evidentiary issue is “harmless unless it was reasonably probable that it changed the outcome” of the case. Brundridge v. Fluor Fed. Servs., Inc., 164 Wn.2d 432, 452, 191 P.3d 879 (2008). Here, the trial court, confronted with the District’s evidentiary objections, said that “the case will not turn on that, of course.” Verbatim Report of Proceedings at 51. The District’s attorney agreed, at least in part, that “it’s not something that I think this matter ought to be turning on one way or another.” Id. After reviewing the record, we agree. If the trial judge committed evidentiary error here, it was harmless.
V. The recent passage of Substitute Senate Bill 5984 does not moot this case
¶41 After we heard argument in this case, the legislature passed a law on March 1,2012, allowing the City to impose a sales tax to pay for the Regional Center without holding an election. See Laws of 2012, ch. 4, § 6. That bill, Substitute S.B. 5984, does not moot our decision in this case. A case is moot if we can no longer provide effective relief. Westerman v. Cary, 125 Wn.2d 277, 286, 892 P.2d *8091067 (1994) (citing Orwick v. City of Seattle, 103 Wn.2d 249, 253, 692 P.2d 793 (1984)). Here, our ability to provide effective relief is not impacted by a slight modification to the City’s taxing authority. This is a declaratory judgment action asking whether it would violate the City’s debt limit to enter into a CLA with the District. That question still urgently requires an answer, and we have received nothing suggesting the City does not still want an answer to it. This case does not turn on whether the City can impose a tax without putting it to a vote. In fact, the City’s capacity to generate revenue to pay for an obligation is completely independent from whether that obligation implicates the City’s debt limit. Municipal debt limits are calculated as a percentage of the value of taxable property within the municipality, and sales taxes simply do not enter into the equation. See Const. art. VIII, § 6; RCW 39.36.020(1), (2)(a)(ii).
¶42 Nor does it moot the case that the District recently passed a ballot measure imposing a sales and use tax within its jurisdiction. See Michelle McNiel, Sighs of Relief after Voters Overwhelmingly Approve Sales Tax Increase for Town Toyota Center, Wenatchee World, Apr. 17,2012. Again, the question we must answer is whether entering into a CLA implicates the City’s debt limit. Our ability to provide effective relief by answering this question is not affected by a slight change in the District’s fiscal situation, particularly where we have no evidence that this change will solve the District’s financial woes or eliminate the need for long-term bonds. Our situation with respect to this case remains the same: We have been called on to answer a constitutional question, and that question still urgently requires an answer. Moreover, on appeal we review the facts in the record before us. It is not our role to supervise the specifics of the District’s finances, particularly where recent changes to those finances are not detailed in the record. This case is not moot.
*810CONCLUSION
¶43 We have a duty under the constitution to enforce limitations on public indebtedness. This case implicates the core concerns of article VIII, section 6. We cannot sit idly by while municipalities creatively attempt to exceed their proper debt limits, frustrating the principles enshrined in our constitution. We affirm the trial court’s ruling and remand to the trial court for further proceedings, if necessary, consistent with this opinion.
C. Johnson, J.M. Johnson, and González, JJ., concur. Stephens, J., concurs in the result only.Chapter 7.25 RCW provides a procedure for obtaining declaratory judgments for the validity of local bond issues.
Article VIII, section 6 differentiates between “voted” indebtedness taken on with the assent of three-fifths of the voting populace and “nonvoted” indebtedness taken on without voter assent. Voted debt has a higher limit — five percent compared with one and one-half percent for nonvoted debt. None of the indebtedness in this case is voted.
ROW 39.36.020(1) and (2)(a)(ii) contain a debt limit that is identical in substance to the constitutional limitation. See Dep’t of Ecology v. State Fin. Comm., 116 Wn.2d 246, 253 n.7, 804 P.2d 1241 (1991).
Article VIII, section 1 forbids the State from contracting debt in an amount for which payments of principal and interest in any fiscal year would equal more than nine percent of a three-year average of state revenues. Article VIII, section 1 also defines “debt” for purposes of the state debt limit:
(d) In computing the amount required for payment of principal and interest on outstanding debt under this section, debt shall be construed to mean borrowed money represented by bonds, notes, or other evidences of indebtedness which are secured by the full faith and credit of the state or are required to be repaid, directly or indirectly, from general state revenues and which are incurred by the state. . . .
The dissent goes astray from the very beginning, dissent at 812-13, equating the term “become indebted in any manner” in article VIII, section 6 limitation on municipalities with the definition of “debt” under article VIII, section 1(d). Section 1(d) applies to the state, not municipalities, and by its terms applies only for the purpose of computing the total amount of the debt, which is not the issue here.
See State ex rel. Wash. State Fin. Comm. v. Martin, 62 Wn.2d 645, 653-54, 384 P.2d 833 (1963).
See State ex rel. Troy v. Yelle, 36 Wn.2d 192, 195, 217 P.2d 337 (1950).
See Comfort v. City of Tacoma, 142 Wash. 249, 255-57, 252 P. 929 (1927).
See Dep’t of Ecology, 116 Wn.2d at 254, 258.
The dissent wrongly suggests that risk of loss is an “entirely new legal analysis.” Dissent at 810. The principle is recognized both in our case law and by scholars and is consistent with nearly all of our public debt cases dating back to early statehood. It has developed haphazardly, but organically, as courts have resolved individual cases on their merits. Recognizing this pattern, we should state it forthrightly and rely on it — not only as a matter of honest jurisprudence but also to give clear guidance to municipal officers, the public, and lower courts. This will help prevent future invalid bond issuances by promoting understanding and encouraging public scrutiny.
See Taxpayers for Improving Pub. Safety v. Schwarzenegger, 172 Cal. App. 4th 749, 762-63, 91 Cal. Rptr. 3d 370 (2009) (holding that the contingency exception applies where a “governmental entity agrees to pay sums in succeeding periods in exchange for property, goods, or services to be provided during those periods”); Knowlton v. Ripley County Mem’l Hosp., 743 S.W.2d 132, 136-37 (Mo. App. 1988) (holding that contingency exception applies to employment contract contingent on services performed, just like payment of a hydrant rental, maintenance of a public market, and payment for water as delivered).
We later confirmed the result in Comfort in Kelly v. City of Sunnyside, 168 Wash. 95, 11 P.2d 230 (1932).
The dissent does not acknowledge these conflicting lines of authority. But where cases conflict, the responsibility of the court is to harmonize them or overrule one line or the other, not to simply ignore the conflict.
Nor does it make sense, as amici suggest, to decide whether a liability counts as debt based on how likely it is to occur. This creates two major problems: first, it is difficult to imagine how a court would make that factual finding. Second, it would potentially make debt a moving target, hinging a liability’s status as debt on changing facts. It is far more sensible to look at what kind of contingency is at issue; i.e., whether it is similar to the pay-as-you-go fact pattern in Walla Walla City where the contingency is receipt of consideration, not some outside event or market condition. See 172 U.S. at 19-20.
The dissent misplaces its reliance on Twichell v. City of Seattle, 106 Wash. 32, 179 P. 127 (1919), dissent at 816, in which the City pledged the revenue of its street car system to fund repayment of a bond issue. Bonds paid exclusively from the operational revenue of a utility such as a street car system do not come within the debt limit because they do not require repayment from general revenues. Nor does the dissent gain any support from citing a case presenting the identical issue as Comfort: establishing a guaranty fund with a five percent cap. Dissent at 816 (citing Kelly v. City of Sunnyside, 168 Wash. 95, 97, 11 P.2d 230 (1932)). Finally, the dissent is unsupported by its citation to State ex rel. Washington Toll Bridge Authority v. Yelle, 56 Wn.2d 86, 94, 351 P.2d 493 (1960), dissent at 818, which turned on statutory interpretation, not the constitutional debt limit.
In contrast, article VIII, section 1(a), (d) says that the state may “contract debt” and that “under this section, debt shall be construed to mean borrowed money. . . .” By its own terms, this definition applies to section 1 (“this section”), not to the article VIII, section 6 limitation on municipal indebtedness.