DELTA AIR LINES INC v. DEPARTMENT OF REVENUE (2025)

Under Oregon law, most businesses have the value of their property—and thus the amount of their tax—determined by the county assessor. Some businesses, however, are assessed centrally, by the Department of Revenue itself. When the county assessor does the assessment, the amount of the tax is calculated based on the value of the taxpayer's real and tangible personal property, but not its intangible property. When a business is centrally assessed, however, the amount of the tax is calculated on the value not just of real and tangible personal property, but also intangible property.

The taxpayers in two closely related cases—Delta Air Lines, Inc., and PacifiCorp—are businesses subject to central assessment. As relevant here, both taxpayers contend that taxing centrally assessed businesses on intangible property violates the state and federal constitutions, because locally assessed businesses are not taxed on their intangible property. Specifically, the taxpayers argue that such a tax is not uniform as required by Article I, section 32, and Article XI, section 1, of the Oregon Constitution, and that the legislature's classification violates the Equal Privileges and Immunities Clause of the Oregon Constitution (Article I, section 20), or the Equal Protection Clause of the Fourteenth Amendment to the United States Constitution.

The Tax Court did not consolidate the cases, but it addressed both taxpayers’ arguments on the constitutional questions in a single opinion: Delta Air Lines, Inc. v. Dept. of Rev., 25 OTR 308, 2023 WL 5425246 (2023). The court agreed with Delta and held that it is unconstitutional to tax the intangible property of air transportation businesses such as Delta. See id. at 351-52 (summarizing conclusion). However, the court reached a different conclusion regarding the intangible property of utilities, and so it rejected PacifiCorp's constitutional claim. See id. at 352-53 (also summarizing conclusion). As to PacifiCorp, the court issued a short separate opinion adopting the reasoning and explanation it had set out in Delta. PacifiCorp v. Dept. of Rev., 25 OTR 367, adh'd to on recons, 25 OTR 419, 2023 WL 6389963 (2023).

We are now presented with the constitutional questions on appeal. The department appeals the Tax Court's holding that the tax on intangible property is unconstitutional as to air transportation businesses such as Delta, while PacifiCorp appeals the Tax Court's holding that the tax on intangible property is constitutional as to utilities such as itself.

For the reasons that follow, we reverse the Tax Court's conclusion that the tax is unconstitutional as to Delta, and, in a separate opinion to follow, affirm the Tax Court's conclusion that the tax is constitutional as to PacifiCorp.¹ As we will explain, taxpayers’ arguments are best understood to challenge the constitutionality of the legislature's classifications under Oregon's Equal Privileges and Immunities Clause and the United States Constitution's Equal Protection Clause. The test under both clauses is similar: whether the legislative classification is rationally related to a legitimate legislative purpose. We conclude that the tax on the intangible property of centrally assessed businesses is constitutional. The state has a legitimate purpose in obtaining revenue, and the taxation of intangible property is rationally related to that purpose. The legislature's decision to limit the taxation of intangible property to centrally assessed businesses rationally promotes various legitimate purposes, including administrative efficiency, developing and keeping expertise in valuing such businesses, promoting fairness among the centrally assessed taxpayers, and balancing the expected revenue return against limited departmental resources.

The uniformity provisions of the Oregon Constitution—Article I, section 32, and Article IX, section 1—do not impose any relevant additional limits on the classes that the legislature may create. Accordingly, the tax on intangible property imposed on both centrally assessed businesses does not violate the uniformity provisions.

I. BACKGROUND

A. Central Assessment and Intangible Property

1. Current Oregon law

In general, most property taxes are assessed at the county level by the county assessor. ORS 308.210(1). Some industries, however, are centrally assessed—that is, they are assessed by the Department of Revenue directly. The businesses that Oregon centrally assesses are listed in ORS 308.515(1).² Delta is centrally assessed under ORS 308.515(1) (e) because it provides “[a]ir transportation.” PacifiCorp is centrally assessed under ORS 308.515(1)(k) because it sells electricity.

Among the ways in which centrally assessed businesses are taxed differently is that they are taxed on intangible property. In general, “intangible personal property is not subject to assessment and taxation,” except as directed in the central assessment statutes, ORS 308.505 to 308.674. ORS 307.030(2). The central assessment statutes confirm that centrally assessed businesses are taxed on intangible property. See ORS 308.505(14)(a) (defining “property” for purposes of central assessment as “all property of any kind, whether real, personal, tangible or intangible”).

The legislature has not defined “intangible property” in the context of central assessment.³ For purposes of this case, it is sufficient to rely on the general principles that we have articulated in past cases: intangible property is “ ‘property representative of a right rather than a physical object,’ ” such as “ ‘patents, * * * goodwill, trademarks, franchises, and copyrights.’ ” Tektronix, Inc. v. Dept. of Rev., 354 Or. 531, 543-44, 316 P.3d 276 (2013) (quoting definition of “intangible assets” in West's Tax Law Dictionary 570 (2013)); see also Powerex Corp. v. Dept. of Rev., 357 Or. 40, 61, 346 P.3d 476 (2015) (contrasting intangible property with tangible property).⁴

When valuing centrally assessed businesses, the department uses a method known as unit valuation. The term itself broadly means valuing a business as a unit—as a “going concern”⁵ —instead of by adding together the value of individual properties. See, e.g., Michael T. Raymond, Why Federal Preemption Is Needed to End Discriminatory Taxation of Telecommunications Property, 6 St & Loc Tax Lawyer 15, 22 (2001); Bruce A. Fowler, Unit Valuation: Oklahoma's Illegal Tax on Intangible Property, 31 Tulsa LJ 367, 370 (1995). In Oregon, unit valuation is authorized by ORS 308.555 (for centrally assessed businesses, department is authorized to “value the entire property, both within and without the State of Oregon, as a unit”).⁶

2. Historical context

The Tax Court implied that Oregon is unusual in listing the industries subject to central assessment, rather than setting out a series of factors used to identify who is subject to central assessment. See 25 OTR at 315 n 9 (“The legislature's use of a list in ORS 308.515(1) contrasts with the more typical approach of stating a definition based on express criteria.”); id. at 350 (“the legislature has exercised its prerogative to draw the classification line by means of a list rather than by specifying criteria”). Implicitly, the Tax Court raises an important question: Why were these particular industries chosen to be centrally assessed and to have their intangible property taxed?

Oregon's central assessment law is not out of the ordinary among the states in terms of the industries to which it applies. Unit valuation and central assessment proceeded across the nation, beginning in the nineteenth century. Central assessment came as a response to the challenges associated with having local taxing districts attempt to fairly apportion the value of property owned by entities that might span numerous districts, such as railroads. Before the second half of the nineteenth century, local assessment mechanisms were already showing their inadequacy to meet the challenges presented by railroads, because even local railroads often spanned multiple taxing districts. James C. Bonbright, 2 The Valuation of Property 635-37 (1937); see Comcast Corp. v. Dept. of Rev., 356 Or. 282, 289-91, 337 P.3d 768 (2014). Local assessors could assess the value at the replacement cost of the railroad property located in their district, or they could determine the “going concern” value of the entire railroad, then attempt to apportion that value to the part of the railroad that lay in their taxing district. Bonbright, 2 Valuation at 635. The courts, however, were often very skeptical of the fairness of having local assessors value the going concern, and so sometimes required local assessors to use the lower replacement cost instead. Id. at 635-37; see Comcast, 356 Or. at 289-90, 337 P.3d 768. Besides which, local assessment was simply not amenable to unit valuation. See Bonbright, 2 Valuation at 637 (“So serious are the practical difficulties of applying the unit rule to separate railroad assessments by small political subdivisions, that the rule itself could hardly have developed under this procedure.”).

The consolidation of the railroads into large interstate businesses only exacerbated the problems. It led to popular demand for assessment methods to reach intangible values, including goodwill, that could not effectively be reached by local assessors. Bonbright, 2 Valuation at 637. “Hence, statutes were passed which set up state boards of assessment and which directed these boards to include the entire values of the railroads as distinct from, or in addition to, the so-called ‘cost values’ of the physical assets.” Id.

The “going concern” value is essentially another name for modern unit valuation. Conceptually, unit valuation reflects the fact that some industries have substantially more value than their tangible assets alone may reflect. “[I]t is a cardinal rule which should never be forgotten that whatever property is worth for the purposes of income and sale it is also worth for purposes of taxation.” Adams Express Company v. Ohio, 166 U.S. 185, 220, 17 S Ct 604, 41 L Ed. 965 (1897) (Adams Express II) (upholding a state property tax that used unit valuation to reach such intangible property as goodwill). “Substance of right demands that whatever be the real value of any property, that value may be accepted by the State for purpose of taxation, and this ought not to be evaded by any mere confusion of words.” Id. at 221, 17 S Ct 604. The Court went on to set out precisely why a state was entitled to tax both tangible and intangible property:

“To the owners [of the business], for the purposes of income and sale, the corporate property is worth hundreds of thousands of dollars. Does substance of right require that it shall pay taxes only upon the thousands of dollars of tangible property which it possesses? Accumulated wealth will laugh at the crudity of taxing laws which reach only the one and ignore the other, while they who own tangible property, not organized into a single producing plant, will feel the injustice of a system which so misplaces the burden of taxation.”

Id.

Although central assessment and unit valuation began in the 19th century with railroads, even before the end of that century the states had expanded those doctrines to reach telegraph companies and express companies. See Bonbright, 2 Valuation at 648-57 (reviewing developments only through 1900).

“Oregon's original central assessment scheme was consistent with the development of unit valuation and central assessment statutes nationally.” Comcast, 356 Or. at 291, 337 P.3d 768. In Comcast, we reviewed the various business types subject to unit valuation and central assessment in Oregon. Id. at 291-93, 337 P.3d 768. The Tax Court also did so here in a lengthy appendix to its decision. Delta, 25 OTR at 355-66.

Today, Oregon's choices of which particular businesses to subject to central assessment and unit valuation do not appear out of the ordinary nationally. Various commentators have provided overarching summaries of national laws regarding unit valuation, and those summaries indicate as much:

“The companies which are usually subject to unit valuation include both regulated and unregulated public utilities and transportation companies. They are usually interstate, capital-intensive businesses that employ systems of interdependent interrelated assets in multiple jurisdictions. Although the businesses were historically almost always publicly regulated utilities, this is no longer necessarily true due to extensive deregulation.”

Fowler, 31 Tulsa LJ at 371 (internal quotation marks and footnote omitted).

“Traditionally, the ‘regulated industries’ have fallen primarily into three categories: (1) energy (gas, electric and pipeline) and water public utilities, (2) telecommunications (telephone and other) public utilities, and (3) regulated transportation companies (air, motor, rail and water carriers).”

James A. Amdur, Property Taxation of Regulated Industries, 40 Tax Lawyer 339, 339 (1987); see also Powell, 30 Cornell J L & Pub Pol'y at 248 (“Today, centrally assessed properties include not only public utilities, but also more broadly multi-jurisdictional properties such as railroads; telegraph, telephone, and telecommunications companies; pipelines; and airlines.”); Bonbright, 2 Valuation at 634-35 (“In ad valorem taxation * * *, assessments under the unit rule have been largely confined to public-utility properties—especially to railroad, telegraph, long-distance telephone, pipe line, and express properties.” (Footnote omitted.)). Oregon's list of businesses subject to unit valuation and central assessment almost entirely overlaps with those standard categories.⁷

As the Tax Court also noted, some states at least purport to exclude intangible property from the taxation of centrally assessed businesses. See 25 OTR at 331-32; National Conference of State Legislatures, Property Taxation of Communications Providers, A Primer for State Legislatures 4 (April 2024), https://documents.ncsl.org/wwwncsl/State-Federal/NCSL-SALT-Property-Taxation-of-Communications-Providers-April-2024.pdf (accessed July 17, 2025); Fowler, 31 Tulsa LJ at 376; Walter Hellerstein, State and Local Taxation of Intangibles Generates Increasing Controversy, 80 J Tax'n 296, 302 (1994).

The extent to which that is actually true is less clear. As indicated above, unit valuation and central assessment were tools specifically intended in part to facilitate the assessment of intangible property. Moreover, it is difficult to engage in unit valuation without considering the value of intangible property. As the Wisconsin Supreme Court noted over 100 years ago:

“One might as well try to value the life-blood of a horse, or his capacity to breathe, as try to place a value upon the visible part of railroad property separate from its rights, franchises, and privileges.”

Chicago & N.W. Ry. Co. v. State, 128 Wis. 553, 621-22, 108 N.W. 557, 573 (1906). States that claim not to tax intangible property have been criticized for doing so indirectly through the mechanism of unit valuation. See National Conference of State Legislatures, Property Taxation of Communications Providers at 4 (unit valuation of telecommunication companies “can result in taxation of more than the actual value of tangible property by including significant intangible value, even in states where intangibles are statutorily or constitutionally exempt from taxation”); Fowler, 31 Tulsa LJ at 379 (“Although a majority of states have made a public policy decision to exclude intangible property from property taxation, their taxing system may be effectively taxing the property through the application of unit valuation techniques in assessing public service companies.”); Hellerstein, 80 J Tax'n at 302 (although many states purport to prohibit taxation of intangible property, “[t]he problem * * * is that under the guise of merely ‘considering’ intangible values in determining the value of tangible property, the intangible property is itself being taxed”). This court has described other states’ attempts to use unit valuation while excluding intangible property as an “artificial focus on tangible property.” DISH Network Corp. v. Dept. of Rev., 364 Or. 254, 292, 434 P.3d 379 (2019).

In sum: To the extent that Delta's and PacifiCorp's arguments are predicated in part on an assumption that Oregon is an outlier in the businesses that it subjects to central assessment and unit valuation, or in taxing the intangible property of those businesses, that assumption is not well-founded.

B. Tax Court Proceedings

Both Delta and PacifiCorp filed actions in the Tax Court challenging the department's assessments: Delta challenged the assessment for tax year 2019-20,⁸ while PacifiCorp challenged the assessment for tax year 2020-21. Both actions challenged the department's determination of the property values used to calculate their tax liability. In addition, Delta and PacifiCorp both contended that it was unconstitutional for Oregon to tax the intangible property only of centrally assessed businesses, and not locally assessed businesses. Both taxpayers raised the issues in their respective cases by motions for partial summary judgment.

Both taxpayers ultimately relied on four constitutional provisions. The first two both relate specifically to uniformity of taxation. First, Article I, section 32, of the Oregon Constitution provides in part:

“ * * * [A]ll taxation shall be uniform on the same class of subjects within the territorial limits of the authority levying the tax.”

Second, Article IX, section 1 reads:

“The Legislative Assembly shall, and the people through the initiative may, provide by law uniform rules of assessment and taxation. All taxes shall be levied and collected under general laws operating uniformly throughout the State.”

The second two constitutional provisions at issue are not tax-specific but, instead, limit the government's ability to treat citizens differently. Oregon's Equal Privileges and Immunities Clause, Article I, section 20, provides:

“No law shall be passed granting to any citizen or class of citizens privileges, or immunities, which, upon the same terms, shall not equally belong to all citizens.”

Finally, the Equal Protection Clause of the Fourteenth Amendment to the United States Constitution provides in part:

“No State shall make or enforce any law which shall * * * deny to any person within its jurisdiction the equal protection of the laws.”

US Const, Amend XIV, § 1.

Under all four provisions, the Tax Court concluded that the governing legal standard required it to consider whether the state's decision to tax the intangible property only of centrally assessed businesses satisfied the so-called “rational basis test.” See 25 OTR at 312-13. The Tax Court also concluded, however, that any tax classification is subject to a preliminary requirement that there must be “genuine differences” in the nature or use of the property being taxed, before the court will consider whether the rational basis test has been met. See id. at 322-27; id. at 353. Based on that understanding, the Tax Court concluded that there were no such genuine differences between the intangible property of centrally assessed air transportation businesses and that of locally assessed bus and trucking companies, and so the tax was unconstitutional as to Delta. Id. at 344-45, 350-51. But the court concluded that there were genuine differences between the intangible property used by a rate-regulated utility and the intangible property used by other businesses, and so the tax was constitutional as to PacifiCorp. Id. at 352-53.

II. DISCUSSION

A. Uniformity Provisions: Article I, Section 32, and Article IX, Section 1

1. Overview

Two of the provisions of the Oregon Constitution raised by the taxpayers relate to taxes specifically: Article I, section 32, and Article IX, section 1. We refer to them collectively as the uniformity provisions. Both were part of the original Oregon Constitution, but both were substantially rewritten by amendment in 1917.

We begin with the text. As noted, the relevant part of Article I, section 32, currently provides:

“[A]ll taxation shall be uniform on the same class of subjects within the territorial limits of the authority levying the tax.”

Article IX, section 1, currently reads:

“The Legislative Assembly shall, and the people through the initiative may, provide by law uniform rules of assessment and taxation. All taxes shall be levied and collected under general laws operating uniformly throughout the State.”

Textually, Article I, section 32, requires only that a tax be uniform within whatever class the taxing authority creates, provided that it is uniform “within the territorial limits of the authority levying the tax.” See Tharalson v. State Dept. of Rev., 281 Or. 9, 16, 573 P.2d 298 (1978) (uniformity provisions were intended to allow legislature to allow taxes “so long as the tax was uniform geographically and within one ‘class of subjects’ ”); Standard Lbr. Co. v. Pierce et al., 112 Or. 314, 335-36, 228 P. 812 (1924) (Standard Lumber) (noting the distinction). “[T]he use of the phrase ‘within the territorial limits of the authority levying the tax’ raised the principle of territorial uniformity to that of an express constitutional requirement.” Jarvill v. City of Eugene, 289 Or. 157, 177, 613 P.2d 1, cert. den., 449 U.S. 1013, 101 S.Ct. 572, 66 L.Ed.2d 472 (1980); see id. at 170-78, 613 P.2d 1 (detailing origin and nature of territorial uniformity requirement); Orval Etter, Municipal Tax Differentials, 37 Or L Rev 1, 40-41 (1957) (while legislature can use “many factors” to classify property for taxation, property's location “apparently is ruled out explicitly and completely by the requirement of territorial uniformity in the Oregon [C]onstitution”).

Article IX, section 1, is similar, save that it does not refer to classes at all. “We have held that these two constitutional provisions requiring tax uniformity are to be read together.” Jarvill, 289 Or. at 171 n 15, 613 P.2d 1 (citing State ex rel v. Malheur County Court, 185 Or. 392, 411, 203 P.2d 305 (1949)).⁹

The present text of both provisions was enacted in 1917. In the original 1859 Oregon Constitution, Article I, section 32, had required taxes to be not just uniform, but also equal.¹⁰ Original Article IX, section 1, similarly required equality and uniformity, but it further limited the reasons why the legislature could exempt an entity from a tax—to such things as educational, scientific, or charitable purposes, and others, all listed.¹¹

Those 1859 restrictions applied only to property taxes, not to other forms of taxes, but there were concerns that the limits might be read more expansively. See Standard Lumber, 112 Or. at 334-35, 228 P. 812 (so noting). The 1917 amendments were intended to remove most of the existing restrictions and allow the legislature to classify people and property for tax purposes. Standard Lumber summarized the reasoning of the voters:

“Demand was made for removal of those constitutional restrictions, which prevented the classification of property in respect to its nature, condition or class, and the imposition thereon of different rates of taxation upon different classes of property; and which excluded considerations of faculty or ability to pay, equality of sacrifice or governmental advantages provided to the taxpayer[.]”

112 Or. at 335, 228 P. 812 (citations omitted). Jarvill, citing an extensive list of cases and contemporary sources, agreed that the amendments

“were intended to permit the reasonable classification of subjects of taxation, the exemption of certain property from taxation, and the imposition of different rates of taxation upon different classes of property.”

289 Or. at 176-77, 613 P.2d 1.

Textually and historically, then, the uniformity provisions would not seem to limit the types of classifications that the government can use in taxation, provided that the classifications are consistent within the territorial limits of the relevant lawmaking body. In fact, the uniformity provisions were amended specifically to remove existing limits, with Article I, section 32, changed to permit classifications, and Article IX, section 1, changed to remove limits on tax exemptions. Within whatever class the taxing authority may create, if the tax applies uniformly across the territory of the taxing jurisdiction, and if the tax is applied uniformly across that territory, then the uniformity provisions have been satisfied:

“Selecting and classifying incomes and fixing different rates and exemptions does not violate section 32 of Article 1, or section 1 of Article IX of the Oregon Constitution as the taxes are uniform on the same classes of subjects and are collected under general laws and are uniform throughout the state.”

McPherson v. Fisher, 143 Or. 615, 622, 23 P.2d 913 (1933); see Standard Lumber, 112 Or. at 335-36, 228 P. 812 (“[T]he Constitution * * * places no restraint upon the power of the legislature in the matter of taxation which was not already enforced upon it by the 14th Amendment to the federal Constitution, with this qualification, * * * that among the members or objects included in a class selected by the legislature, inherent uniformity as well as territorial uniformity is required.”).

Nevertheless, some of our cases have suggested that the uniformity provisions themselves also require that the classification be rational. It is possible that those statements were merely conflating the requirements of the uniformity provisions with the requirements of equal protection and/or equal privileges, as those provisions are all usually at issue at the same time. For example, Mathias v. Dept. of Rev., 312 Or. 50, 59, 817 P.2d 272 (1991), stated that “a classification, to survive the protections of the uniformity in taxation clauses of the state constitution, must be based on real differences between the subjects disparately treated by the classification.” The cases that Mathias cited for that proposition, however, explicitly referenced only equal protection principles. See Jarvill, 289 Or. at 180, 613 P.2d 1; Huckaba v. Johnson, 281 Or. 23, 25-26, 573 P.2d 305 (1978); Dutton Lbr. Corp. v. Tax Com., 228 Or. 525, 539, 365 P.2d 867 (1961).

For purposes of this opinion, we do not need to resolve whether the uniformity provisions impose any independent restrictions on classification beyond the requirement of territorial uniformity. Even if they do, neither the text, the history, nor that prior case law suggests that the requirement would be different from, or more stringent than, the requirements imposed by equal protection and equal privileges and immunities. Thus, our analysis of Oregon's Equal Privileges and Immunities Clause and the federal Equal Protection Clause—discussed below—would also fully address any such rationality standard under Oregon's uniformity provisions.

The only legal principle unique to the uniformity provisions is the requirement of territorial uniformity. Territorial uniformity, however, is not at issue in this case. The statutes that tax the intangible property of centrally assessed businesses apply uniformly across the state.

2. As applied

Although it is questionable whether the uniformity provisions themselves restrict the permissible classifications, the uniformity provisions do prohibit taxing authorities from choosing to apply that law to taxpayers in a way that violates uniformity. “The taxing authorities may not single out one taxpayer for discriminatory, or selective, enforcement of a tax law that should apply equally to all similarly situated taxpayers.” Penn Phillips Lands v. Tax Com., 247 Or. 380, 385-86, 430 P.2d 349 (1967); see Pacificorp Power Marketing v. Dept. of Rev., 340 Or. 204, 219, 131 P.3d 725 (2006) (quoting Penn Phillips).¹² The uniform reduction of all assessments save that of a single taxpayer would state such a claim, see Reynolds Metals Co. v. State Tax Com., 227 Or. 467, 471-73, 362 P.2d 705 (1961), but evidence that a few tracts had been undervalued would not, see Robinson et ux v. State Tax Com., 216 Or. 532, 537, 339 P.2d 432 (1959) (otherwise, “[i]t would be a rare case in which undervaluation in a few tracts could not be pointed out” (internal quotation marks and citation omitted)).

PacifiCorp asserts that Oregon's intangible property tax is no longer being applied uniformly, relying on federal case law interpreting a federal statute. Under 49 USC § 11501(b)(4), states are prohibited from imposing taxes that “discriminate[ ] against a rail carrier” by treating it differently from other “commercial and industrial property in the same assessment jurisdiction.” In BNSF Railway Co. v. Oregon Department of Revenue, 965 F.3d 681 (9th Cir 2020), the Ninth Circuit held that that statute prohibits Oregon from taxing the intangible property of railroads, because Oregon law treats railroads differently from Oregon commercial and industrial taxpayers generally. See id. at 691-93 (rejecting department's argument that treatment of railroads should only be compared to other centrally assessed businesses). PacifiCorp argues that Oregon's statutes direct the state to tax the intangible property of a class that includes railroads, but railroads are no longer being taxed on their intangible property. Because the tax is no longer being applied to all members of the class listed in the Oregon statutes, PacifiCorp contends that the tax violates the uniformity provisions.

We are not persuaded. Congress has, by federal law, prohibited Oregon from including railroads in the class of businesses taxed on intangible property. Pursuant to the Supremacy Clause of the United States Constitution, that law is controlling. See US Const., Art. VI, cl 2 (“This Constitution, and the Laws of the United States which shall be made in Pursuance thereof * * *, shall be the supreme Law of the Land; and the Judges in every State shall be bound thereby, any Thing in the Constitution or Laws of any State to the Contrary notwithstanding.”). Our “as applied” cases under the uniformity provisions do not consider actions at the level of lawmaking; instead, they focus on the actions of the “taxing authorities,” as to which the uniformity provisions prohibit only “an intentional and systematic pattern of discrimination.” Pacificorp Power, 340 Or. at 219, 131 P.3d 725; see Meadowland Ranches v. Dept. of Rev., 277 Or. 769, 776, 562 P.2d 183 (1977) (claim of unconstitutional discrimination requires “arbitrary and systematic discrimination” (internal quotation marks and citations omitted)). Conceptually, an Act of Congress is not the same as an assessor systematically refusing to properly assess the properties of certain taxpayers. The nature of PacifiCorp's argument seems more properly addressed as a challenge to the validity of the classifications as a whole under equal protection and equal privileges and immunities, and we will return to it later in this opinion.

B. Federal and State Constitutional Limits on Classification

One limit on the legislature's authority to create classifications comes from Oregon's Equal Privileges and Immunities Clause, Article I, section 20. As noted, it provides:

“No law shall be passed granting to any citizen or class of citizens privileges, or immunities, which, upon the same terms, shall not equally belong to all citizens.”

Another limit comes from the Equal Protection Clause of the Fourteenth Amendment to the United States Constitution:

“No State shall make or enforce any law which shall * * * deny to any person within its jurisdiction the equal protection of the laws.”

Although both those provisions involve questions of equality, they differ not just in their text, but also in their history and in how they have been interpreted. We have described Oregon's Equal Privileges and Immunities Clause as the “ ‘antithesis’ ” of the federal Equal Protection Clause: The latter was adopted to prevent the government from unfairly disadvantaging classes, while the former was adopted to prevent the government from unfairly favoring classes. E.g., Hewitt v. SAIF, 294 Or. 33, 42, 653 P.2d 970 (1982) (citations omitted; discussing and citing authorities).

That is not to say that the provisions are unrelated; in fact, both provisions often require the courts to apply similar analytical schemes. Under both, most classifications will be held constitutional on very minimal showings. See, e.g., Cleburne v. Cleburne Living Center, Inc., 473 U.S. 432, 440, 105 S Ct 3249, 87 L Ed. 2d 313 (1985) (federal equal protection); Kramer v. City of Lake Oswego, 365 Or. 422, 454-57, 446 P.3d 1, adh'd to as modified on recons., 365 Or. 691, 455 P.3d 922 (2019) (discussing and rejecting higher standards of scrutiny under equal privileges and immunities). But both federal equal protection and state equal privileges also recognize some classifications—such as race, alienage, or national origin—as “suspect,” requiring an extremely strong justification before the classification will be held to be constitutional. See, e.g., Cleburne, 473 U.S. at 440, 105 S.Ct. 3249 (such classifications “are so seldom relevant to the achievement of any legitimate state interest that laws grounded in such considerations are deemed to reflect prejudice and antipathy”); Hewitt, 294 Or. at 45, 653 P.2d 970 (classification is “ ‘suspect’ when it focuses on ‘immutable’ personal characteristics,” which “can be suspected of reflecting ‘invidious’ social or political premises, that is to say, prejudice or stereotyped prejudgments”). Although federal equal protection recognizes a third category of classification as subject to an intermediate level of scrutiny—classifications based on gender being one of these—Oregon subjects such classifications to the same scrutiny as other “suspect” classifications. See Kramer, 365 Or. at 454-56, 446 P.3d 1 (so noting).

In this case, it is undisputed that the classifications should be analyzed under the most lenient scrutiny standards of state and federal law. In keeping with our traditional practice, we first consider the Oregon Constitution's Equal Privileges and Immunities Clause before we turn to the federal Equal Protection Clause. See State v. Babson, 355 Or. 383, 432-33, 326 P.3d 559 (2014) (discussing court's practice).

1. Oregon's Equal Privileges and Immunities Clause

a. General

As noted, the parties agree that this case does not involve any “suspect” class, and so our analysis should apply the most lenient standards that exist under both state equal privileges and immunities law and federal equal protection law.¹³ As to the former, this court has concluded that the test is whether the classification is rationally related to a legitimate governmental purpose. See Kramer, 365 Or. at 456-57, 446 P.3d 1 (reviewing different tests this court has used over time for equal privileges and immunities); id. at 461, 446 P.3d 1 (summarizing test). That framing matches similar terms used to describe the lenient test under federal equal protection law. See U.S. Railroad Retirement Bd. v. Fritz, 449 U.S. 166, 174-76, 101 S Ct 453, 66 L Ed. 2d 368 (1980) (reviewing different formulations United States Supreme Court had used over time for equal protection). Both tests may thus be described as the “rational basis test” or “rational basis review.”

In the following discussion, we will cite numerous United States Supreme Court decisions interpreting federal equal protection law. For purposes of this part of the analysis, we cite United States Supreme Court decisions for only their persuasive value as to why a legislative classification might be rationally related to a legitimate governmental purpose. We will address the Equal Protection Clause separately later in this opinion.

Rational basis review is especially lenient in the context of tax classifications. This court has explained that neither Oregon's Equal Privileges and Immunities Clause nor the federal Equal Protection Clause requires

“a formula of rigid uniformity in framing measures of taxation. [The legislature] may tax some kinds of property at one rate, and others at another, and exempt others altogether, and it may lay an excise on the operations of a particular kind of business, and exempt some other kind of business closely akin thereto.”

Garbade and Boynton v. City of Portland, 188 Or. 158, 192, 214 P.2d 1000 (1950) (internal quotation marks and citation omitted), overruled in part on other grounds by Multnomah County v. Mittleman, 275 Or. 545, 556-57, 552 P.2d 242 (1976); see Wittenberg et al v. Mutton et al, 203 Or. 438, 447, 280 P.2d 359 (1955) (quoting Garbade with approval); see also Knight v. Dept. of Rev., 293 Or. 267, 271, 646 P.2d 1343 (1982) (“The legislature has wide discretion in classifying subjects of taxation.”); Jarvill, 289 Or. at 178, 613 P.2d 1 (“we have also recognized and expressly held that a taxing authority has a wide range of discretion to classify subjects of taxation”); Wittenberg, 203 Or. at 446-47, 280 P.2d 359 (state legislature “has the widest possible latitude” regarding taxation (internal quotation marks and citations omitted)).

To satisfy the “legitimate purpose” test, the legislature need not have actually stated its purpose; it is sufficient if the court can conceive of one. See Smith et al v. Columbia County et al, 216 Or. 662, 684, 341 P.2d 540 (1959), appeal dismissed, 362 U.S. 215, 80 S.Ct. 668, 4 L.Ed.2d 736 (1960) (citing cases for the proposition that “the legislature is not required to give any index or catalog of its reasons for the classification”); Garbade, 188 Or. at 192, 214 P.2d 1000 (explaining that city council did not need to “record a complete catalogue of the considerations which moved its members to enact” the ordinances (internal quotation marks omitted)); Standard Lumber, 112 Or. at 328, 228 P. 812 (“it is not necessary that the basis of the classification must be deducible from the nature of the things classified”); see also Huckaba, 281 Or. at 26, 573 P.2d 305 (court will uphold statute under rational basis review if “any conceivable state of facts * * * would support it”); Carmichael v. Southern Coal Co., 301 U.S. 495, 509, 57 S Ct 868, 81 L Ed. 1245 (1937) (articulating same principle for equal protection).¹⁴

Any number of legitimate purposes can support tax classifications drawn by the legislature. For example, the legislature may draw lines for administrative convenience or efficiency. Huckaba, 281 Or. at 30-31, 573 P.2d 305; see Carmichael, 301 U.S. at 511, 57 S.Ct. 868. It may want to protect or foster a particular industry. State v. Pyle, 226 Or. 485, 490, 360 P.2d 626 (1961); see Carmichael, 301 U.S. at 512, 57 S.Ct. 868; Etter, 37 Or L Rev at 42. It may want to encourage publicly desirable enterprises such as charities and educational institutions. See Corporation of Sisters of Mercy v. Lane Co., 123 Or. 144, 161-64, 261 P. 694 (1927) (legislature did not violate equal privileges and immunities by granting tax exemption to hospitals, because “the establishment and maintenance of charitable hospitals serves the public welfare”); see also Carmichael, 301 U.S. at 512, 57 S.Ct. 868 (noting general principle in equal protection context). It may choose to tax one industry rather than another because of downstream economic effects. See Thomas M. Cooley, A Treatise on the Law of Taxation 124-25 (1st ed. 1876) (giving example of a tax on breadstuffs, which would disproportionately harm the poor).

The classification is not required to serve a single purpose or be derived from a single rule. The legislature can create a general rule for one reason and an exemption for a different reason. See Pyle, 226 Or. at 489, 360 P.2d 626 (“The mere fact that the principal purpose of the legislation was to preserve the highways does not mean that the legislature may not also have had other considerations in mind.”); id. (“Legislation usually is the product of the adjustment of various interests.”); Wittenberg, 203 Or. at 446, 280 P.2d 359 (legislature “may make distinctions of degree having a rational basis” (internal quotation marks and citation omitted)); see also Carmichael, 301 U.S. at 510-14, 57 S.Ct. 868 (noting that state may decide not to include a group within a tax because it is administratively difficult or unduly expensive, or because it chooses to protect or foster specific industries).

The “rational” part of rational basis review requires only that the classification drawn by the legislature rationally serve the purpose(s). See Knapp v. City of Jacksonville, 342 Or. 268, 276, 151 P.3d 143 (2007) (differences must have “reasonable relationship to the legislative purpose” (internal quotation marks omitted)); Huckaba, 281 Or. at 26, 573 P.2d 305 (requiring both that there be differences between groups of taxpayers treated differently, and that those differences have “a reasonable relationship to the legislative purpose”). But the classification need not be perfect. “A legislature is not bound to tax every member of a class or none.” Wittenberg, 203 Or. at 446, 280 P.2d 359 (internal quotation marks and citation omitted). “The broad legislative classification must be judged by reference to characteristics typical of the affected class rather than by focusing on selected atypical examples.” Huckaba, 281 Or. at 30, 573 P.2d 305. General rules that promote administrative efficiency are permitted, even if the result might be “seemingly arbitrary” in individual cases. Huckaba, 281 Or. at 30-31, 573 P.2d 305.¹⁵ In other words, a classification is not invalid merely because a better one could be made. School Dist. No. 12 v. Wasco County, 270 Or. 622, 629, 529 P.2d 386 (1974); see Cooley, Taxation at 125 (“[A] tax cannot be attacked on averment and proof that some other tax for the same purpose would have been more just and more equal.”).

A legislative solution may be rational even if it does not fully address the problem. As Justice Holmes once explained regarding the Equal Protection Clause, it “is not a pedagogical requirement of the impracticable”; a legislature “may do what it can to prevent what is deemed an evil and stop short of those cases in which the harm to the few concerned is thought less important than the harm to the public that would ensue if the rule laid down were made mathematically exact.” Dominion Hotel v. Arizona, 249 U.S. 265, 268, 39 S Ct 273, 63 L Ed. 597 (1919); see, e.g., Mallatt v. Luihn et al., 206 Or. 678, 702, 294 P.2d 871 (1956) (citing Dominion Hotel with approval and holding that “a classification having some reasonable basis does not offend against the Federal Constitution or the Constitution of this state merely because it is not made with mathematical nicety or because in practice it results in some inequality”).

The foregoing statements stand for the proposition that legislative judgments in the area of taxation are granted a degree of flexibility. They do not represent carte blanche permission for any classification that the legislature may make or any end that the legislature may seek. For example, a law may be drawn so broadly, or so narrowly, that it fails to rationally promote any legitimate end. See State ex rel Huddleston v. Sawyer, 324 Or. 597, 629, 932 P.2d 1145, cert. den., 522 U.S. 994, 118 S.Ct. 557, 139 L.Ed.2d 399 (1997) (“A law may be so overinclusive or underinclusive that no rational relationship can be detected.”); Romer v. Evans, 517 U.S. 620, 633, 116 S Ct 1620, 134 L Ed. 2d 855 (1996) (striking law down under rational basis review because it was “at once too narrow and too broad”); see generally Ronald D. Rotunda and John E. Nowak, 3 Treatise on Constitutional Law—Substance & Procedure § 18.2(b) (Westlaw July 2024 update) (discussing rational basis review of overinclusive and underinclusive laws).

b. Mathias and “genuine differences”

The Tax Court's analysis appears to have rested on two parts of this court's opinion in Mathias, 312 Or. at 50, 817 P.2d 272. The first was Mathias’s reference to “genuine differences” as an aspect of the rational basis test. See, e.g., id. at 60, 62-63, 817 P.2d 272. The second was Mathias’s suggestion that, in property tax classifications, courts may consider only differences in property, and not in who owns or uses the property. See id. at 60, 62-63, 817 P.2d 272. Based on those parts of Mathias, the Tax Court functionally understood the “genuine differences” test to overshadow—if not substitute for—the rational basis test. In other words, as the Tax Court approached the analysis, the department was required to first identify “genuine” differences between the intangible property of centrally assessed businesses and locally assessed businesses. See 25 OTR at 322 (concluding that the analysis was a two-step process beginning with whether there were “genuine” differences). If there were no “genuine” differences between the intangible property of those business classes, or in how those business classes used their intangible property, then the tax would be unconstitutional—without further need to consider whether the tax was rationally related to a legitimate governmental purpose. See id. at 353 (concluding that statute was unconstitutional because there were no “genuine” differences in how different industries used intangible property).

We acknowledge the difficulties that Mathias presents to any court attempting to discern its rationale. This case, however, does not require us to revisit all the details of Mathias’s reasoning or result. Instead, we address only the two points that, in our view, misdirected the Tax Court's analysis.

Mathias considered the constitutionality of a property tax statute that applied to subdivision lots. In general, all properties were assessed at their market value. 312 Or. at 52, 817 P.2d 272. Under the statute, however, a different valuation method was used for some lots, depending on how many other lots the taxpayer owned. If one person owned four or more lots in a subdivision, then those lots would be valued using a method that “ ‘recognizes the time period over which those lots must be sold in order to realize current market prices for those lots.’ ” Id. (quoting ORS 308.205(3) (1989)). The taxpayers, who owned fewer than four lots, argued that it was unconstitutional to value their property higher than identical adjacent lots in the same subdivision. Id. Mathias ultimately agreed with the taxpayers. Id. at 67, 817 P.2d 272.

The first point discussed in Mathias, “genuine differences,” was drawn from Jarvill and Huckaba. See id. at 59, 817 P.2d 272 (quoting both decisions). Those cases show, however, that this court's references to “genuine differences” were not intended as an independently meaningful concept but were merely a restated part of the rational basis test. A classification is not valid merely because some difference between the classes exists; those differences must somehow relate to the governmental purpose. See Jarvill, 289 Or. at 180, 613 P.2d 1 (attributing source of “genuine differences” to Huckaba); Huckaba, 281 Or. at 25-26, 573 P.2d 305 (discussing principle of federal equal protection law); see also Metropolitan Co. v. Brownell, 294 U.S. 580, 583, 55 S Ct 538, 79 L Ed. 1070, reh'g den., 295 U.S. 767, 55 S Ct 647, 79 L Ed. 1708 (1935) (question is not whether classes are different, “but whether the differences between them are pertinent to the subject with respect to which the classification is made”; if “those differences have any rational relationship to the legislative command, the discrimination is not forbidden” (citations omitted)); Power Co. v. Saunders, 274 U.S. 490, 493-94, 47 S Ct 678, 71 L Ed. 1165 (1927) (equal protection requires “that the classification be not arbitrary but based on a real and substantial difference having a reasonable relation to the subject of the particular legislation”; a valid classification “must rest on differences pertinent to the subject in respect of which the classification is made”). “Genuine differences” thus is not a substitute for the rational basis test. Instead, it merely emphasizes that the rational basis test requires more than identifying some difference between the classes.

The second point addressed in Mathias was the extent to which a classification for property taxes required courts to narrowly consider only differences in the classes of property being taxed, as opposed to differences in ownership. That appears to us to be, at least in part, a function of the order in which the opinion addressed the department's proposed reasons for the classification.

After reviewing the constitutional principles, Mathias had stated in passing that, in the context of ad valorem property taxes, classifications must be based on genuine differences “between the classes of property” that were being treated differently. 312 Or. at 60, 817 P.2d 272 (emphasis added). The court then rejected the department's initial contention that there were differences in the property being classified for purposes of the tax. Id. at 60-65, 817 P.2d 272. The court explained that the particular distinguishing characteristic used by the legislature—“[t]he amount of other property that a taxpayer owns”—was not rationally related to the characteristics of the property. Id. at 62, 817 P.2d 272 (adding a footnote identifying two cases that had struck down classifications based purely on characteristics of the owner). Arguably, that statement suggests that characteristics of ownership are not pertinent.

Those statements earlier in the opinion, however, must be considered in the context of later parts of the analysis. Mathias went on to address an “alternative” basis for the classification proposed by the department: an intent “to encourage investment and effort by those engaged in subdividing land by providing them a tax incentive or subsidy.” Id. at 65, 817 P.2d 272. If the court had intended its earlier statements to mean that property tax classifications can be justified only by differences in the nature or use of property, then that alternative proposed basis for the classification would have already been refuted. Instead, the court concluded that the characteristics of the owner (that is, how many other properties he owned) were not rationally related to the particular classification being challenged. The department had proposed that the justification was to encourage developers, but “the benefits of the statute [were] not limited to subdividers or developers or to property being developed.” Id. Moreover, the court explained, the valuation of multiple lots was not limited to “such ownerships still held by the original subdivider or to bulk sales of subdivision lots.” Id. The court did not find any indication that the legislature intended to create a tax exemption “from which only owners of four or more lots would benefit.” Id. The statutory text, context, and legislative history did not support the conclusion that the legislature had intended to treat taxpayers differently. See id. at 66, 817 P.2d 272 (text has “no indication that the legislature intended to require other taxpayers to pay a part of a subdivider's property taxes”); id. (“tax benefits for individual landowners” were not being considered). The court concluded:

“The argument that the legislature intended partial exemption or subsidy for those owning four or more lots at the expense of other property taxpayers, including those owning fewer lots, is not supportable.”

Id. at 67, 817 P.2d 272.

Again, if Mathias had meant to hold that the characteristics of a property owner can never be considered in evaluating the rational basis for a property tax, then the court would have had no need to even discuss the proffered alternative justification. The statute would have been unconstitutional even if the legislature had expressly limited the valuation to developers and had expressly intended it to be a partial tax exemption. Properly understood, therefore, Mathias narrowly held that the characteristics of the owner were, in that particular instance, insufficient to justify the classification method used to value some subdivision lots differently.

That narrow reading is supported by our prior case law concluding that a tax exemption for property held by charitable organizations was rationally related to a legitimate governmental purpose, and thus constitutional. Corporation of Sisters of Mercy, 123 Or. at 161-64, 261 P. 694; see Carmichael, 301 U.S. at 512, 57 S.Ct. 868 (noting that equal protection permits “the exemption of charitable institutions” from taxation).

Having explained what we understand to be the proper reading of Mathias, we have no further reason to consider whether that case correctly applied those principles.¹⁶ Insofar as this case is concerned, however, Mathias does not establish that “genuine differences” must be identified as a preliminary step before applying the rational basis test. Nor does Mathias require that classifications in the property tax context must always and only be justified by reference to the nature or use of the property.

c. Application

The Tax Court held that the relevant “classification” in this case is between intangible property used by a centrally assessed business and intangible property used by a locally assessed business. See 25 OTR at 322 (concluding that classes were “(1) intangible property used in a business listed in ORS 308.515(1) (taxable), and (2) all other intangible property (not taxable)”). The parties largely do not dispute that classification, and it is sufficient for purposes of this opinion. We point out only that, while intangible property is the thing being taxed differently, it is not something that distinguishes the classes: intangible property is something that both classes (centrally assessed businesses and locally assessed businesses) have in common.

As noted earlier, the Tax Court had concluded that Mathias required a prerequisite showing of “genuine differences” between the intangible property held by centrally assessed businesses and the intangible property held by locally assessed businesses, either in the nature of that property or how it is used. See id. at 325-26 (discussing Mathias). We have explained, however, that a showing of “genuine differences” is not a prerequisite to rational basis review. If there are differences between the classes that are rationally related to a legitimate governmental purpose, then those differences are “genuine differences.”

Moreover, as we also have explained, rational basis review is not limited to differences in the nature of the intangible property held by those businesses, or in how those businesses use that property. The rational basis test does not require the legislature to put on such limiting blinders when making policy choices about taxation.

We turn, then, to how the rational basis test applies here. As we will explain, we conclude that the legislature's classifications are rationally related to legitimate governmental purposes. There are rational reasons why the legislature might legitimately choose to (1) tax intangible property; (2) tax the intangible property of centrally assessed businesses; and (3) not tax the intangible property of locally assessed businesses.

Axiomatically, obtaining revenue is a legitimate governmental purpose. Moreover, the taxation of intangible property is rationally related to the purpose of obtaining revenue. That proposition seems self-evidently true, nor do we understand taxpayers to contend otherwise.

Furthermore, as we have explained in some detail, unit valuation and central assessment are intertwined. They developed specifically to permit states to tax the intangible property of certain types of businesses. In the judgment of the legislature, those businesses have substantial amounts of their value bound up in such intangible property as goodwill, and the nature of those businesses—often distributed across multiple taxing jurisdictions—makes it particularly likely that that value would escape taxation. Taxpayers are parts of industries very similar to railroads, the classic centrally assessed business: railroads, airlines, and electrical utilities are all heavily regulated industries that use specific and limited corridors to transport things within and without the state (railroads mainly transport cargo; Delta mainly transports people; PacifiCorp transports electricity).

We also conclude that there are conceivable, rational reasons why the legislature would choose to limit the tax on intangible property to centrally assessed businesses.¹⁷

To begin with, the legislature might have intended to promote efficiency and fairness in taxation by having a statewide agency develop expertise in valuing the intangible property of some or all of the listed businesses, most of which are specialized in nature and often subject to extensive governmental regulation. But because the Department of Revenue has limited resources, the legislature may reasonably have concluded that central assessment and the associated tax on intangible property should be limited to those industries that offer the highest potential revenue return to Oregon.

Relatedly, different industries may vary both in how difficult it is to assess their intangible property and in how much additional revenue would be generated by doing so. The legislature could rationally balance those considerations in choosing certain industries to be centrally assessed and taxed on their intangible property. Further, the legislature might consider the administrative cost imposed on taxpayers who are required to report the value of their intangible property, and then selected those industries that—in the view of the legislature—could best bear that expense. All those considerations would rationally serve legitimate purposes.

Beyond all that, the legislature may conceivably have had independent reasons not to tax the intangible property of various other businesses.

For example, the legislature may have determined that ordinary property taxes come close enough to capturing the value of what the legislature may have considered to be businesses more heavily invested in capital or physical products—manufacturing, merchants, or the bus and trucking companies pointed to by Delta. For that reason, the legislature might have declined to add the extra complication of taxing their intangible property.

As for bus and trucking companies specifically, the department has noted that bus and trucking companies are already subject to a different taxation scheme that is prescribed in part by the Oregon Constitution. See Or. Const., Art. IX, § 3a (tax revenues obtained by taxing motor vehicle fuel, or use or ownership of motor vehicles, must be put toward construction and maintenance of public roads and related works). The legislature could rationally consider the entirety of the tax system applicable to the bus and trucking industries when making policy decisions about the individual parts of that system—such as whether to tax the intangible property of that industry.¹⁸

Or the legislature could have decided not to tax the intangible property of certain industries on the ground that the revenue obtained from taxing that industry as a whole was not worth the cost. That is, the legislature might have believed that the total revenue gained by taxing the intangible property of all businesses within a particular industry might, on average, fail to offset the total cost to the state of assessing the intangible property of all the businesses in that industry.

Or the legislature could have concluded that a particular industry faced additional regulatory burdens, or taxes, or competition, that made an additional tax burden inappropriate as a matter of public policy.

In identifying those conceivable purposes, we are simply echoing similar conceivable reasons that the United States Supreme Court recognized and approved in Carmichael, when it held that Alabama might lawfully restrict its tax for unemployment insurance to those employers having eight or more employees while exempting certain types of employers. The Court first explained that “[a]dministrative convenience and expense in the collection or measurement of the tax are alone a sufficient justification for the difference[.]” 301 U.S. at 511, 57 S.Ct. 868. The state legislature might have concluded that “the expense and inconvenience of collecting the tax from small employers” would be “disproportionate to the revenue obtained,” id.; “that generally the number of employees bears a relationship to the size of the payroll and therefore to the amount of the tax,” id.; and that “the large number of small employers and the paucity of their records of employment would entail greater inconvenience in the collection and verification of the tax than in the case of larger employers,” id. Similarly, “[r]elatively great expense and inconvenience of collection may justify the exemption from taxation of domestic employers, farmers, and family businesses, not likely to maintain adequate employment records, which are an important aid in the collection and verification of the tax.” Id. at 513, 57 S Ct 868. Furthermore, the Court explained, reasons of public policy would also justify the legislature in deciding to exclude particular industries: the legislature may “withhold the burden of the tax in order to foster what it conceives to be a beneficent enterprise” or “to aid a depressed industry such as shipping.” Id. at 512, 57 S Ct 868.

Neither equal protection nor equal privileges and immunities require the legislature to draw “mathematically exact” lines. Dominion Hotel, 249 U.S. at 268, 39 S.Ct. 273; see Mallatt, 206 Or. at 702, 294 P.2d 871 (“mathematical nicety” not required). In deciding to tax the intangible property of centrally assessed businesses, the legislature did not need to identify and weigh a set of controlling criteria, then “spreadsheet” every industry in the entire economy, accurately placing each on the proper side of the line, on penalty of any error rendering the entire classification invalid. Any such requirement would indeed convert equal protection and equal privileges into “a pedagogical requirement of the impracticable.” Dominion Hotel, 249 U.S. at 268, 39 S.Ct. 273; Mallatt, 206 Or. at 702, 294 P.2d 871.

2. Federal Equal Protection Clause

We turn now to the Equal Protection Clause of the United States Constitution. As long ago as 1890, the United States Supreme Court noted the leniency granted the states in making classifications for tax purposes:

“[A state] may, if it chooses, exempt certain classes of property from any taxation at all, such as churches, libraries and the property of charitable institutions. It may impose different specific taxes upon different trades and professions, and may vary the rates of excise upon various products; it may tax real estate and personal property in a different manner; it may tax visible property only, and not tax securities for payment of money; it may allow deductions for indebtedness, or not allow them.”

Bell's Gap R'd Co. v. Pennsylvania, 134 U.S. 232, 237, 10 S Ct 533, 33 L Ed. 892 (1890). The Court continues to emphasize the breadth of permissible classifications. See, e.g., Nordlinger v. Hahn, 505 U.S. 1, 11-12, 112 S Ct 2326, 120 L Ed. 2d 1 (1992) (equal protection standard “is especially deferential in the context of classifications made by complex tax laws”; “the States have large leeway in making classifications and drawing lines which in their judgment produce reasonable systems of taxation” (internal quotation marks and citations omitted)).

Beyond that, we see little to add to what we have already said. Our analysis of Oregon's Equal Privileges and Immunities Clause had cited numerous United States Supreme Court decisions for their persuasive authority (and in fact many of our older cases relied heavily on equal protection principles). The decisions of the Court are controlling in the context of the federal Equal Protection Clause, but the fundamental analysis under the federal standard is very similar to that which applies under state law. All the points that led to our conclusion that the classification is rational for purposes of equal privileges and immunities also show that the classification is rational for purposes of equal protection.

We would add that the primary arguments made here—that property cannot be classed based purely on who holds or uses it, and that some businesses in the state cannot be taxed on their intangible property when others are not—were rejected by the United States Supreme Court in the nineteenth century. In the Adams Express cases, express companies had contended that it violated equal protection for the State of Ohio to tax their intangible property such as goodwill, when other businesses in Ohio were not subject to such a tax. The express companies’ arguments, which are printed only in the United States Reports and the Lawyer's Edition, contended that the law violated equal protection because

“[p]roperty cannot be classified in respect to mere ownership. The same kind and character of property devoted to the same uses, within the same taxing districts, cannot be taxed by one rule against one class of persons and by a different rule against another class.”

Adams Express Company v. Ohio, 165 U.S. 194, 206, 17 S Ct 305, 41 L Ed. 683 (1897) (Adams Express I); see id. (arguing that “the property owned by express companies within the State of Ohio is not different in its character, uses or situations from other similar property within the State, nor is there any greater difficulty in ascertaining its value for purposes of taxation”); Adams Express II, 166 U.S. at 209-10, 17 S.Ct. 604 (on rehearing, express companies asserted that law violated equal protection because it taxed express companies differently from “[m]erchants, manufacturers, banks, brokers, newspapers, gas companies, street railway companies, indeed all persons and corporations engaged in business in the State”).

The Court rejected those contentions, holding that Ohio's tax did not violate the Equal Protection Clause. See Adams Express I, 165 U.S. at 221-22, 17 S.Ct. 305 (the property of the express companies, “whether represented in tangible or intangible property, * * * possessed a value in combination * * * which could as rightfully be recognized in the assessment for taxation” as it could be for railroad, telegraph, and sleeping-car companies); id. at 228-29, 17 S Ct 305 (rejecting equal protection argument); Adams Express II, 166 U.S. at 225, 17 S.Ct. 604 (denying petition for rehearing without further addressing equal protection). Delta's similar contentions do not appear more persuasive today than they were when the Court decided Adams Express.¹⁹

Finally, we return to PacifiCorp's argument regarding the Ninth Circuit's holding in BNSF. PacifiCorp asserts that the holding in BNSF amounts to a determination by a federal court that the classification at issue here is not rationally related to a legitimate governmental purpose. The Ninth Circuit, however, was not applying the rational basis test; it was applying a more restrictive test prescribed by statute. As we mentioned previously, BNSF was interpreting a federal statute, 49 USC § 11501(b)(4), that prohibited states from taxing railroads differently from other commercial and industrial taxpayers. 965 F.3d at 684; id. at 686; id. at 691-93. The rational basis test for equal protection uses a substantially less restrictive test, as the United States Supreme Court explained in Alabama Dept. of Revenue v. CSX Transp., 575 U.S. 21, 27-28, 135 S Ct 1136, 191 L Ed. 2d 113 (2015) (noting that construing 49 USC § 11501(b)(4) to use the looser equal protection standard would “deprive [the statute] of all real-world effect, providing protection that the Equal Protection Clause already provides”). The Court emphasized how forgiving the test for equal protection was:

“In the Equal Protection Clause context, very few taxpayers are regarded as similarly situated and thus entitled to equal treatment. There, a State may tax different lines of businesses differently with near-impunity, even if they are apparently similar.”

Id. at 27-28, 135 S.Ct. 1136. As a result, nothing the Court said in BNSF undermines the conclusions that we reach in this case.

III. CONCLUSION

The tax at issue here is rationally related to any number of legitimate purposes. Our analysis under Oregon's Equal Privileges and Immunities Clause shows that the tax is valid. Moreover, the tax also is constitutional under the Equal Protection Clause. The tax presents no additional issues under the uniformity provisions of the Oregon Constitution.

The judgment of the Tax Court is reversed, and the case is remanded to the Tax Court for further proceedings.

I concur in the result and the reasoning of the majority. I write separately, however, to address an issue that I perceive as an essential predicate step of our analysis under Article I, section 20, of the Oregon Constitution: whether a “true class” exists. The majority omits that step because the parties did not address it in their briefing; the majority therefore assumes it, without deciding it. I would have preferred that we address it.

Article I, section 20, prohibits granting privileges or immunities to one citizen or class of citizens that are not equally available to all citizens. As we explained in State v. Clark, 291 Or. 231, 237, 630 P.2d 810, cert. den., 454 U.S. 1084, 102 S.Ct. 640, 70 L.Ed.2d 619 (1981), the clause “forbids inequality of privileges or immunities not available ‘upon the same terms,’ first, to any citizen, and second, to any class of citizens.” Id. at 237, 630 P.2d 810 (quoting Article I, section 20).

“Class” is a term of art for Article I, section 20, analysis. Disparate treatment implicates Article I, section 20, only when it involves a “true class.” State ex rel Huddleston v. Sawyer, 324 Or. 597, 610, 932 P.2d 1145, cert. den., 522 U.S. 994, 118 S.Ct. 557, 139 L.Ed.2d 399 (1997). In attempting to describe precisely what is meant by a “true class,” our cases draw a distinction between classes that are created by the challenged law or government action itself and classes that are defined in terms of characteristics that are shared apart from the challenged law or action.

We have referred to “true classes” as denoted by “ad hominem characteristic[s],” Van Wormer v. City of Salem, 309 Or. 404, 408, 788 P.2d 443 (1990), by “personal characteristic[s],” Zockert v. Fanning, 310 Or. 514, 523, 800 P.2d 773 (1990), or by “antecedent personal or social characteristics or societal status,” Hale v. Port of Portland, 308 Or. 508, 525, 783 P.2d 506 (1989), abrogated on other grounds by Smothers v. Gresham Transfer, Inc., 332 Or. 83, 23 P.3d 333 (2001). We have described “true class” characteristics as (not exclusively) “sex, ethnic background, legitimacy, past or present residency or military service.” Clark, 291 Or. at 240-41, 630 P.2d 810.

In contrast, we have recognized that

“every law itself can be said to ‘classify’ what it covers from what it excludes. For instance, the rule of this court that limits the time for filing a petition for review * * * ‘classifies’ persons by offering the ‘privilege’ of review to those who file within 30 days and denying it to those who file later.”

Id. at 240, 630 P.2d 810. Those types of classes do not invoke Article I, section 20, at all. Huddleston, 324 Or. at 610, 932 P.2d 1145. As we explained, “[a]ttacks on such laws as ‘class legislation’ therefore tend to be circular and * * * have generally been rejected whenever the law leaves it open to anyone to bring himself or herself within the favored class on equal terms.” Clark, 291 Or. at 241, 630 P.2d 810. Later commentary framed the question as whether the classification represents a “true class” or a “pseudo-class,” with pseudo-classes being outside the scope of the state's equal privileges clause. David Schuman, The Right to Equal Privileges and Immunities: A State's Version of Equal Protection, 13 Vt L Rev 221, 232-33 (1988).¹

That “true class” distinction is an essential predicate step of an Article I, section 20, analysis. As we said in Sealey v. Hicks, 309 Or. 387, 397, 788 P.2d 435, cert. den., 498 U.S. 819, 111 S.Ct. 65, 112 L.Ed.2d 39 (1990),

“[i]n evaluating whether a class exists under Article I, section 20, we must first determine whether the class ‘is created by the challenged law itself’ or ‘by virtue of characteristics * * * apart from the law in question.’ [Clark, 291 Or. at 240, 630 P.2d 810]. Classes of the first type are entitled to no special protection and, in fact, are not even considered to be classes for the purposes of Article I, section 20.”

(Emphasis added, ellipses in original.)

I readily acknowledge that we have not articulated the clearest test to assist litigants, or lower courts, in sifting true classes from non-true classes. The Court of Appeals has expressed “frustration” with its inability to identify the principles that would harmonize our decisions on the point. See Neher v. Chartier, 124 Or App 220, 225 n. 3, 862 P.2d 1307 (1993), rev'd on other grounds, 319 Or. 417, 879 P.2d 156 (1994).² That difficulty may well explain why the parties have ignored the issue. But to me, that is all the more reason for us to engage with that aspect of our Article I, section 20, analysis, and perhaps offer clarity.

In the evidentiary context, I recently wrote of my growing unease with our “worn-threadbare tactic of assuming without deciding.” State v. Taylor, 372 Or. 536, 557, 551 P.3d 924 (2024) (James, J., concurring). My unease is particularly acute in the area of state constitutional law. Unlike the majority of states that employ a lockstep or interstitial approach to constitutional questions, Oregon employs an independent state constitutional model. As we have explained, our “first things first” doctrine requires a court to consider state law claims first because “the state does not deny any right claimed under the federal Constitution when the claim before the court in fact is fully met by state law.” Sterling v. Cupp, 290 Or. 611, 614, 625 P.2d 123 (1981).

Our “first things first” approach yields many benefits, but two in particular I highlight. First, rigorous adherence to a full state constitutional analysis is responsible for Oregon's independent protection of civil rights and liberties above the federal floor. See, e.g., State v. McCarthy, 369 Or. 129, 153-54, 501 P.3d 478 (2021) (explaining that this court has rejected federal law construing the Fourth Amendment as a “reasonable expectations of privacy” test because “ ‘the privacy protected by Article I, section 9, is not the privacy one reasonably expects but the privacy to which one has a right’ ” (quoting State v. Campbell, 306 Or. 157, 164, 759 P.2d 1040 (1988) (emphasis in Campbell)); State v. Dixson, 307 Or. 195, 766 P.2d 1015 (1988) (concluding that unlike the Fourth Amendment, the protections of Article I, section 9, could, in some circumstances, extend beyond the curtilage of a person's home); State v. Henry, 302 Or. 510, 515, 732 P.2d 9 (1987) (recognizing that the protections Article I, section 8, provides to Oregonians are “broader” than the First Amendment to the United States Constitution.)

Second, as former Justice Carson noted, “reliance on state law [helps to] eliminate the practical consequences of the troublesome effects of shifts in the United States Supreme Court's interpretation of the federal constitution.” Wallace P. Carson, Jr., “Last Things Last”: A Methodological Approach to Legal Arguments in State Courts, 19 Willamette L Rev 641, 648-49 (1983); see also State v. Caraher, 293 Or. 741, 750, 653 P.2d 942 (1982) (“The resolution of this case under the federal analysis provides no guidance for the next. * * * The goal of simplification is, in our view, better served by relying on Article I, section 9 of our own Constitution * * * than by hypothesizing how the U.S. Supreme Court would consider this case in light of its past decisions and then deciding whether to adopt that rule.” (Footnote omitted)). In short, rigorously applied state constitutionalism is a breakwater against rough federal seas.

As former Justice Linde cautioned, “when a court ties a state constitutional guarantee as a tail to the kite of the corresponding federal clause, it may simply find the state ground ignored on certiorari[.] * * * [T]he habit that developed in the 1960's of making a federal case of every claim and looking for all law in Supreme Court opinions dies hard.” Hans Linde, First Things First: Rediscovering the States’ Bill of Rights, 9 U of Baltimore L Rev 379, 390 (1980).

Even when our state constitutional analysis differs from federal constitutional analysis, consistently addressing and applying those differences may be important to establishing a claim of independent adequate state grounds. When analysis of the state constitutional ground is tied too closely to related federal constitutional grounds, federal courts might not treat the state constitutional ground as independent. As the United States Supreme Court cautioned in Michigan v. Long:

“Apart from its two citations to the state constitution, the court below relied exclusively on its understanding of [Terry v. Ohio, 392 U.S. 1, 88 S Ct 1868, 20 L Ed. 2d 889 (1968),] and other federal cases. Not a single state case was cited to support the state court's holding that the search of the passenger compartment was unconstitutional. Indeed, the court declared that the search in this case was unconstitutional because ‘[t]he Court of Appeals erroneously applied the principles of Terry * * * to the search of the interior of the vehicle in this case.’ [Michigan v. Long, 413 Mich. 461, 471, 320 N.W.2d 866, 869 (1982)]. The references to the state constitution in no way indicate that the decision below rested on grounds in any way independent from the state court's interpretation of federal law. Even if we accept that the Michigan constitution has been interpreted to provide independent protection for certain rights also secured under the Fourth Amendment, it fairly appears in this case that the Michigan Supreme Court rested its decision primarily on federal law.”

463 U.S. 1032, 1043-44, 103 S Ct 3469, 77 L Ed. 2d 1201 (1983) (emphases in original).

In espousing first things first, Justice Linde related two anecdotes, that I recount here in full:

“[L]awyers once came to our court trying to fit a woman's right to operate a day care center within the due process analysis of Goldberg v. Kelly[, 397 U.S. 254, 90 S Ct 1011, 25 L Ed. 2d 287 (1970)]. Only after the argument did our own examination show that she was entitled to prevail under the state administrative procedure act, which counsel apparently had not read.

“In another recent case, a defendant charged with speeding demanded the maintenance records of the radar sets used by the Portland police. The case was argued below and in our court as a federal due process claim under the Supreme Court's rule in Brady v. Maryland, [373 U.S. 83, 83 S Ct 1194, 10 L Ed. 2d 215 (1963),] which dealt with prosecution suppression of evidence favorable to the defense. After the argument, it occurred to us that the records were apparently available to anyone on request under the state's public records law. We asked the parties for additional memoranda. The state agreed that the records were, indeed, available; all defendant had to do was pick them up for himself instead of demanding that the district attorney get them for him. Defense counsel's response was that this might be so, but that the ‘threshold question’ in the case was the duty of the state in light of the United States constitutional standards; the merits of other legal ways of getting the information were irrelevant.

“As I say, that perspective dies hard. In both of these cases, we happened to notice on our own motion that the state law protected the interest at stake. But if we had not noticed the state issues on our own, the shaky federal due process claims might well have failed.

“What this should mean in practice is simply good lawyering, both at the bar and on the bench. If we expect to get careful attention to a state statute, we should give equally careful attention to the state constitution.”

Linde, 9 U of Baltimore L Rev at 390-91.

I share Justice Linde's perspective—preserving the independence of our Oregon Constitutional approach is a community effort. We all must do our part. Lawyers must address the state constitutional claim, and do so fully, giving each step attention and analysis. And courts, likewise, must do their part, resisting attempts by litigants to short circuit the state constitution, or an aspect of our state constitutional analysis. And, in my view, appellate courts should be very reticent to omit consideration of the state constitution, or an aspect of our state constitutional analysis, even when the parties have failed to brief it. To that end, I call upon lawyers to be mindful of addressing all the steps of our state constitutional analysis, and I call upon this court to resume our historical practice of asking for supplemental briefing from parties who fail to address the state constitution, or a state constitutional step, rather than continuing our trend of assuming without deciding such matters.

However, in this case, because the parties have not briefed the issue of whether the tax law affects a true class, and because this court did not require supplemental briefing by the parties to address that missing analysis, I will reserve my resolution of that question for a future case. I am therefore able to concur in the rest of the reasoning of the majority and join its result.

I respectfully concur.

FOOTNOTES

1.   Although the case is captioned for Delta alone, we explain the constitutional standard and address the arguments made by both Delta and PacifiCorp. Those taxpayers have appealed from cases that were not consolidated below, the Tax Court reached different conclusions as to the different taxpayers, and the taxpayers present somewhat different arguments regarding constitutionality here. Accordingly, we follow the Tax Court's lead and write a single opinion regarding the constitutional questions. We will issue a separate opinion that incorporates our resolution here and otherwise addresses the department's cross-appeal, which makes an unrelated challenge to the Tax Court's valuation holding regarding the 2020-21 tax year. See PacifiCorp v. Dept. of Rev., 25 OTR 227, 2023 WL 4571446 (2023) (addressing valuation issue).

2.   ORS 308.515(1) identifies the following centrally assessed business types:“(a) Railroad transportation;“(b) Railroad switching and terminal;“(c) Electric rail transportation;“(d) Private railcar transportation;“(e) Air transportation;“(f) Water transportation upon inland water of the State of Oregon;“(g) Air or railway express;“(h) Communication;“(i) Heating;“(j) Gas;“(k) Electricity;“(L) Pipeline;“(m) Toll bridge; or“(n) Private railcars of all companies not otherwise listed in this subsection, if the private railcars are rented, leased or used in railroad transportation for hire.”

3.   ORS 307.020 defines “intangible personal property” fairly precisely, but since 1977 that definition no longer applies to the central assessment statutes. ORS 307.020(1)(a) (defining “intangible personal property”); ORS 307.020(2) (definition does not apply to central assessment statutes); see Or. Laws 1977, ch. 602, § 1 (amending prior version of statute to make inapplicable to central assessment).

4.   Those broad outlines of intangible property are sufficient for purposes of our decision. Note, however, that the legislature has excluded some types of intangible property from central assessment. See ORS 308.505(14)(c) (intangible property excludes “[c]laims on other property, including money at interest, bonds, notes, claims, demands or any other evidence of indebtedness, secured or unsecured” and “[a]ny shares of stock in corporations, joint stock companies or associations”). The Tax Court's opinion also identified other types of intangible property that are not taxed for various reasons. See Delta, 25 OTR at 331 n. 25.

5.   Broadly speaking, “going concern” valuation “considers a company's market value as a whole and does not, either in practice or in theory, purport to assess the various component parts that go into that whole.” DISH Network Corp. v. Dept. of Rev., 364 Or. 254, 292, 434 P.3d 379 (2019). Cf. OAR 150-308-0260(2) (in valuing industrial property, “[t]he going concern concept recognizes that the value of an assembled and operational group of assets usually exceeds the value of an identical group of assets that are separate or not operational”).

6.   Some authorities use the term to refer not just to the valuation of a business as a going concern, but also to the apportionment of the appropriate share of that valuation to the taxing state. See K.E. Powell, Identifying Exempt Intangible Assets in State Property Tax: Urging Stricter Application of Burden of Proof, 30 Cornell J L & Pub Pol'y 235, 242 (2020); James C. Bonbright, 2 The Valuation of Property 633 n. 1 (1937).

7.   We additionally note that listing specific business types instead of abstract criteria may help remove initial questions—and potential litigation—over which taxing body should be assessing a particular business. It thus promotes certainty for taxpayers, local assessors, and the department alike.

8.   The Tax Court consolidated Delta's appeals for a number of other tax years. See 25 OTR at 309 n 2.

9.   We recognized in Jarvill that the provisions are not identical. Article IX, section 1, does not expressly require territorial uniformity, and it appears limited to statewide taxes. 289 Or. at 171 n 15, 613 P.2d 1. The parties here do not rely on those textual differences, and we need not address them here.

10.   Specifically, the relevant part of the 1859 text read, “[A]ll taxation shall be equal and uniform.”

11.   The 1859 text read:“The Legislative Assembly shall provide by law for a uniform and equal rate of assessment and taxation; and shall prescribe such regulations as shall secure a just valuation for taxation of all property, both real and personal, excepting such only for municipal, educational, literary, scientific, religious, or charitable purposes as may be specially exempted by law.”

12.   The prohibition on discriminatory enforcement under the uniformity provisions seems functionally identical to the similar prohibitions imposed by equal protection and equal privileges and immunities. See Penn Phillips, 247 Or. at 385-86, 430 P.2d 349 (explaining that “[a]rbitrary or systematic discrimination in assessment” not only violates equal protection and equal privileges and immunities, but it “also offends the uniformity clauses of our own constitution”).

13.   Our prior cases have indicated that a classification defined by the statute itself is not a “true class,” and that such classifications do not violate Article I, section 20, for that reason alone. Because the parties agree that the classification here is subject to rational basis review, and we conclude that it satisfies that test, we need not decide in this case whether the classification creates a “true class” for purposes of Article I, section 20.

14.   See, e.g., Knapp v. City of Jacksonville, 342 Or. 268, 276, 151 P.3d 143 (2007) (upholding differing surcharges for developed property because of something the city “could have” concluded (internal quotation marks and citation omitted)); People's Util. Dist. et al v. Wasco Co. et al, 210 Or. 1, 22, 305 P.2d 766 (1957) (upholding tax statute based on “possible bases for the enactments”); State v. Kozer, 116 Or. 581, 588-90, 242 P. 621 (1926) (upholding classification because the legislature “might well” have classified commercial vehicles, based on a policy “apparent” from the motor vehicle statutes).

15.   In Huckaba, the legislature taxed retirement benefits for military service differently from retirement benefits for other civil services, because military servicepeople could retire earlier and thus might be able to start a new career. The plaintiff argued that the distinction failed rational basis review because some civil servants also could retire early. Rejecting the argument, this court explained:“General rules are essential if a system of the magnitude and complexity of the Personal Income Tax Act is to be administered with a modicum of efficiency, even though application of the rule may produce seemingly arbitrary consequences in some cases. A nonmilitary federal retiree may, in fact, after retirement obtain employment and create an additional retirement fund. Or conversely an Armed Forces retiree may be unable to enter a new career and be required to subsist on his military retirement pay. Making these determinations would require individualized proof as each income exclusion was claimed. The legislature could reasonably choose between a system of individualized inquiry and a general rule based on the source of the retirement benefit. The former method would introduce complexities in the administration of an already complex tax system and increase the expense of administration. The choice between these competing policies is a legislative determination and the decision to accord the benefit on the basis of an easily ascertainable criterion does not offend constitutional principles.”Huckaba, 281 Or. at 30-31, 573 P.2d 305.

16.   There are reasons to question whether it did. Among other things, Mathias asserted that land use statutes and regulations cannot support a constitutional classification. 312 Or. at 61, 817 P.2d 272. Jarvill had expressly held that they could. 289 Or. at 180-81, 613 P.2d 1. Mathias’s additional assertion that “[t]he amount of other property that a taxpayer owns is not a rational basis for distinguishing between otherwise identical lots for tax purposes,” 312 Or. at 62, 817 P.2d 272, would seem to contradict the long-standing precedent allowing taxes to be graduated based precisely on such numerical amounts as income.

17.   The Tax Court appears not to have reached that question, having had held that there were no “genuine differences” in the nature or use of intangible property. See 25 OTR at 342-51 (considering whether there were “genuine differences,” and ultimately concluding that there were not).

18.   The Tax Court rejected the department's argument regarding the separate taxation scheme applicable to motor vehicles. See 25 OTR at 343-45. The court's reasons for doing so appear to have depended heavily on its having understood Mathias to mean that the legislature, when deciding which businesses to tax on intangible property, could not give any consideration to how bus and trucking companies were subject to a separate tax system for tangible property (motor vehicles). We have now clarified that Mathias does not go so far.

19.   Ten years ago, the Court suggested that the same principles remain valid. In a case focused on a different issue, the Court offhandedly rejected (in dictum) the claim that it would be unconstitutional to tax a railroad differently from a motor carrier:“It would be permissible—as far as the Equal Protection Clause is concerned—for a State to tax a rail carrier more than a motor carrier, despite the seeming similarity in their lines of business.”Alabama Dept. of Revenue v. CSX Transp., 575 U.S. 21, 28, 135 S Ct 1136, 191 L Ed. 2d 113 (2015). It would be difficult to understand why equal protection would permit different treatment for railroads and motor carriers, but prohibit different treatment for airlines and motor carriers.

1.   Professor Schuman suggested that a “pseudo-class” was “any group of people who would never have conceived of themselves as a ‘class,’ and would not have been treated as a class, if the statute or policy allegedly disadvantaging them did not exist.” Schuman, 13 Vt L Rev at 233.

2.   In that footnote in Neher, the Court of Appeals gave the following review of the case law as it then existed:“We share plaintiff's frustration in attempting to discern the correct analysis for Article I, section 20, challenges. The [Oregon] Supreme Court's opinions have been inconsistent regarding what is a ‘true class.’ In [Clark, 291 Or. at 240, 630 P.2d 810], the court said that classes are based on personal characteristics that persons or groups have apart from the law itself, such as sex, ethnic background, legitimacy, residency or military service. In [Hale, 308 Or. at 525, 783 P.2d 506], the court applied a similar standard to hold that a classification of victims of governmental torts is not an identifiable class, because it is not based on ‘antecedent personal or social characteristics or societal status.’ However, in [Sealey, 309 Or. at 387, 788 P.2d 435], the court held that persons injured by products do constitute a class for purposes of Article I, section 20, because the class exists apart from the statute. Compare [Van Wormer, 309 Or. at 408 n 7, 788 P.2d 443] (persons who suffered governmentally inflicted wrongful death not true class); Eckles v. State of Oregon, 306 Or. 380, 387, 760 P.2d 846 (1988), cert. dismissed, 490 U.S. 1032, 109 S.Ct. 1928, 104 L.Ed.2d 400 (1989) (‘’classes’ of private insurers, insureds and [workers’ compensation] claimants on the one hand, and SAIF insureds and claimants on the other,’ are not true classes); Norwest v. Presbyterian Intercommunity Hosp., 293 Or. 543, 652 P.2d 318 (1982) (children of disabled parents not a class under Article I, section 20); with State ex rel Adult & Fam. Ser. v. Bradley, 295 Or. 216, 666 P.2d 249 (1983) (illegitimacy is a true class); Hewitt v. SAIF, 294 Or. 33, 653 P.2d 970 (1982) (gender is a true class). The court also has not been clear about what analysis applies if there is a ‘true class’ involved. Compare [Hewitt, 294 Or. at 33, 653 P.2d 970] (gender classification reviewed for whether it was based on intrinsic differences between the sexes) with Seto v. Tri County Metro. Transportation Dist., 311 Or. 456, 814 P.2d 1060 (1991) (geographic classification reviewed for rational basis) and [Hale, 308 Or. at 524, 783 P.2d 506] (rational basis test ‘has been superseded’). Further, it is unclear whether there is any judicial scrutiny at all if the class not a ‘true class,’ but is created by the statute itself. See, e.g., [Sealey, 309 Or. at 397, 788 P.2d 435] (classes ‘created by the challenged law itself’ are ‘entitled to no special protection and, in fact, are not even considered to be classes for the purposes of Article I, section 20’); [Hale, 308 Or. at 525, 783 P.2d 506] (victims of governmental torts are not an identifiable class based on personal or social characteristics and, therefore, there is no violation of Article I, section 20); [Eckles, 306 Or. at 387, 760 P.2d 846] (classes that exist only by virtue of the statute do not violate Article I, section 20); but see [Clark, 291 Or. at 240, 630 P.2d 810] (attacks on classes created by the legislative scheme itself have generally been rejected whenever the law leaves it open to anyone to join); Cole v. Dept. of Rev., 294 Or. 188, 655 P.2d 171 (1982) (law not directed at true class passed Article I, section 20, challenge because the privilege was available upon the same terms equally to all citizens); Hunter v. State of Oregon, 306 Or. 529, 761 P.2d 502 (1988) (grant of post-conviction relief to persons convicted of a state crime but not to persons convicted of a municipal crime is a classification created by the statute itself, but it does not violate Article I, Section 20, because the same standard applies to all).”Neher, 124 Or App at 225 n. 3, 862 P.2d 1307.

GARRETT, J.

James, J., concurred and filed an opinion, in which Bushong, J., joined.