¶53 (dissenting) — When citizens legislate through the initiative process, they are held to the same standards as our elected lawmakers. When corporations legislate by initiative, we must also hold them to those same standards.
¶54 Article II, section 19 of our state constitution declares:
No bill shall embrace more than one subject, and that shall be expressed in the title.
This provision has been part of the basic framework of our state’s government for over 150 years, appearing not only in our state constitution at its original adoption but in the Organic Act of 1853 establishing Washington as a territory.6 Article II, section 19 is a cornerstone of good government, ensuring that our lawmakers legislate honestly. Article II, section 19’s single-subject and subject-in-title rules require our elected legislators to enact laws with forthrightness and clarity. We demand the same of initiatives.
¶55 Initiative Measure 1183 (1-1183), which privatizes our state liquor industry, allows hard liquor to be sold at grocery stores and other retail establishments, and dramatically changes state regulation of liquor distribution and sales. But 1-1183 also imposes new taxes and contains a $10 million earmark7 that has nothing to do with liquor. An initiative can impose new taxes, but the ballot title cannot misleadingly imply that it does not. Likewise, earmarking a portion of the new tax revenue for public safety is not inherently problematic, but article II, section 19 *667precludes combining a substantive liquor privatization law with an earmark that has no rational relation to liquor privatization and may have been included only to win votes. I respectfully dissent.
I. 1-1183 violates the subject-in-title rule because it fails to properly notify voters that it raises taxes
¶56 Under article II, section 19’s subject-in-title rule, a law is unconstitutional if its title fails to put the public, legislators, or voters on notice of its contents. Power, Inc. v. Huntley, 39 Wn.2d 191, 198, 235 P.2d 173 (1951). In essence, the rule requires legislators to be honest about what is contained in a law. Seymour v. City of Tacoma, 6 Wash. 138, 148-49, 32 P. 1077 (1893) (“The object of the requirement that the subject of an act shall be expressed in its title is, that no person may be deceived as to what matters are being legislated upon.”); Howlett v. Cheetham, 17 Wash. 626, 635, 50 P. 522 (1897) (“[A] title which is misleading and false is not constitutionally framed.”); State ex rel. Wash. Toll Bridge Auth. v. Yelle, 32 Wn.2d 13, 24-25, 200 P.2d 467 (1948) (one purpose of the subject-in-title rule is to prevent surprise or fraud). A law is unconstitutional if its title fails to “ ‘give[ ] notice that would lead to an inquiry into the body of the act, or indicate to an inquiring mind the scope and purpose of the law.’ ” Wash. Fed’n of State Emps. v. State, 127 Wn.2d 544, 555, 901 P.2d 1028 (1995) (quoting Young Men’s Christian Ass’n v. State, 62 Wn.2d 504, 506, 383 P.2d 497 (1963)). The rationale behind this rule is obvious: when laws are passed, people should know what is in them, especially those voting on the laws. Amalgamated Transit Union Local 587 v. State, 142 Wn.2d 183, 207, 11 P.3d 762, 27 P.3d 608 (2000). No one should be unpleasantly surprised to learn after voting for a law that it includes provisions one opposes. The rule is particularly important for initiatives because “often voters will not reach the text of a measure or the explanatory statement, but may instead cast their votes based upon the ballot title.” Id. at 217.
*668¶57 The title of 1-1183 fails to notify the public that it imposes taxes, instead misleadingly implying that it does not. Our subject-in-title rule condemns such a classic bait- and-switch, rendering 1-1183 unconstitutional. This case is controlled by Amalgamated Transit, in which the title of the initiative referred to “taxes” but the substantive provisions of the initiative defined “taxes” far more broadly than the public understanding of the word, violating the subject-in-title rule. Here, the title of 1-1183 refers to “license fees” but then imposes “license fees” that are actually taxes and would be understood by the public to be taxes. Just as in Amalgamated Transit, the title violates the subject-in-title rule of article II, section 19.
a. I-1183’s ballot title implies that it imposes only “license fees based on sales,” not taxes
¶58 When we analyze initiatives under the subject-in-title rule, we look to the ballot title. Wash. Fed’n, 127 Wn.2d at 555. By statute, the ballot title consists of a statement of the subject of the measure, a concise description of the measure, and the question of whether or not the measure should be enacted into law. RCW 29A.72.050. We treat the whole ballot title as the initiative’s “title.” Amalgamated Transit, 142 Wn.2d at 211-12.
¶59 I-1183’s ballot title reads as follows:
Initiative Measure
1183
Proposed by initiative petition:
Initiative Measure No. 1183 concerns liquor: beer, wine, and spirits (hard liquor).
This measure would close state liquor stores and sell their assets; license private parties to sell and distribute spirits; set license fees based on sales; regulate licensees; and change regulation of wine distribution.
Should this measure be enacted into law?
*669[] Yes
[] No
State of Washington Voters’ Pamphlet, General Election 19 (Nov. 8, 2011).
¶60 This ballot title is consistent with I-1183’s text, if not its substance. Section 103 of the initiative imposes so-called “license fees” on hard liquor retailers:
(4) Each spirits retail licensee must pay to the board, for deposit into the liquor revolving fund, a license issuance fee equivalent to seventeen percent of all spirits sales revenues under the license ....
(5) In addition to the payment required under subsection (4) of this section, each licensee must pay an annual license renewal fee of one hundred sixty-six dollars. . . .
Laws of 2012, ch. 2, § 103. Similarly, section 105 imposes license fees on hard liquor distributors':
(3) (a) . . . each spirits distributor licensee must pay to the board for deposit into the liquor revolving fund, a license issuance fee calculated as follows:
(i) In each of the first two years of licensure, ten percent of the total revenue from all the licensee’s sales of spirits made during the year for which the fee is due, respectively; and
(ii) In the third year of licensure and each year thereafter, five percent of the total revenue from all the licensee’s sales of spirits made during the year for which the fee is due, respectively.
(4) In addition to the payment set forth in subsection (3) of this section, each spirits distributor licensee renewing its annual license must pay an annual license renewal fee of one thousand three hundred twenty dollars for each licensed location.
Finally section 301 of the initiative states that these license fees are not taxes:
This act does not increase any tax, create any new tax, or eliminate any tax.
*670Thus, the title of 1-1183 is consistent with the text of 1-1183, since the initiative claims to impose only “license fees.” But the title is not consistent with the substance of 1-1183.
b. In substance, 1-1183 imposes new taxes, not just fees
¶61 To determine the meanings of “license fee” and “tax” for a subject-in-title challenge, we look to their common and ordinary meanings. Wash. State Grange v. Locke, 153 Wn.2d 475, 495, 105 P.3d 9 (2005). We determined the common and ordinary meaning of “tax” in Amalgamated Transit: it is a “ ‘pecuniary charge imposed by legislative or other public authority upon persons or property for public purposes : a forced contribution of wealth to meet the public needs of a government.’ Webster’s Third New International Dictionary 2345 (1986)” Amalgamated Transit, 142 Wn.2d at 219-20. A “fee” is “a charge fixed by law or by an institution . . . for certain privileges or services.” Webster’s Third New International Dictionary 833 (2002). But a license fee also connotes something different from a tax; the money collected for a fee does not go toward public needs, but toward the cost of regulation or it is actually a key part of the regulation. Franks & Son, Inc. v. State, 136 Wn.2d 737, 750, 966 P.2d 1232 (1998). In other words, the common and ordinary understanding is that a tax supplements the public treasury, while a fee circles back to the regulator to cover the cost of regulation. Id.; Amalgamated Transit, 142 Wn.2d at 220-21.
¶62 We have explored this common and ordinary meaning in more detail in our case law, specifically articulating the difference between a tax and a regulatory fee. We have said that a tax is a levy made for the purpose of raising revenue for a general governmental purpose, while a regulatory fee is enacted principally as an integral part of the regulation of an activity or to cover the cost of regulation. See Franks & Son, 136 Wn.2d at 750. To determine whether a charge is a tax or a fee, we consider three factors:
“[W]hether the primary purpose ... is to accomplish desired public benefits which cost money, or whether the primary *671purpose is to regulate” [;] [W]hether the money collected must be allocated only to the authorized regulatory purpose [; and] [W]hether there is a direct relationship between the fee charged and the service received ... or between the fee charged and the burden produced by the fee payer.
Covell v. City of Seattle, 127 Wn.2d 874, 879, 905 P.2d 324 (1995) (quoting Hillis Homes, Inc. v. Snohomish County, 97 Wn.2d 804, 809, 650 P.2d 193 (1982)).8 These factors reflect commonsense concepts. Despite respondents’ contentions, the Covell test is not a technical, legalistic definition unknown to the common voter. It reflects commonly held understandings, as we held in Amalgamated Transit. 142 Wn.2d at 220-21, 226. The test is essentially a rigorous way to flesh out a difference that we have previously held the average voter understands. Id.
¶63 Applying the Covell factors to this case leaves no doubt that the 1-1183 “license issuance fees” are taxes. All three factors point to a tax, leaving little room for serious debate. First, the purpose of the “license issuance fees” is not to regulate liquor at all, nor is it to provide services. The purpose is to replace and supplement the revenue stream that previously came from profits on the sale of liquor. That revenue stream is not dedicated solely to alcohol regulation — in fact, much of it goes straight to the public treasury, either to the general fund or to cities and counties. See RCW 66.08.190; Laws of 2012, ch. 2, § 101 (2)(k) (the initiative will “[m]aintain the current distribution of liquor revenues to local governments . . .”). Under CovelVs second factor, to be considered a fee instead of a tax, it is “ ‘essential’ that the money collected be segregated and ‘allocated only to the authorized regulatory purpose.’ ” Samis Land Co. v. City of *672Soap Lake, 143 Wn.2d 798, 809-10, 23 P.3d 477 (2001) (footnote omitted) (quoting CoveLl, 127 Wn.2d at 879). Here, there is no requirement that the money collected be allocated to regulating liquor or even to alcohol-related purposes. Instead, the money is funneled to a variety of sources, with much of it going to local government programs and the general fund. See Laws or 2012, ch. 2, § 102. Finally, under the third factor, there is no relationship between the fee charged and either services provided or regulatory burdens produced by the fee payer. The fee increases with sales even though increased sales do not create equivalent increases in services or burdens.
¶64 Analysis of these factors settles the issue: the so-called “license fees” are really taxes. This conclusion is confirmed by the initiative itself, which includes flat yearly “license renewal fees,” which likely cover the actual regulatory burden of issuing a license. These are the real license fees. The percentage-based charges are essentially sales taxes imposed on the seller.
¶65 During Abraham Lincoln’s trial lawyer career, Lincoln is said to have cross-examined a witness as follows:
“How many legs does a horse have?”
“Four,” said the witness.
“Right,” said Abe.
“Now, if you call the tail a leg, how many legs does a horse have?”
“Five,” answered the witness.
“Nope,” said Abe, “callin’ a tail a leg don’t make it a leg.”
Lamon v. McDonnell Douglas Corp., 19 Wn. App. 515, 534-35, 576 P.2d 426 (1978) (Andersen, J., dissenting). Calling a tax a license fee does not make it a license fee.
c. 1-1183 violates the subject-in-title rule and is unconstitutional because its title misleadingly implies that it imposes only “license fees,” not taxes
¶66 A ballot title that implies something untrue fails to put the public on notice of what is contained in the *673initiative. Seymour, 6 Wash. at 148-49; Howlett, 17 Wash. at 635. The core reason for including article II, section 19 in our constitution was to ensure fairness, forthrightness, and clarity in the legislative process. 1-1183 fails this purpose because it misleads voters, falsely suggesting it contains no new taxes when in reality it does.
¶67 This case is controlled by Amalgamated Transit. There, we found a subject-in-title violation in 1-695 because the title referred to “taxes,” but in the text of the act, “taxes” was defined to include not just taxes but also license fees and other charges. Amalgamated Transit, 142 Wn.2d at 220-21. 1-1183 has the same defect. The title of 1-1183 refers to “license fees,” but those “license fees” include not only bona fide “license renewal fees,” but also “license issuance fees,” which are in fact taxes. Thus, the common meaning of “license fees” is broadened in the same way “taxes” was broadened in Amalgamated Transit, resulting in the same failure of notice. Id. Moreover, the drafters of the initiative in Amalgamated Transit likely used the word “taxes” for the same reason (voters will vote to limit them) that the drafters of 1-1183 avoided it (voters will not vote to raise them). Since this case follows the precise logical pattern as Amalgamated Transit, we are bound by our prior reasoning and cannot ignore it. When reviewing initiatives and recent legislation, perhaps more than in any other context, it is crucial that we adhere to our past precedents to avoid any appearance that we are just voting our preferences to undo results achieved at the ballot box or in the halls of our legislature.
¶68 The majority reasons that the ballot title’s reference to “ ‘license fees based on sales’ ” puts inquiring minds on notice of what the bill does — that it raises taxes. Majority at 663-65. In essence, this argument assumes that the average voter simply does not know or care whether a given charge is a tax or a fee. We held to the contrary in Amalgamated Transit that the average voter considers taxes and fees to be two different things. 142 Wn.2d at 220-21, 226. We specifi*674cally stated that “[t]he average informed voter would not conclude that the charge for a state’s nurse’s license, for example, is a ‘tax’ in its traditional sense.” Id. at 221. Under Amalgamated Transit, the title of 1-1183 put the voter on notice that it imposed license fees; it did not put the voter on notice of the imposition of a new tax.
¶69 Common sense confirms this distinction. Voters do not like taxes. See Joshua D. Rosenberg, The Psychology of Taxes: Why They Drive Us Crazy, and How Can We Make Them Sane, 16 Va. Tax Rev. 155, 157 (1996) (“Among all kinds of legislative and judicial decision making, only tax laws seem capable of engendering nearly universal anger, anxiety, paranoia and outright hatred . . . .”). If we were to hold that license fees and taxes are basically the same to an ordinary voter, we would be assuming that the average voter is either mistaken about, uninformed of, or unable to appreciate the difference between a tax and a fee. This is not only wrong, it is contrary to our case law. Amalgamated Transit, 142 Wn.2d at 220-21.
¶70 It could be argued (although certainly none of the parties take this position)9 that all our constitution requires is for the ballot title to match the text of the initiative, not for the ballot title to match the substance of the initiative. Under this view, article II, section 19 is a procedural technicality, not a substantive good government provision that ensures honest and forthright legislation. Article II, section 19 is not merely a tidy drafting provision; it is “the most salutary provision in our state constitution.” State ex rel. Arnold v. Mitchell, 55 Wash. 513, 516, 104 P. 791 (1909).
*675¶71 We should not create a loophole that would invite the drafters of initiatives to mislead the public. Just as it is misleading to call a liquor tax a “license fee,” the next law could raise the business and occupation tax, calling it a “business regulations fee,” or the sales tax, calling it a “transactional license fee.” A law is not unconstitutional merely because it raises taxes without expressly saying so in the title. To be sure, plenty of broadly titled laws raise taxes to achieve legislative goals without the title stating, “This bill raises taxes.” But a title cannot contain false or misleading information. If a title affirmatively implies that it does not raise taxes whereas the legislation in reality raises taxes, it violates article II, section 19’s subject-in-title rule.
II. 1-1183 also violates the single-subject rule because it contains a $10 million public expenditure earmark that has nothing to do with liquor
¶72 The Washington State Constitution does not allow lawmakers to cobble together unrelated pieces of legislation to garner votes. Nor does it allow legislators to attach unpopular laws to popular laws in order to gain approval for legislation that would not otherwise pass, a practice commonly known as “logrolling.”10 Wash. Fed’n, 127 Wn.2d at 552; Amalgamated Transit, 142 Wn.2d at 212.
¶73 The single-subject rule was included in our constitution to prevent logrolling. Wash. Fed’n, 127 Wn.2d at 552. By requiring bills to contain only one subject, our constitution ensures that every bill passes or fails on its own merits, not because a bill drafter was able to creatively ice the legislative cake with a popular provision. See, e.g., Wash. Toll Bridge Auth. v. State, 49 Wn.2d 520, 525, 304 P.2d 676 *676(1956). Like the subject-in-title rule, the single-subject rule is a critical safeguard of honest governance.
¶74 I agree with the majority that most provisions in 1-1183 do not violate the single-subject rule. But I depart from the majority because 1-1183 violates the single-subject rule in that the $10 million public safety earmark has no rational unity with liquor privatization.
a. 1-1183 contains a $10 million public expenditure earmark
¶75 Section 302 of 1-1183 does two things: (1) it preserves existing fund distributions to local governments under prior liquor laws and (2) it guarantees local governments an additional $10 million under the new “license fee” revenue stream for “the purpose of enhancing public safety programs”:
The distribution of spirits license fees under sections 103 and 105 of this act through the liquor revolving fund to border areas, counties, cities, towns, and the municipal research center must be made in a manner that provides that each category of recipients receive, in the aggregate, no less than it received from the liquor revolving fund during comparable periods prior to the effective date of this section. An additional distribution of ten million dollars per year from the spirits license fees must be provided to border areas, counties, cities, and towns through the liquor revolving fund for the purpose of enhancing public safety programs.
Laws of 2012, ch. 2, § 302. Although the $10 million set-aside must be spent on “public safety programs,” neither party disputes that there is no requirement the money be spent on anything directly related to alcohol. Further, the money is not put toward carrying out any other part of the initiative. Instead, it goes straight to local government treasuries.
*677b. The $10 million earmark has no rational unity with the main subject of 1-1183 (liquor) or to other provisions contained therein
|76 We begin our analysis of the single-subject rule by determining whether the title of the bill is general or restrictive. Where, as here, the bill has a general title,11 we look to see whether there is “rational unity” between the general subject and the incidental subdivisions. State v. Grisby, 97 Wn.2d 493, 498, 647 P.2d 6 (1982). Rational unity exists where all subjects contained in the bill are (1) germane to the general subject and (2) germane to one another. City of Burien v. Kiga, 144 Wn.2d 819, 826, 31 P.3d 659 (2001). The single-subject rule does not, “ ‘by restricting the contents of an “act” to one subject, contemplate a metaphysical singleness of an idea or thing, but rather that there must be some rational unity between the matters embraced in the act, the unity being found in the general purpose of the act and the practical problems of efficient administration.’ ” State ex rel. Wash. Toll Bridge Auth. v. Yelle, 61 Wn.2d 28, 33, 377 P.2d 466 (1962) (quoting State ex rel. Test v. Steinwedel, 203 Ind. 457, 468, 180 N.E. 865 (1932)).
¶77 On its face, the earmark for public safety has no relationship to liquor (or the practical problems of efficiently administering 1-1183) other than that the money paying for it comes from a liquor tax. This alone is not enough. The overriding purpose of the single-subject rule is to prevent logrolling, i.e., cobbling together unrelated legislation to garner votes. Wash. Fed’n, 127 Wn.2d at 552. Earmarking is a classic form of logrolling. In many ways, it is the most pernicious form of all because it calls to mind explicit vote-buying, securing support for legislation by funding pet projects or otherwise catering to the desires of undecided voters.
*678¶78 We will allow earmarks where they are sufficiently related to the topic of the underlying legislation. See, e.g., Wash. Ass’n of Neighborhood Stores v. State, 149 Wn.2d 359, 371, 70 P.3d 920 (2003) (upholding an earmark for tobacco-related programs only because the earmark related to the subject of the initiative: tobacco); Yelle, 61 Wn.2d at 38 (upholding an earmark for highway funds only because it related to the subject of the initiative, toll bridges and ferries — which were part of the highway system). This strikes a fair balance between curtailing logrolling on the one hand and allowing lawmakers a free hand to legislate on the other. This approach makes sense because the more an earmark has to do with the subject of a bill, the less likely it is that it was inserted solely to garner votes, and vice versa. However, we have never explicitly held that it violates the single-subject rule to combine substantive law with a wholly unrelated earmark.12 We should do so now. This is a fair and sensible rule supported by our case law that balances the competing concerns at stake and fulfills the purpose of article II, section 19.
¶79 Respondents make several unconvincing arguments that the public safety earmark is in fact related to liquor. First, respondents argue that “public safety” is rationally connected to liquor. On a theoretical level, this is undoubtedly true. The consumption of liquor increases public safety risks; drunk driving and alcoholism come quickly to mind. Thus, the State contends that funneling liquor tax money into public safety ameliorates the damage done by the initiative or is a hedge against the increased public safety risk that would come along with higher liquor consumption and easier access to liquor. If this were the case, there would certainly be rational unity between the earmark and the rest of the initiative.
*679¶80 The State’s argument is flawed because the proponents of 1-1183 never acknowledged that privatizing liquor would lead to increased consumption or public safety risk. Those concerns were raised by the initiative’s opponents. More importantly, the $10 million earmark simply does not hedge against alcohol-related risk. It could have done so — if the initiative had devoted $10 million to alcohol-related programs, there would be a clear hedge against risk and a rational unity between the substantive law and the earmark. Instead, the initiative allows money to be allocated to any local government public safety program, potentially including programs as unrelated to alcohol as tsunami warning systems, earthquake preparedness, or building code compliance. The entire range of public safety programs cannot possibly have rational unity with 1-1183. When the connection between an earmark and substantive law is so remote, the danger of logrolling is high, and such a connection cannot meet our constitution’s requirement that earmarks be related to the underlying substantive law.13
¶81 Second, respondents make an “historical practice” argument that liquor money has historically been devoted to local government public safety programs, creating rational unity between liquor and these programs. Thus, the earmark is merely a continuation of historical practice. There does appear to be some support in our case law for the idea that historical practice can inform our rational unity analysis, albeit in concurrences and dissents rather than in majority opinions. See, e.g., Wash. Fed’n, 127 Wn.2d *680at 575 (Talmadge, J., concurring in part/dissenting in part). If historical practice were relevant to determining rational unity, and if the $10 million earmark simply replaced a preexisting liquor revenue stream with the earmark, one might argue that the earmark satisfies the rational unity test. See Ass’n of Neighborhood Stores, 149 Wn.2d at 370-71. But 1-1183 instructs the State how it must spend new revenues that did not exist before. Thus, the earmark is not a subject that has historically been combined with liquor because the revenue it appropriates never existed prior to the passage of the initiative. Instead, the law designates a new distribution of funds going forward that does not reflect historical practice — in effect a reallocation of legislative priorities rather than a proportional slice of a larger pie.
¶82 Heavy reliance on historical practice also misses the point of the single-subject rule. The point is not that liquor money can never be spent on public safety programs unrelated to liquor. Of course it can. The single-subject rule imposes no limitations on how tax money can be spent. The point is that every law must rise or fall on its own merits. A $10 million earmark to local governments belongs in a different bill than a liquor privatization law, and if they had been enacted separately, there would be no violation of the single-subject rule.
f 83 Our responsibility is not to continue past historical practices but to determine the constitutionality of 1-1183 as passed. For all the reasons stated here, 1-1183 fails to satisfy our constitution. In any event the history of liquor privatization initiatives suggests a particularly high danger of logrolling. The year before 1-1183 passed, two liquor privatization initiatives were rejected by the voters. Following their initial defeat, the initiative’s proponents met with various stakeholders and added several provisions to the law, most notably the $10 million earmark for local government programs. The second time around, the bill passed. We cannot know for certain whether the earmark was added to *6811-1183 only to garner votes or whether the drafters legitimately sought to address public safety concerns. But the fact remains that the initiative did not pass until the earmark was added.
III. The constitutional defects in 1-1183 are not sever-able, and the law should be struck down in its entirety
¶84 A legislative act is unconstitutional in its entirety if the invalid provisions cannot be severed, meaning that (1) it cannot reasonably be believed that the act would have passed without the invalid portions or (2) elimination of the invalid portion would render the remaining part useless to accomplish the legislative purpose. Amalgamated Transit, 142 Wn.2d at 227-28. While a severability clause may sometimes “provide the assurance that the legislative body would have enacted remaining sections even if others are found invalid,” id. at 228, the existence of a severability clause is not dispositive of the issue. Leonard v. City of Spokane, 127 Wn.2d 194, 201, 897 P.2d 358 (1995).
¶85 I-1183’s severability clause cannot save the initiative from complete invalidation. Because the earmark provision is not rationally related to the subject of privatizing liquor sales, it violates the single-subject provision. We have held that an initiative that violates the single-subject rule must be voided in its entirety. Kiga, 144 Wn.2d at 825. “When an initiative embodies two unrelated subjects, it is impossible for the court to assess whether either subject would have received majority support if voted on separately. Consequently, the entire initiative must be voided.” Id.
¶86 But even if we were concerned only with the subject-in-title violation, it still does not follow that a majority of voters would have voted for the initiative without the new taxes. The State admits that the new taxes were added to replace the revenue stream provided by liquor sales. The legislative purpose of reforming liquor distribution would not be practicable without replacing that revenue stream, *682nor would the initiative likely have garnered popular support if it left a giant hole in the State’s finances. Consequently, the constitutionally infirm portions of 1-1183 cannot be severed from the valid portions, and 1-1183 should be struck down in its entirety.
CONCLUSION
¶87 This case demonstrates precisely why our founding fathers included article II, section 19 in our constitution: so that laws are passed fairly, honestly, and on their individual merits. Article II, section 19 holds our lawmakers to exacting standards to ensure that legislation is honestly come-by; we must hold citizen and corporate legislators to no less. Kiga, 144 Wn.2d at 824. This initiative would violate the constitution if our legislature had passed it, and it is equally unconstitutional as an initiative of the people.
¶88 I respectfully dissent.
C. Johnson and Fairhurst, JJ., concur with Wiggins, J.See ch. 90, § 6, 10 Stat. 172, 175 (1853); see also City of Fircrest v. Jensen, 158 Wn.2d 384, 403-04, 143 P.3d 776 (2006).
An “earmark” is money that is “set aside for a specific purpose or recipient.” Black’s Law Dictionary 584 (9th ed. 2009).
Respondents are incorrect in suggesting we have applied Covell only to taxes at the municipal level, not the state level. See, e.g., Franks & Son, 136 Wn.2d at 750-51 (applying Covell test to Washington Utilities and Transportation Commission charge to determine whether it was a tax or a regulatory fee); Dean v. Lehman, 143 Wn.2d 12, 27-28, 18 P.3d 523 (2001) (applying Covell test to Washington State Department of Corrections charge to determine whether it was a tax).
There is one possible exception. Counsel for intervenor-respondent Costco Wholesale Corporation appeared to endorse this position at oral argument, suggesting that a title would not violate article II, section 19 even if it included a falsehood. Wash. Supreme Court Oral Argument, Wash. Ass’n for Substance Abuse & Violence Prevention v. State, No. 87188-4 (May 17, 2012), at 46 min., 28 sec., video recording by TVW, Washington State’s Public Affairs Network, available at http://www.tvw.org.
The phrase refers to an old frontier custom of neighbors working together to move logs that each had cut, which would be more difficult to move without help. Black’s Law Dictionary defines “logrolling” as “[t]he legislative practice of including several propositions in one measure ... so that the legislature or voters will pass all of them, even though these propositions might not have been passed if they had been submitted separately.” Black’s, supra, at 1026.
The parties do not dispute that the title is general. Clerk’s Papers at 1617.
However, we have held on numerous occasions that it violates article II, section 19 to include substantive law in an appropriations bill. Wash. State Legislature v. State, 139 Wn.2d 129, 145, 985 P.2d 353 (1999).
It is something of an open question whether our “rational unity” test resembles a “rational basis” test in that we are required to give credence to hypotheticals that are unlikely to occur. See, e.g., Loparex, LLC v. MPI Release Technologies, LLC, 964 N.E.2d 806 (Ind. 2012) (“ c[I]f there is any reasonable basis for grouping together in one act various matters of the same nature, and the public cannot be deceived thereby, the act is valid.’ ” (quoting Stith Petroleum Co. v. Dep’t of Audit & Control, 211 Ind. 400, 409, 5 N.E.2d 517, 521 (1937))). For example, in theory, local governments could allocate every penny of the money they receive from 1-1183 to alcohol-related programs. But this possibility is, at best, extremely improbable. In any event, given the dangers of logrolling here, the mere possibility that money could be allocated to alcohol-related purposes is too tenuous to uphold an otherwise deeply flawed law.