Apportionment apoplexy: throwback, throwout, or just throw up your hands.

Apportioning corporate taxable income for state tax purposes may, at first blush, appear analogous to a will contest over a decedent's estate, with multiple parties each vying for a piece of the pie. Each party fighting over an estate wants as much as he or she can get, often caring little--if at all--for the needs or rights of the others involved. Each has his or her own idea of what the rules should be for determining how large a piece of the pie is deserved.

But there is an important difference between this situation and that involving state tax apportionment: Unlike the will contest scenario, in multistate corporate income taxation it is possible for each of the states to win. In a will contest, ultimately one arbiter will decide the law to be applied to the corpus of the estate, and all the slices of the "pie" will add up to the exact size of decedent's estate. In multistate corporate income taxation, that is not necessarily the case inasmuch as the Supreme Court has determined that, in the absence of a congressional requirement of uniformity, the states have a great deal of discretion in selecting and applying apportionment formulae. Each state may, by and large, use its own formula to determine its share. As a consequence, the states--in the aggregate--may tax more than 100 percent of the income of a corporation or corporate group. This state of affairs has not only been acknowledged by the Court, it has been--with certain limitations--condoned.

Fortunately, corporate taxpayers are not wholly without recourse. First, the Constitution does place real limitations on state apportionment of corporate income. In addition, the statutory language prescribing a state's formula may be open to interpretation, and may not support the state taxing authority's interpretation or application of the formula. Many states have specific statutory provisions that allow deviations from standard apportionment formulae (constitutional or statutory interpretation arguments can, if nothing else, be a powerful buttress when seeking such deviation). Finally, the states cannot simply make up the rules as they go. State law may prevent a state taxing authority from applying a particular interpretation or aspect of an apportionment law to the extent that the state taxing authority has not followed administrative law requirements for promulgating rules or regulations.

These issues converge in different ways in the context of the battery of apportionment formulae that corporate taxpayers face. Throwback rules, throwout rules, and even single-factor apportionment formulae can raise any or all of these bases for challenge, and the corporate taxpayer must be prepared to identify which base, or bases, can be helpful.

This article first explores the reasons formulary apportionment is used, and then discusses the constitutional limitations on the apportionment methods that states may use. Then, certain specific apportionment approaches are analyzed: single factor apportionment, throwback rules, and throwout rules. Finally, this article summarizes the weapons a taxpayer needs (and should be aware of) when attacking a taxing authority's apportionment position.

I. Why Apportionment? Slicing a Shadow.

A. What Is the Right Way to Carve Up the Income of a Multistate Business?

Before delving into the nitty gritty of apportionment law "quirks" such as throwout or throwback, it is worthwhile to explore why states use formulary apportionment in the first place. After all, apportionment is not found in nature--it is just an arbitrary but generally reasonable and efficient way to determine the share of a corporate tax base that a state is entitled to tax.

1. Separate Accounting

Historically, separate accounting was often employed in state tax systems to attribute to each relevant state the portion of the taxpayer's income that was generated there. "Early state income tax laws permitted corporations to treat separately the income earned in each state as long as they maintained separate geographic accounting records that enabled them to ascertain that income with reasonable accuracy." (1)

Separate accounting, however, has lost its luster over the years, at least with the governmental bodies that select apportionment formulae. Some have questioned whether attempting to carve up an enterprise into 50 different income generating components, for example, is ever feasible. Of course, beyond the issue whether separate accounting is an acceptable (or should be the default) means of carving up a corporate tax base, maintaining separate geographic records may be impractical from the taxpayer's perspective.

"[A]pportionability often has been challenged by the contention that ... the source of [particular] income may be ascertained by separate geographical accounting. The Court has rejected that contention so long as the intrastate and extrastate activities formed part of a single unitary business. Butler Bros. v. McColgan, 315 U.S. 501, 506-508 (1942); Ford Motor Co. v. Beauchamp, 308 U.S. 331, 336 (1939); cf. Moorman Mfg. Co. v. Bair, 437 U.S., at 272. In these circumstances, the Court has noted that separate accounting, while it purports to isolate portions of income received in various States, may fail to account for contributions to income resulting from functional integration, centralization of management, and economies of scale. Butler Bros. v. McColgan, 315 U.S., at 508-509. Because these factors of profitability arise from the operation of the business as a whole, it becomes misleading to characterize the income of the business as having a single identifiable 'source.'" (2)

In essence, the Supreme Court has concluded that even though formulary apportionment is necessarily artificial, separate accounting may be also to the extent it fails to take into account the synergies among business operations in different states. This reasoning led the Court to conclude in 1991 that "[a]lthough separate geographical accounting may be useful for internal auditing, for purposes of state taxation it is not constitutionally required." (3)

2. Formulary Apportionment

In contrast to separate accounting, formulary apportionment applies a percentage-based formula to the total tax base in order to determine the share attributable to the state in question. While certain items may be removed from the tax base as non-business or non-operational income, (4) in most states for most types of corporations the balance of taxable income is apportioned using a statutory formula. The product of formulary apportionment has become--generally--accepted as a reasonable proxy for the real amount of income (or other base) attributable to a particular state.

States use a wide variety of different formulae to apportion the income or franchise tax base that is computed under their tax laws. The baseline approach is the "three-factor" formula with fractions comparing a taxpayer's in-state property, payroll, and receipts to the respective total amounts of these items everywhere. Other variations include three-factor formulae with disproportionate weight given to certain factors, such as double-weighting the receipts fraction, and even certain single-factor formulae, such as a single-factor formula based only on the fraction of the taxpayer's in-state to its total receipts.

The Supreme Court has largely accepted these variations in methods of formulary apportionment because of "the difficulty of identifying the geographic source of the income earned by a multistate enterprise." (5) Because formulary apportionment is a synthetic means of determining the portion of corporate income, receipts, or other value that may properly be assigned to one state, there seemingly can be no single, perfect approach to formulary apportionment. Indeed, as the Court acknowledged, "[a]llocating income among various taxing jurisdictions bears some resemblance ... to slicing a shadow." (6)

B. State Variations In Apportionment Formulae Are Allowed

When the Supreme Court contemplated invalidating the Iowa single-factor receipts formula in 1978, the Court determined that the Commerce Clause did not flatly prohibit "any overlap in the computation of taxable income by the States." (7) The Court acknowledged that unless all of the states in which a corporation did business followed identical apportionment rules, there would always be some risk of duplicative taxation.

Nonetheless, the Court found that if the freedom of each state to choose its own, independent apportionment formula rules was ever to have to yield to "an overriding national interest in uniformity," the determination of which approach to apportionment should govern all states "should be determined only after due consideration is given to the interests of all affected States." (8) The Court refused to make a "policy" decision of which of the many different approaches to formulary apportionment should be the single, uniform rule across the country. Instead, it concluded such a determination was the sole province of Congress as part of its power under the Commerce Clause. To date, Congress has declined to exercise that power to address this issue.

II. Constitutional Requirements

Just what are the constitutional limitations on the apportionment methods that states employ? They include the Commerce Clause fair apportionment requirements, the Due Process Clause preclusion of disproportionate taxation, and finally the Commerce Clause and Equal Protection Clause discrimination prohibitions.

A. Fair Apportionment

The Supreme Court has identified four requirements for any state tax to pass muster under dormant Commerce Clause jurisprudence, and one of those four requirements is that the tax scheme fairly apportion the base that it taxes. (9) ("Dormant Commerce Clause jurisprudence" refers to the body of common law developed by the Supreme Court in the absence of specific congressional action.) That fair apportionment requirement has been further developed by the Court in the form of consistency testing. The Court's tests for "internal consistency" and "external consistency" of apportionment formulae were first identified as such in Container Corp. of America v. Franchise Tax Board, 463 U.S. 159 (1983).

1. Internal Consistency Test

When it first articulated the internal consistency test, the Supreme Court pronounced that the "first, and again obvious, component of fairness in an apportionment formula is what might be called internal consistency--that is, the formula must be such that, if applied by every jurisdiction, it would result in no more than all of the unitary business' income being taxed." (10)

To test internal consistency, one must evaluate the tax burden that would result if a multistate corporation were subjected in every state to the same formula employed by the one state in question. This is a hypothetical exercise by its very nature; the actual tax burden borne by the corporation in other states is not relevant. For example, when evaluating New Jersey's apportionment rules, a corporation doing business in both New Jersey and Pennsylvania would hypothesize that both states employed New Jersey's apportionment rules. If there is a risk that more than 100 percent of the business' income will end up in the two states' tax bases in the aggregate, then New Jersey's rules are internally inconsistent and, hence, unconstitutional.

The Supreme Court has on three separate occasions found state tax provisions to violate the internal consistency requirement--two involving business and occupation tax provisions, and another involving an unapportioned axle tax on trucking companies.

In 1984, the Court found that a multiple activities exemption from a West Virginia business and occupation tax violated internal consistency. (11) The exemption applied to wholesale businesses conducted by in-state manufacturers in West Virginia, purportedly justified on the ground those entities were already subject to the state's tax on manufacturers. The same type of wholesale business conducted by out-of-state manufacturers, however, was taxed. The Court found the scheme to cause inconsistency, even though in-state manufacturers were subject to the manufacturing component of the tax. If every state in the nation applied that scheme, the wholly in-state businesses would only bear one level of tax because they would be subject to tax as a manufacturer but not as a wholesaler. A business that crossed state lines, however, would be subject to tax as a manufacturer in its home state and then again as a wholesaler in the other state being examined. The multiple activities exemption thus caused the tax to be internally inconsistent.

The second case also involved a multiple activities exemption--this time the manufacturer's exemption from the Washington State business and occupation tax. (12) The exemption was available to manufacturers that also conducted and paid the Washington business and occupation tax on a wholesaling business. This multiple activities exemption was found to be internally inconsistent because an interstate manufacturer/wholesaler could, theoretically, have to pay both the manufacturing and the wholesaling tax, while an intrastate manufacturer/wholesaler would pay only the wholesaling tax.

Finally, in the third case, the Court struck an unapportioned axle tax imposed on trucking companies as internally inconsistent. (13) Trucking companies doing business across the country could be subjected to the same "flat" amount of tax by every state, while an intrastate trucking company would bear the burden of only one tax.

Several state courts have similarly struck down internally inconsistent tax provisions. (14) For example, the New Jersey Supreme Court recently invalidated an unapportioned hazardous waste fee as internally inconsistency, relying heavily on the Scheiner decision. (15) The court found that the State had, and failed to meet, the burden to prove the flat fee was not unconstitutional. In addition, the court found the fee violated two of the other prongs of the Complete Auto Transit Commerce Clause test because it "discriminates against interstate commerce by charging a flat fee unrelated to a transporter's level of activity in the State; and it places an undue burden on interstate commerce."

In Sprint Communications Co., (16) the District of Columbia Court of Appeals found the District's gross receipts tax violated internal consistency by providing certain credits for and exemptions from the District's sales and use and personal property taxes. The court noted that the long distance telecommunications carriers located outside the District that were challenging the tax would not be able to obtain the gross receipts tax offset that domestic carriers (that were subject to those other personal property and sales taxes) received.

The Alabama Supreme Court similarly struck a gross receipts-based tax, there a municipal business license tax imposed on corporations engaged in the business of soliciting or selling in the jurisdiction. (17) The tax was computed on the basis of 100 percent of a licensee's unapportioned gross receipts from sales regardless of where the sale was solicited, the contract was consummated or the delivery was made. The court concluded the tax was internally inconsistent "because, if local governments in other states in which [the taxpayer] does business ... were to impose license taxes based on gross receipts from sales made within their respective jurisdictions, then multiple state taxation of interstate commerce would result."

2. External Consistency Test

The Supreme Court recognized at its first articulation of the internal and external consistency concepts that external consistency is more convoluted: "The second and more difficult requirement is what might be called external consistency--the factor or factors used in the apportionment formula must actually reflect a reasonable sense of how income is generated." (18)

Beyond the fact that reasonable minds can often differ about what constitutes a reasonable sense of how income is generated, proving an inaccurate reflection of that "reasonable sense" quantitatively is indeed difficult. There is no conceptual formula that one can use to test external consistency--a taxpayer must be wary of (and, ultimately, able to prove) how much of its tax base is being apportioned to the state, how much of its business is really conducted there, and whether there is a mismatch. Like the Supreme Court's test for pornography, to some extent one just has to know it when he or she sees it. For practical purposes, this requirement is similar to the Due Process Clause prohibition against disproportionate taxation.

In 2003, the Pennsylvania Supreme Court found that the application of the Philadelphia business privilege tax to media receipts received by the Philadelphia Eagles Football Club violated the external consistency requirement. (19) The city taxing authority imposed the tax on 100 percent of the club's media receipts for the television broadcast of its football games regardless of whether those games were played in Philadelphia or in another location. The court found the club demonstrated by clear and cogent evidence that the Commerce Clause was violated because a tax on receipts derived from all of the team's games even though half of those games were played outside of Philadelphia was plainly out of all proportion to the Club's business activities in Philadelphia that generated the payment of the receipts. That the club "was commercially domiciled in Philadelphia and played some of its games there only meant that the City was entitled to tax its fair share of the receipts, not all of the receipts...."

B. Due Process--"Out of All Appropriate Proportion"

Long before it identified the Commerce Clause external consistency test for fair apportionment, the Court required--for Due Process Clause purposes--that the income attributed to a state by that state's apportionment formula be rationally related to the values connected with the taxing state. (20)

The only case in which the Supreme Court found in favor of a taxpayer making this argument was Hans Rees' Sons, Inc. v. North Carolina. (21) The Court found that the employment of a single-factor property formula to tax 83 percent of the income of a taxpayer that only had 17 percent of its income sourced to the state (based on its manufacturing activities there) was out of all appropriate proportion to the business transacted in the state. This disproportionate result violated the Due Process Clause. The Court explained that, in general, unless an apportionment formula is "intrinsically arbitrary" the method will be sustained until proof is offered of an unreasonable and arbitrary application in particular cases (as had been established there). This is one of the situations where separate accounting is useful to taxpayers--namely, as a means to prove the amount of income generated by activities in a particular state in order to compare it with the results yielded by that state's apportionment formula.

In Trinova Corp. v. Michigan Department of Treasury, a decision rendered 60 years after Hans Rees' Sons, the Supreme Court reiterated that a Due Process Clause challenge requires the taxpayer to establish that there is no rational relationship between the tax base measure attributed to the state and the contribution of the taxpayer's business activity in that state to the tax base. (22) Indeed, in order to prevail, the taxpayer would be required to prove by "'clear and cogent evidence' that the income attributed to the State is in fact 'out of all appropriate proportions to the business transacted ... in that State', or has 'led to a grossly distorted result.'" (23)

Trinova Corporation challenged the apportionment method employed by the Michigan Single Business Tax (SBT)--a value-added tax--under both the Due Process Clause and the Commerce Clause. As explained by the Court, the SBT base consisted of a taxpayer's cost of labor, depreciation, interest, and profit. To the extent that a taxpayer did business both within and outside of Michigan, the base must be adjusted to determine that portion of each component that is attributable to Michigan. Apportionment was accomplished under the SBT by multiplying the total tax base by a three-factor apportionment formula that compared the Michigan amounts to the total amounts of payroll, property, and sales. The SBT also provided for two deductions relevant to Trinova--one for capital acquisitions and one for labor costs that exceeded 63 percent of the total tax base.

During the year at issue, Trinova, an auto parts manufacturer, was based in Ohio but also had a sales office in Michigan. More than 26 percent of its sales were made to Michigan customers. Trinova argued that it was able to and should have been allowed to specifically allocate its labor and depreciation costs to locations within and without Michigan--instead of apportioning those two elements of the tax base using three-factor apportionment. Although Trinova agreed to three-factor apportionment of its income (as one component of the tax base), Trinova had no income in that year.

The Court acknowledged that Trinova could identify the exact location of its plant, equipment, and most of its payroll, but concluded that Trinova's proposed approach was "incompatible with the rationale of a [value-added tax], and is unsupported in the record." Noting that the components of the SBT base are not each "separate and independent taxes," the Court found that Trinova's approach failed to take into account the integration of those various components as adding value "[i]n a unitary enterprise." Picking apart certain components of the tax for specific geographic localization failed to account for the "remainder or residual [of value added] that cannot be located with economic precision."

In specifically addressing Trinova's Due Process argument that the Michigan formula resulted in taxation "out of all appropriate proportions to the business transacted" in Michigan, the Court did note that Trinova's position "finds some support among economists." The Court found, however, that it was unable to conclude which of three alternative approaches to apportionment gave the most accurate calculation of Trinova's value added in Michigan because--

   Trinova has not convincingly demonstrated which
   figure is most accurate. Trinova gives no estimate
   of the value added that would take account of both
   its Michigan sales activity and Michigan market
   demand for its products.... Trinova has failed to
   meet its burden of proving "by 'clear and cogent evidence,'"
   Moorman Mfg. Co., 437 U.S., at 274, that
   the State of Michigan's apportionment provides a
   distorted result. (24)

C. Discrimination

In addition to its fair apportionment requirement, the Commerce Clause requires that a tax "not discriminate against interstate commerce." (25) Indeed, in addition to its finding of internal inconsistency, the Supreme Court in Armco, Inc. v. Hardesty found the imposition of the West Virginia wholesaler gross receipts tax on a non-domiciliary corporation to be an impermissible interference with free trade. While there are many ways in which a tax might discriminate, a "tax that unfairly apportions income from other States is a form of discrimination against interstate commerce." (26)

Similarly, to the extent an apportionment formula effectively caused a greater tax burden on out-of-state corporations, that might be "the very sort of parochial discrimination that the Equal Protection Clause was intended to prevent." (27)

III. Three-Factor Apportionment--The "Benchmark" A. The Three-Factor Formula

The "three-factor" apportionment formula--averaging the fractions of: (1) in-state receipts to total receipts, (2) instate property to total property, and (3) in-state payroll to total payroll--is the standard formula upon which most state apportionment approaches are based. There are, not surprisingly, many variations on this general theme. For example, many states double-weight the receipts fraction so that receipts account for 50 percent of the overall formula. Other variations may be found in various states' rules for determining which of a taxpayer's receipts, property or employees are considered to be "in-state."

B. Three-Factor Formula is the Benchmark

Although it has approved other approaches to formulary apportionment, the Supreme Court has declared the standard three-factor formula to be "something of a benchmark against which other apportionment formulas are judged." (28) In blessing the three-factor apportionment formula of the Michigan single business tax, the Court noted that this method was "first approved for apportionment of income in Butler Bros. v. McColgan, 315 U.S. 501 (1942)." (29) As the Court stated in Trinova:

   The three-factor formula is widely used, and is included
   in the Uniform Division of Income for Tax
   Purposes Act, 7A U.L.A. 331 (1990 Cum. Supp.)
   (approved in 1957 by the National Conference of
   Commissioners on Uniform State Laws and the
   American Bar Association). (30)

Why has this method has gained such broad acceptance? Even though the Supreme Court has acknowledged that it is "not a precise apportionment for every case," nonetheless the three factors--payroll, property, and sales--"appear in combination to reflect a very large share of the activities by which value is generated." (31) This makes sense. The Court has told us that an apportionment formula must yield an amount that is rationally related to the business actually conducted in that state. What are the factors that generate income--the labor that makes a company run (reflected by payroll) and the capital that those people employ to do so (reflected by property). The sales factor reflects the marketplace for what the company sells. The Court has stated that "in combination" these three factors generally provide a good representation of what creates any company's income.

Most states, however, have either modified the standard three-factor approach or have opted not to use it at all. One approach apportions income based on a receipts factor (receipts sourced to that state divided by all receipts) standing alone. While one Supreme Court case did address this type of "single-factor" formula, it is not at all clear that this approach is free from further constitutional scrutiny.

IV. Single-Factor Apportionment

A. Does Moorman Really Bless Single-Factor Receipts-Based Apportionment?

The Supreme Court has acknowledged that a variety of different apportionment formulae may pass constitutional muster. Indeed, the Court in 1920 approved a formula that apportioned based on a property factor alone. (32) Many would point to the Court's 1978 decision in Moorman as authority for using a single-factor receipts-based apportionment formula. (33) The Court's decision to uphold the formula in that case, however, largely turned on the taxpayer's failure to develop an adequate record.

Moorman was a manufacturing corporation that had more than 500 salesmen and six warehouses in Iowa; it could not deny that it had a significant level of business activity within Iowa's borders. Moorman attacked Iowa's single-factor receipts-based formula on the grounds that it caused duplicative taxation when viewed together with Illinois taxes, yet failed to provide the factual evidence necessary to reach that conclusion.

The Supreme Court expressly attributed its rejection of this argument to Moorman's failure to make its record: "[s]ince the record does not reveal the sources of appellant's profits, its Commerce Clause claim cannot rest on the premise that profits earned in Illinois were included in its Iowa taxable income and therefore the Iowa formula was at fault for whatever overlap may have existed." (34) Thus, while many point to the decision in Moorman as blessing single-factor apportionment, the decision perhaps is more fairly seen as a challenge to that method, as applied, that failed to establish the facts necessary to show excessive taxation. One can imagine that Moorman would have prevailed had it shown where and how its profits were generated.

Moreover, in the time since the Moorman decision was rendered in 1978, the standards for fair apportionment and discrimination have been refined by subsequent decisions, such as the external consistency test articulated in Container Corp. The Supreme Court's Moorman decision did not, because it could not, specifically address the external consistency of a single-factor receipts-based formula. At the very least, there remains fertile ground for a fresh challenge to one of the single-factor formulae currently employed in various states across the country.

V. Throwback/Throwout

A. Throwback Rules

1. What Do They Do, and What Does It Mean to Be "Subject to Tax"?

As a general matter, a throwback rule allows a state to include a receipt in the numerator of its receipts fraction when that receipt would otherwise not be so included. Certain state statutes require this when another state that would "normally" include the receipt in its own numerator either cannot or has chosen not to impose tax on the corporation. Typically, throwback rules are employed by a state from which a shipment of tangible personal property originates as a means to grab the receipt back from the destination state--even though the origination state normally would assign the receipt for apportionment purposes to the state of shipment destination. The origination state's justification is that if the destination state does not tax the corporation, the tax attributable to the receipt will not be collected by any state.

The UDITPA model rule requires throwback when the taxpayer is not taxable in the purchaser's state. A taxpayer is considered taxable in another state if it meets either of two tests: "(1) By reason of business activity in another state, the taxpayer is subject to one of the types of taxes specified in Article IV.3.(1), namely: A net income tax, a franchise tax measured by net income, a franchise tax for the privilege of doing business, or a corporate stock tax; or (2) By reason of such business activity, another state has jurisdiction to subject the taxpayer to a net income tax, regardless of whether or not the state imposes such a tax on the taxpayer." This is, of course, just a model provision. The actual statutory language used by the 24 states that have enacted throwback provisions varies in many different ways from this model.

Common causes of throwback among different states are the taxpayer not being subject to tax in the destination state (because it is protected by constitutional nexus requirements or by federal Public Law No. 86-272), (35) or the taxpayer being subject to tax, but the destination state choosing not to impose tax on it. (36) For example, the Indiana Department of Revenue contends that "[i]n every sales transaction, at least one state has the power to tax income derived from the sale of tangible personal property; if the state wherein the sale occurred is forbidden to do so by 15 U.S.C. [section] 381, then that income is 'thrown back' to the originating state." (37) Other triggering events for some throwback statutes include situations where the taxpayer's customer is the federal government (38) or the taxpayer's sales were made into a foreign country. (39)

Because there are many variations in throwback rules, one must carefully review the precise language of any individual state's statute (and the case law interpreting it). Is the taxpayer protected against throwback whenever the other state has jurisdiction to tax it, or must the taxpayer have actually filed a return and paid a tax in the other state? Which state's law is used to determine whether Public Law No. 86-272 shields the taxpayer from tax in the destination state? What if, in a combined reporting state, the taxpayer is not taxable in the shipment destination state but another corporation in its combined group is? These scenarios have created numerous thorny issues.

Perhaps most important, should the origination state--from a tax policy perspective--be allowed to employ an apportionment scheme that increases its share of the tax base merely because of the laws of and activities in another state and not because of any increase in the taxpayer's amount of business activity in the origination state? Is that increased share of the tax base rationally related to the taxpayer's values connected with the origination state, or proportionate to the business transacted in the origination state? (40) Arguably, no.

2. What Must a Corporation Do to Avoid Throwback ?

In situations where a throwback provision is triggered when the destination state is unable (constitutionally or under Public Law No. 86-272) to tax that corporation or sale, the taxpayer may find itself in the odd position of having to prove that it was subject to tax or had nexus in the shipment destination state. Not surprisingly, states' views of nexus tend to be much more restrictive when they are seeking to employ a throwback rule against one of their own as compared to when they are seeking to assert nexus over an out-of-state corporation. Is that fair?

In one recent decision by the Massachusetts Appellate Tax Board, a taxpayer was required to make exactly that type of showing in arguing that the Commissioner incorrectly applied throwback to a subsidiary included in its combined return. (41) The Commissioner contended that the Commonwealth's throwback rule required the subsidiary's sales delivered to 33 other states to be thrown back to the numerator of the taxpayer's Massachusetts receipts fraction. That throwback rule applied when "the corporation is not taxable in the state of the purchaser and the property was not sold by an agent or agencies chiefly situated at, connected with or sent out from the premises for the transaction of business owned or rented by the corporation outside this commonwealth." (42)

The taxpayer defeated the Commissioner's throwback position on two different grounds. First, the taxpayer argued, and the Appellate Tax Board found, that because the subsidiary's activities in the shipment destination states went beyond Public Law No. 86-272 the subsidiary was "taxable" in those states. The taxpayer offered evidence of the general practices of the subsidiary across the country. The Board concluded that the subsidiary's training on product usage, product demonstrations, and trouble-shooting went beyond activities ancillary to solicitation--in other words, they exceeded the types of activities protected by Public Law No. 86-272. In reaching this conclusion, the Board applied Massachusetts law interpreting Public Law No. 86-272 to determine that the subsidiary was "taxable" on its sales into the 33 other states. The Board also rejected the Commissioner's argument that the taxpayer should be required to "'pinpoint' exactly where and when specific activities occurred throughout every corner of the thirty-three disputed jurisdictions." (43) Given that the Commissioner had produced no contrary evidence, the Board concluded that "appellant met its burden of proving that [the subsidiary] engaged in 'in-service' advice and troubleshooting activities 'as a matter of regular company policy, on a continuing basis' throughout the disputed jurisdictions where it registered sales." (44)

For 23 of the states from which the Commissioner contended throwback was required, the taxpayer also successfully argued that the subsidiary's sales to those states were not initiated from Massachusetts but rather were connected with the subsidiary's offices outside the Commonwealth. The language of the Massachusetts throwback rule is unique; most throwback rules apply only if the state seeking to enforce it is the state from which actual ship-merit originated.

Similarly, several decisions have rejected application of the Michigan single business tax throwback rule where the taxpayer could establish substantial nexus for Commerce Clause purposes in the other states into which its sales were destined. (45)

A recent decision from the Illinois Circuit Court of Cook County rejected the application of throwback to Illinois where the taxpayer paid tax to the jurisdiction to which its sales were destined, albeit not on the sales in question. (46) The Illinois Department of Revenue sought to throwback the taxpayer's $13 million in taxable sales made to foreign countries. The company did pay a net income tax in those countries, but paid that tax on income unrelated to the sales at issue. The court noted that the Illinois statute required throwback only when "the person is not taxable in the state of the purchaser." (47) The court found it could not require throwback because the taxpayer was in fact taxable in the foreign countries in question--even though the receipts at issue were not and its decision resulted in "nowhere sales."

3. Issues Unique to Combined Groups--California and New York

a. California--Joyce and Finnigan (Huffy and Citicorp, too)

Two significant decisions from California highlight the complex issue of apportionment and throwback in the context of a combined group. While only Appeal of Finnigan Corp. (48) specifically dealt with a throwback rule, the decision in Appeal of Joyce, Inc. (49)--and the contradictory outcomes in the two cases--reveal this important apportionment issue: Should "taxability" of a corporate member of a combined group be governed by looking in isolation at the single corporation that conducted the transaction(s) in question, or instead by looking at the activities of both that corporation and the activities of all others (or any other) in the taxing state's combined group?

In Finnigan, the California State Board of Equalization (SBE) determined that California had no right to claim the throwback of receipts from sales made by a member of the combined group into states where that particular member was not itself taxable, even though a combined affiliate was so taxable. Noting that the California statute allowed throwback only if the "taxpayer is not taxable in the state of the purchaser," the SBE determined that the "taxpayer" to be analyzed included all of the corporations in the combined unitary group. Because other corporations in the group were taxable in those states, the SBE concluded that the receipts could not be thrown back into the group's receipts fraction numerator. The Finnigan case actually involved two separate decisions. In the first ("Finnigan I"), (50) the SBE determined that throwback was not required so long as one corporation in the unitary group was taxable in the state in question. In the second ("Finnigan II"), (51) on rehearing, the SBE stuck by its initial decision but also acknowledged that its interpretation of the throwback rule required it to specifically overrule its prior decision in Appeal of Joyce--a case that did not involve throwback.

Many years earlier the SBE had decided Joyce, which addressed in-bound California sales that originated from out-of-state (and, thus, necessarily could not have involved California throwback). In Joyce, the SBE had concluded that taxability should be determined on an individual corporation basis for each of the members of the combined group. Under Joyce, the in-bound sales of a combined group member that was not subject to tax in California could not be used to apportion additional taxable income to the taxpayer-member's California tax base. Although the story does not end here, the Finnigan II decision in 1990 concluded that Joyce was "analytically and philosophically incompatible" with its reasoning, and Joyce was therefore overruled.

In a nutshell, the distinction between these cases is in whether one tests for subjectivity to tax on a separate corporation basis for a combined group (Joyce), or whether one instead tests to see if any corporation within the combined group is subject to tax in that state, and if so, then considers every corporation in the group to be subject to tax in that jurisdiction--at least for apportionment purposes (Finnigan).

In Finnigan II, the SBE rectified the inconsistency between Finnigan and Joyce, and Finnigan/Finnigan II remained the law in California for several years--both for in-bound and out-bound transactions. Not even 10 years later, however, the SBE rejected its earlier Finnigan and Finnigan II positions and returned to the rule employed in Joyce. (52)

b. New York--Alpharma and Disney--and Other States

The New York Division of Tax Appeals has recently been presented with two cases that raise the Joyce/Finnigan II issue. The Administrative Law Judges (ALJs) in both cases, and the Tax Appeals Tribunal in Alpharma, followed the Finnigan II approach of including in the New York receipts fraction numerator of a combined group the New York receipts of group members that are not subject to tax in New York. (53) The flaw in this approach, according to the taxpayers, is that including these receipts (the receipts of a corporation that is not subject to New York taxation) in the New York numerator for the combined group has the same effect as taxing the corporation directly.

In Matter of Alpharma, (54) the New York State Tax Appeals Tribunal affirmed the decision of an ALJ (albeit on other grounds) to include the New York receipts of a nexus-protected member of the unitary group in the numerator of the sales factor when computing the combined apportionment formula. The Tax Appeals Tribunal concluded that an apportionment formula is only a formula for dividing income and it does not determine (and is not impacted by) nexus or subjectivity to tax. Thus, according to the Tribunal, the nexus-protected status of any particular member of the unitary group is not relevant to the computation of the "New York State receipts" numerator of the receipts fraction for apportionment purposes.

In a similar vein, a New York State Administrative Law Judge (ALJ) in Matter of Disney Enterprises, (55) determined that receipts of a nexus-protected member of a unitary group could be included in the "New York State receipts" numerator of the receipts fraction in calculating the combined apportionment formula. The ALJ determined that Public Law No. 86-272 did not prohibit this treatment because in this instance it could not be concluded that the only business activities conducted within New York by the subsidiaries at issue or on behalf of those entities was the solicitation of orders for sales of tangible personal property.

Furthermore, the ALJ determined that protections afforded to nontaxpayer corporations by the Commerce Clause did not prohibit inclusion of nexus-protected subsidiaries' receipts in the New York numerator of the receipts fraction because the company's activities were part of a unitary business that had sufficient nexus. That "sufficient nexus" of course came through the activities of other members of the unitary group who were separate corporate entities. An appeal is pending.

Several other states have weighed in on the Joyce/ Finnigan II issue. The Texas Comptroller, in the context of a Texas taxpayer trying to avoid throwback to Texas on its outbound sales, noted that "Texas looks only at the single corporate entity as the 'taxpayer.'" (56) The Comptroller concluded that the taxpayer could not establish it was subject to tax in (and thus could avoid throwback from) other states in which the taxpayer did not individually have nexus but was included in a combined return based on the nexus of other members of its combined group.

In Illinois, the state Appellate Court has determined that the Joyce rule will be applied when evaluating whether throwback is required. (57) The taxpayer had made sales into jurisdictions in which it was not subject to income tax, but in which other corporations included in its unitary group were taxable. The court noted that throwback is triggered whenever the "person" is not taxable in the state of the purchaser, (58) and found that the "taxpayer" referred to the particular corporation in question--not to its unitary group as a whole. Thus, the taxpayer's sales in question could be thrown back to Illinois.

4. Policy Consideration: Is "Nowhere Income" Really a Problem?

One must ask whether it is really appropriate, as a policy matter, to throwback any receipts? After all, the Supreme Court has expressly accepted the fact that the price of allowing each state to adopt its own apportionment formula is that some taxpayers inevitably will be taxed on more than 100 percent of their income. Why should it not be equally acceptable that some taxpayers may be taxed on less than 100 percent of their income, especially when that result is dictated by another state's policy decisions? Why should "nowhere income" be any more unpalatable than the over-taxation or "imaginary income" that results for some taxpayers?

B. Throwout Rules

1. What Are They?

Instead of a throwback rule, two states have adopted "throwout" rules that similarly affect the state apportionment formula in the event that there are "nowhere" receipts. Unlike a throwback rule, throwout functions to throw items out of the denominator of the apportionment formula--generally based on those items' not being picked up in any other state's numerator. The result is the same as with a throwback rule: a higher apportionment percentage for the state employing the rule, even though there has been no increase in economic activity in that state.

2. The West Virginia and New Jersey Throwout Rules

Long the only state statutory throwout rule on the books (until New Jersey's 2002 enactment of a throwout rule, discussed below), West Virginia employs a throwout rule that eliminates from the denominator of the West Virginia receipts fraction any receipts from sales of tangible goods destined to states in which the taxpayer is not taxed. The statute also throws out sales to the federal government. Like a traditional throwback rule, the pro vision only applies to sales of tangible personal property:

Sales of tangible personal property are in this state if:

***

(ii) The property is shipped from an office, store, warehouse, factory or other place of storage in this state and the purchaser is the United States government.

(B) All other sales of tangible personal property delivered or shipped to a purchaser within a state in which the taxpayer is not taxed, as defined in subsection (b) of this section, shall be excluded from the denominator of the sales factor. (59)

To be deemed taxable in another state, the statute requires that "the taxpayer [be] subject to a net income tax, a franchise tax measured by net income, a franchise tax for the privilege of doing business, or a corporation stock tax," or "that the destination state [have] jurisdiction to subject the taxpayer to a net income tax, regardless of whether, in fact, that state does or does not subject the taxpayer to the tax."

Unlike the West Virginia provision, the throwout rule enacted by New Jersey in 2002 is not limited to sales of tangible personal property, and is not even limited to sales that originate in New Jersey. In fact, New Jersey has taken it upon itself to effectively seek custody of all "wayward receipts," even if New Jersey is neither the state of origination nor the state of destination for that transaction:

   if receipts would be assigned to a state, a possession
   or territory of the United States or the District of
   Columbia or to any foreign country in which the
   taxpayer is not subject to a tax on or measured by
   profits or income, or business presence or business
   activity, then the receipts shall be excluded from
   the denominator of the sales fraction. (60)

Qualifying taxes of other jurisdictions that will prevent a receipt from being thrown out include net worth taxes, gross receipts taxes, and the Michigan single business tax, but not property taxes, excise taxes, payroll tax, or sales tax. (61) There is a statutory limitation on the dollar impact of New Jersey's throwback rule. The increased liability for all members of an affiliated or controlled group (compared to what would have been calculated without the throwback rule) cannot exceed $5 million. (62)

One must keep in mind that New Jersey's throwout rule does not apply only to receipts from sales of tangible personal property, but does extend to receipts generated by any other means (financial services, etc.). Furthermore, unlike West Virginia, New Jersey will throwout receipts even if New Jersey was neither the state of origination nor the state of destination. Indeed, New Jersey contends that "[r]eceipts which are included in the numerator of a jurisdiction's receipts fraction by reason of the operation of a throwback provision are deemed not to be receipts assigned to that jurisdiction and are, therefore, excludible from this State's receipts fraction denominator." (63)

What is New Jersey's right to claim a greater share of income based on transactions that never had any New Jersey connection? While the Division of Taxation has "recognize[d] there could be cases where distortion occurs" and offered to "entertain requests for relief on a case by cases basis," (64) these systematic flaws cry out for a systematic cure--namely, through judicial challenge or legislative amendments.

3. Administratively Derived Throwout Rules

Beyond the statutory throwout rules just discussed, some states have attempted to impose throwout on an administrative basis.

Although Pennsylvania did not have a statutory throwout rule, until 1984 the Commissioner applied a throwout rule to certain taxpayers (on a case-by-case basis) pursuant to his administrative powers of adjustment. Throwout was applied where the standard apportionment formula did not "fairly represent" the taxpayer's business activity in the state. As with statutory throwout rules in place in other states, receipts were eliminated from the receipts fraction denominator where the taxpayer was neither taxed nor subject to tax in the destination state.

The state's highest court initially upheld this use of the Commissioner's statutory power of adjustment. (65) Six years later, in a different case challenging the Commissioner's authority to assert throwout, the Pennsylvania Supreme Court reversed its position and overruled its earlier approval, finding the throwout technique to be beyond the Commissioner's statutory power. (66) Specifically, the court noted that "the express legislative intent that the apportionment provisions be utilized to tax 'business activity in this State' ... is unequivocal, and application of the 'throw out' rule so as to tax business activities in other states is plainly in derogation of that intent."

   We perceive no defect in the logic, or fairness, of a
   statutory scheme that results in income being apportioned
   to states not imposing a tax, provided the
   scheme assures, as does the present one, that tax
   burdens are borne so as to correlate with benefits
   derived from business activities in fact conducted
   in this Commonwealth. To deem as business activities
   in this Commonwealth such activities as occur
   elsewhere, simply because those activities are not
   elsewhere taxed, is to vitiate the apportionment
   provision designed to tax "business activity in this
   State." (67)

As the high court in Pennsylvania noted, the statutory scheme in Pennsylvania was designed to tax business activities conducted in Pennsylvania. "[W]hether other states in which a corporation does business choose or do not choose to levy an income tax on that corporation is irrelevant to a determination of the corporation's taxable income in Pennsylvania."

Indeed, as a general matter (beyond the limitations of Pennsylvania's statutory scheme), the Constitution requires that an apportionment scheme reach only the portion of the base that reflects a taxpayer's business activity in that state. Throwout and throwback rules are not designed to reflect an increased level of activity in the state that uses the rule to increase its share of the tax base. Instead, throwout and throwback increase that state's share of apportioned income (or other base) due to the wholly arbitrary fact of whether another state can or does tax that piece of the tax base pie.

VI. Non-Constitutional Arguments Against Apportionment Positions

A. Proper Promulgation--Administrative Procedure Act Requirement

Many states have an Administrative Procedure Act that requires a government agency (including a taxing authority) to promulgate formal regulations to articulate a rule of general applicability. If those specific procedural "rule-making" requirements--notice to the public perhaps chief among them--have not been followed, then the tax agency may not enforce the "unwritten rule" against a taxpayer. (68)

Some courts have in fact rejected tax agency interpretations of an apportionment rule where the agency failed to follow the state's Administrative Procedure Act. In 1984, the New Jersey Supreme Court rejected the Director's use of audience share as a means to apportion the receipts of a multistate television or radio station because that method constituted a rule of general applicability. (69) The Director contended he was authorized to employ such a method pursuant to his discretionary authority to modify the formula to achieve "a fair and proper allocation." The court found, however, that because the method would apply to other similarly situated broadcasters, it constituted de facto rule-making that had to comply with the Administrative Procedure Act requirements. Those requirements had not been met, and the method could not be applied against the taxpayer.

When faced with a taxing authority's apportionment position that is not embodied in a formally promulgated rule or regulation, a taxpayer should consider whether the position is one that would apply to other taxpayers as well. If so, then the taxpayer should explore whether a defense can be made on the grounds of failure to comply with that state's Administrative Procedure Act.

B. Outside the Scope of the Statute

Often, the statutory language employed by a state does not unambiguously cover the full breadth of issues that arise in apportionment. Thorny issues may be presented on questions of sourcing, determining what it means to be subject to tax in another jurisdiction, and so on. The wary taxpayer must be attuned to the possibility that a taxing authority's position simply lies outside the scope of the statute.

Furthermore, a taxpayer should never simply assume that the taxing authority's discretion is unbridled. Careful review of the precise statutory language creating discretionary power that impacts apportionment may reveal that the taxing authority's position exceeds that discretionary power.

VII. Fighting Back: What It Takes; Relief Provisions

A. Proving Distortion--The Taxpayer's Burden

The common and visible defect in most taxpayer challenges to apportionment rules is the failure to substantiate the existence of the flaw in the apportionment rules as they apply to the taxpayer. Two factors likely contribute to this--the complexity of quantifying the distortion or excessive taxation that results from the allegedly improper rules and the strong presumption in favor of the government. Indeed, when faced with a difficult constitutional question, it is all too easy for a court to avoid the issue by determining that the taxpayer simply failed to establish the facts necessary to support a constitutional challenge to apportionment rules.

For example, in its 1991 decision against Trinova Corp., the Supreme Court swiftly rejected any obligation to determine which of the three alternative methods of apportionment urged by the parties "gives the most accurate calculation" simply because "Trinova has not convincingly demonstrated which figure is most accurate." (70)

   Trinova gives no estimate of the value added that
   would take account of both its Michigan sales activity
   and Michigan market demand for its products.
   Michigan, on the other hand, has consistently applied
   a formula, the elements of which appear to
   reflect a very large share of the activities by which
   value is generated, with further relief for labor intensive
   taxpayers such as Trinova. (71)

The Court found Trinova failed to meet its burden of proving by clear and cogent evidence that Michigan's apportionment formula resulted in distortion; no further consideration of the issues was required.

Similarly, in Moorman, the Supreme Court noted the taxpayer's failure to even "suggest that it has shown that a significant portion of the income attributed to Iowa in fact was generated by its Illinois operations; the record does not contain any separate accounting analysis showing what portion of appellant's profits was attributable to sales, to manufacturing, or to any other phase of the company's operations." (72) Here, too, Moorman's challenge to Iowa's single-factor apportionment formula was flatly rejected.

Taxpayers mounting future challenges to apportionment must learn from the Court's ready rejection of these taxpayers' arguments. Effective and reliable factual evidence of distortion must be obtained and properly introduced at the trial level in order to build a potentially successful challenge to an apportionment scheme.

B. Statutory Relief as a Constitutional Circuit Breaker

State taxing authorities and taxpayers alike have each, at times, attempted to employ statutory apportionment relief provisions to avoid what they acknowledge or contend would otherwise be an unconstitutional result. These provisions are sometimes described as "constitutional circuit breakers" that can function to save an apportionment formula against an unconstitutional application. (73)

Most statutory schemes provide the Commissioner or Director with discretionary authority to adjust an apportionment formula under certain circumstances for a particular taxpayer. For example, many statutes provide authority to exclude one or more of the factors, or to include additional factors, if the apportionment provisions do not fairly represent the extent of the taxpayer's business activity in the state. The model Uniform Division of Income for Tax Purposes Act includes such a power in Section 18. In fact, these statutory adjustment powers typically also provide the option of using separate accounting or any other method that will effectuate an equitable apportionment of the tax base.

Taxpayers have successfully argued that discretionary apportionment relief must be provided when the normal statutory apportionment formula would result in unconstitutional or otherwise excessive taxation. In the matter of Just Born Inc., the New York City Tax Appeals Tribunal found that because the taxpayer's two income streams did not flow from a single unitary business, the standard apportionment formula under the City's general corporation tax could not be applied to the taxpayer's total income. (74) The Tribunal agreed with the Administrative Law Judge that the taxing authority improperly refused to employ its discretionary authority to employ a "different [apportionment] method calculated to apply a fair and proper allocation of the income ... reasonably attributable to the City." The Tribunal concluded that "separate accounting, that takes into account only [the income stream properly taxable by the City], is the appropriate computational model to capture the income fairly attributable to the City" for general corporation tax purposes.

Discretionary adjustment power has been applied by some judicial or quasi-judicial bodies as further justification for awarding a taxpayer relief through statutory interpretation. In Grumman Corporation, an Administrative Law Judge found both that the statutory provision in question was not meant to be applied in the manner suggested by the Commissioner under "the very unique circumstances" of the case and also that "[a]t the very least, given the underlying purpose of the statute, the Commissioner should have exercised his discretion by allowing the deduction in these circumstances." (75) In contrast, taxpayers have been denied adjustment where no distortion was found and thus "no justification or legal authority for a discretionary adjustment" existed. (76)

On the flip side, a taxing authority is similarly able to employ its powers of discretionary adjustment to its benefit when the standard apportionment formula fails to fairly represent a taxpayer's income. For example, in Union Pacific Corp., the Idaho Supreme Court upheld the Tax Commission's application of an alternative apportionment formula to exclude the taxpayer's sales of accounts receivable from its sales factor denominator. (77) Similarly, in Microsoft Corp., a California appellate court found it did not even need to reach a statutory interpretation issue about the scope of the sales factor for returns on short-term investments because the court found it could decide in favor of the Franchise Tax Board under the discretionary adjustment power. (78) These cases raise particular concern in that they allow the taxing authorities to apply apportionment rules that have broad applicability on an ad hoc basis under discretionary adjustment powers. If a taxing authority seeks a deviation from the normal apportionment rules that will have general applicability for a class of taxpayers (as opposed to an isolated adjustment for a single taxpayer with a unique set of facts), Due Process notice principles and governmental administrative procedure obligations ought to require promulgation of a formal rule or, if necessary, a statutory change.

Taxpayers must be wary if they plan to seek to take advantage of a discretionary adjustment. Many states require that taxpayers petition for permission to use an alternative formula, and create strict time deadlines (such as prior to the filing of the original return) within which such applications must be submitted. (79)

Conclusion

Although governments do have broad discretion in selecting, and applying, apportionment formulae, that discretion certainly is not without limitation under the Constitution and various principles of state law. One must be aware of those constitutional and other legal limitations in order to be able to identify apportionment approaches that go too far.

(1) Jerome R. Hellerstein & Walter Hellerstein, State and Local Taxation 410-11 (6th ed. 1997).

(2) Trinova Corp. v. Michigan Department of Treasury, 498 U.S. 358, 378 (1991) (quoting Mobil Oil Corp. v. Commissioner of Taxes of Vermont, 445 U.S. 425, 438 (1980)).

(3) Trinova Corp, 498 U.S. at 378 (quoting Mobil Oil Corp., 445 U.S. at 438).

(4) Many states apply formulary apportionment only to "business income," with "non-business income" being specifically assigned to a single state. In addition, a state may be constitutionally barred from taxing a corporation's non-operational income. Allied-Signal, Inc. v. Director, Division of Taxation; 504 U.S. 768 (1992).

(5) Trinova Corp., 498 U.S. at 373 (citing Underwood Typewriter Co. v. Chamberlain, 254 U.S. 113, 120-121 (1920) (legislature "'faced with the impossibility of allocating specifically the profits earned by the [taxpayer's] processes conducted within its borders.'")).

(6) Container Corp. of America v. Franchise Tax Board, 463 U.S. 159, 192 (1983).

(7) Moorman Manufacturing Co. v. Bair, 437 U.S. 267, 278 (1978).

(8) Moorman Manufacturing Co., 437 U.S. at 280.

(9) Complete Auto Transit, Inc. v. Brady, 430 U.S. 274, 279 (1977).

(10) Container Corp., 463 U.S. at 169.

(11) Armco, Inc. v. Hardesty, 467 U.S. 638, 644-45 (1984).

(12) Tyler Pipe Indus. v. Washington Department of Revenue, 483 U.S. 232 (1987).

(13) American Trucking Association v. Scheiner, 483 U.S. 266 (1987).

(14) But see E.I. du Pont Nemours & Co. v. State Tax Assessor, 675 A.2d 82 (Me. 1996), abandoning the court's earlier internal consistency decision in Tambrands v. State Tax Assessor, 595 A.2d 1039 (Me. 1991).

(15) American Trucking Associations v. State of New Jersey, 852 A.2d 142 (N.J. July 19, 2004).

(16) Sprint Communications Co., Cable & Wireless Commissions, Inc. v. Kelly, 642 A.2d 106 (D.C. Ct. App. 1994), cert. den., 513 U.S. 916 (1994).

(17) M & Associates, Inc. v. City of Irondale, 723 So. 2d 592 (Ala. 1998).

(18) Container Corp., 463 U.S. at 169.

(19) Philadelphia Eagles Football Club, Inc. v. City of Philadelphia, 823 A.2d 108 (Pa. 2003).

(20) Norfolk & Western Ry. Co. v. Missouri, 390 U.S. 317 (1968) (testing for a grossly distorted result); Moorman Manufacturing Co., 437 U.S. 267.

(21) 283 U.S. 123 (1931).

(22) Trinova Corp., 498 U.S. at 373 (citing Container Corp., 463 U.S. at 180-181).

(23) Trinova Corp., 498 U.S. at 380 (citing Hans Rees' Sons, Inc., 283 U.S. at 135; Norfolk & Western Ry. Co., 390 U.S. at 326; Moorman Manufacturing Co., 437 U.S. at 274).

(24) Trinova Corp., 498 U.S. at 384.

(25) Complete Auto Transit, 430 U.S. at 279.

(26) 467 U.S. at 644.

(27) Metro. Life Ins. Co. v. Ward, 470 U.S. 869, 878 (1985).

(28) Container Corp., 463 U.S. at 170.

(29) Trinova Corp., 498 U.S. at 380.

(30) Trinova Corp., 498 U.S. at 381.

(31) Container Corp., 463 U.S. at 183.

(32) Underwood Typewriter Co. v. Chamberlain, 254 U.S. 113 (1920).0 See also Hans Rees" Sons, Inc., 283 U.S. 123.

(33) Moorman Manufacturing Co. v. Bair, 437 U.S. 267 (1978). Although the Supreme Court refused to enforce a single-factor sales formula in General Motors Corp. v. District of Columbia, 380 U.S. 553 (1965), the Court did not rule on constitutional grounds but instead concluded that the regulation prescribing the formula was not authorized by the District of Columbia Code.

(34) Moorman Manufacturing Co., 437 U.S. at 277.

(35) The Multistate Tax Commission position is that "[i]f any sales are made into a state which is precluded by P.L. 86-272 from taxing the income of the seller, such sales remain subject to throwback to the appropriate state which does have jurisdiction to impose its net income tax upon the income derived from those sales." Statement of Information Concerning Practices of Multistate Tax Commission and Signatory States Under Public Law 86-272 (July 27, 2001).

(36) In contrast, for purposes of Utah's throwback rule, the taxpayer need not throw back receipts from shipments destined to a state that does not impose a tax. The taxpayer is considered "taxable in the state of the purchaser" for purposes of the throwback rule in Utah Code Ann. [section] 59-7-318(2), as long as "that state has jurisdiction to subject the taxpayer to a net income tax regardless of whether, in fact, the state does or does not." Utah Code Ann. [section] 59-7-305.

(37) Indiana Department of Revenue, Letter of Findings Nos. 01-127 and 01-128 (June 1, 2004).

(38) In Utah (a UDITPA state), the throwback rule provides that "[s]ales of tangible personal property are in this state if ... the property is shipped from an office, store, warehouse, factory, or other place of storage in this state, and: (a) the purchaser is the United States Government; or (b) the taxpayer is not taxable in the state of the purchaser." Utah Code Ann. [section] 59-7-318(2).

(39) Indiana's throwback rule can apply to sales made into foreign countries. Rule 45 IAC 3.1-1-64.

(40) Trinova Corp., 498 U.S. at 380; Norfolk & Western Ry. Co., 390 U.S. 317; Moorman Manufacturing Co., 437 U.S. 267.

(41) Colgate-Palmolive Co. v. Massachusetts Commissioner of Revenue, No. C255116 (App. Tax Bd. April 3, 2003). See also Cambridge Brands, Inc. v. Massachusetts Commissioner of Revenue, No. C259013 (App. Tax Bd. July 16, 2003).

(42) Mass. G.L.c. 63, [section]38(f)(2).

(43) Colgate-Palmolive Co., supra.

(44) Id.

(45) Magnetek Controls, Inc. v. Michigan Department of Treasury, 562 N.W.2d 219 (Mich. Ct. App. 1997); Michigan Sugar Co. v. Michigan Department of Treasury, Tax Tribunal No. 220328 (Jan. 1, 1998) (but no substantial nexus existed in states into which only a few sporadic trips were made).

(46) Morton International Inc. v. Illinois Department of Revenue, No. 01 L 50752 (Ill. Cir. Ct., Cook Cty., July 8, 2004).

(47) Id. (citing 35 ILCS 5/304(a)(3)(B)(ii)).

(48) No.88-SBE-022-A (Cal. SBE Jan. 24, 1990).

(49) Cal. SBE, Nov. 11, 1966.

(50) No. 88-SBE-O22-A (Aug. 25, 1988).

(51) No. 88-SBE-O22-A (Jan. 24, 1990).

(52) Appeal of Huffy Corp.(Cal. SBE April 22, 1999). See also Citicorp North America Inc. v. California Franchise Tax Board, 83 Cal.App.4th 1403 (2000), review denied (Cal. S.Ct., Jan. 10, 2001) (involving the application of Finnigan in the wake of the return to the Joyce rule).

(53) Alpharma, Inc., NYS Division of Tax Appeals, AL3 Unit, TA No. 817895 (ALJ Unit Sept. 12, 2002); Disney Enterprises, Inc. & Combined Subsidiaries, DTA No. 818378 (Feb. 12, 2004).

(54) NYS Division of Tax Appeals, Tax Appeals Tribunal, DTA No. 817895 (Aug. 5, 2004).

(55) NYS Division of Tax Appeals, ALJ Unit, DTA No. 818378 (Feb. 12, 2004).

(56) Texas Comptroller's Decision, Hearing No. 38,716 (June 15, 2000) (citing Comptroller's Rule 3.557(e)(37)(1)).

(57) Follett Corp. v. Illinois Department of Revenue, No. 4-02-0938 (Ill. App. Ct., Nov. 13, 2003).

(58) See 35 ILCS 5/304(a)(3)(B).

(59) WV Code See. 11-24-7(e)(11).

(60) N.J.S.A. 54:10A-6.

(61) N.J.A.C. 18:7-8.7(t).

(62) N.J.S.A. 54:10A-5(h).

(63) N.J.A.C. 18:7-8.7.

(64) New Jersey Division of Taxation, Q&A Regarding the Business Tax Reform Act of 2002.

(65) Hellertown Manufacturing Co. v. Commissioner, 480 Pa. 348, 390 A.2d 732 (1978).

(66) Paris Manufacturing Co. v. Commissioner, 505 Pa. 15,476 A.2d

890 (1984). On different legal grounds, one taxpayer challenged the Connecticut Commissioner's application of a throwout rule as a rule of general applicability that was required to have been (but was not) formally adopted as a regulation under the Connecticut Administrative Procedure Act. Control Module, Inc. v. Commissioner of Revenue Services, 41 Conn. Sup. 271, 567 A.2d 1264 (Conn. Super. 1989).

(67) Id. (emphasis added).

(68) As mentioned above, one Connecticut taxpayer argued that the state's Administrative Procedure Act prevented the Commissioner's imposition of a throwout rule with respect to the taxpayer's apportionment fraction. Control Module, Inc., 41 Conn. Sup. 271, 567 A.2d 1264.

(69) Metromedia, Inc. v. Director, Division of Taxation, 97 N.J. 313 (1984).

(70)Trinova Corp., 498 U.S. at 384.

(71) Trinova Corp., 498 U.S. at 384.

(72) Moorman Manufacturing Co., 437 U.S. at 272.

(73) Trinova Corp., 498 U.S. at 371-72 (discussing the Michigan Supreme Court use of the phrase).

(74) In the Matter of Just Born Inc., New York City Tax Appeals Tribunal, TAT (E) 93-456 (GC) (Mar. 30, 1998), motion to vacate denied, TAT(E)93-456R(GC) (Feb. 22, 1999).

(75) In the Matter of Grumman Corporation, NYS Division of Tax Appeals, ALJ Unit, DTA No. 813147 (July 11, 1996).

(76) In the Matter of Barclays Group, Inc., NYS Division of Tax Appeals, Tax Appeals Tribunal, DTA No. 818789 (Jan. 27, 2005).

(77) Union Pacific Corp. v. Idaho State Tax Commission, 83 P.2d 116 (Id., Jan. 4, 2004).

(78) Microsoft Corp. v. Franchise Tax Board, No.A105312 (Ca. 1st App. Dist. 2005) (appeal pending).

(79) See, e.g., Leathers v. Jacuzzi, Inc., 935 S.W.2d 252 (Ark. 1996) (corporations had neither requested nor received permission from the Commissioner to file combined returns; Arkansas Supreme Court reversed lower court decision that combination was mandated in order to achieve a clear reflection of income and expenses as "the judicial branch attempt[ing] to exercise the power of the executive branch").

MARGARET C. WILSON is a partner with McDermott Will & Emery LLP in New York. She received her B.A. and J.D. degrees from the University of Michigan, and her LL.M (Taxation) degree from New York University. She wrote the Corporation Business Tax chapter of the New Jersey Tax Handbook and a member of the editorial board of the Journal of Multistate Taxation and Incentives. This is her first article for The Tax Executive.