To apportion a multistate taxpayer’s income, state statutes and regulations provide standard apportionment formulas that are intended to produce an approximation of a taxpayer’s income attributable to the state. In certain instances, however, the standard formula does not provide a reasonable division of the income tax base, and an alternative method of dividing income among the states in which the taxpayer does business may be permitted or required.

Most states permit or require deviations from the standard formula in special circumstances through alternative apportionment (sometimes called “equitable apportionment”). Most states’ alternative apportionment provisions are modeled after Section 18 of UDITPA.1 The Multistate Tax Commission issued regulations interpreting UDITPA Section 18 stating that it is designed to serve as an exception and should only apply in limited factual situations when deviation from the standard formula is necessary to fairly reflect the taxpayer’s activities in the state. According to UDITPA § 18:

If the allocation and apportionment provisions of this Act do not fairly represent the extent of the taxpayer’s business activity in this state, the taxpayer may petition for or the tax administrator may require, in respect to all or any part of the taxpayer’s business activity, if reasonable:

  1. Separate accounting;
  2. The exclusion of any one or more of the factors;
  3. The inclusion of one or more additional factors that will fairly represent the taxpayer’s business activity in this state; or
  4. The employment of any other method to effectuate an equitable allocation and apportionment of the taxpayer’s income.

Generally, the first step for a party attempting to invoke alternative apportionment is to prove that use of the standard apportionment method does not fairly reflect the taxpayer’s in-state business activity.2 The second step in successfully invoking alternative apportionment—the step that is the focus of this article—is to prove that the proposed alternative method is a reasonable approximation of the taxpayer’s activities and income in the state.3 Under UDITPA Section 18, this requires a showing that the proposed method is both reasonable and equitable.

In the many years since UDITPA was drafted, few courts or commentators have attempted to define or explain what makes a proposed alternative apportionment method a “reasonable” one. The most notable attempt to provide explicit guidance for determining the reasonableness of a proposed alternative is found in the Oregon Supreme Court’s 1985 decision in Twentieth Century Fox.4 That court wrote that

in the context of UDITPA, reasonableness has at least three components: (1) the division of income fairly represents business activity and if applied uniformly would result in taxation of no more or no less than 100 percent of taxpayer’s income; (2) the division of income does not create or foster lack of uniformity among UDITPA jurisdictions; and (3) the division of income reflects the economic reality of the business activity engaged in by the taxpayer in Oregon.5

Insofar as the Oregon court’s guidance helps taxing authorities, taxpayers and courts test the reasonableness of a proposed alternative method, it is better than no guidance at all. However, that three-factor test provides essentially no guidance on the front end for deciding which alternative method should be applied by a taxing authority or a taxpayer to the particular set of facts at hand, instead implicitly suggesting that once the first threshold for the imposition of alternative apportionment has been satisfied, any apportionment method that is occasionally applied by taxing jurisdictions will be an acceptably “reasonable” alternative, regardless of the facts.

Furthermore, the Oregon court’s three-factor test is essentially impossible to satisfy anyway, given that deviating from the standard UDITPA apportionment method necessarily fosters a “lack of uniformity among UDITPA jurisdictions.”

Thus, if Twentieth Century Fox’s three-factor test is not a useful guide, what set of guidelines should a party use in trying to choose a reasonable alternative apportionment method? While it is improbable that any single test or multi-factor guidelines could ever sufficiently anticipate all potential scenarios, a close review of the cases in which alternative apportionment has been upheld reveals that when courts readily accept a proposed alternative apportionment method, it is often because the proposed alternative method satisfies a narrowly drawn test. As described above, for a taxing authority or taxpayer to successfully invoke alternative apportionment, it must prove that the standard formula does not “fairly represent” that taxpayer’s business activity in the state. It follows, then, that a “reasonable” alternative method should not simply be any apportionment method drawn from the universe of generally acceptable apportionment methods; rather, a proposed alternative method should be accepted as “reasonable” when the party asserting alternative apportionment is able to tie the proposed alternative method to the basis for deviating from the standard formula.

While no court appears to have explicitly advanced or endorsed this approach, an examination of the facts and reasoning in a number of cases in which alternative apportionment has been upheld reveals that a close connection between the basis for deviating from the standard apportionment method and the proposed alternative is a common feature of those cases. For example, in Microsoft,6 the California Supreme Court addressed whether the taxpayer’s sales factor was distorted by the inclusion in the denominator of gross receipts from the sale of short-term securities conducted by the company’s treasury department. The court found that “operation of a large treasury department unrelated to a taxpayer’s main business [of selling software] is a paradigmatic example of circumstances warranting” the use of alternative apportionment; such treasury operations frequently distort the sales factor, the court explained, because the taxpayer can engage in a large number of financial transactions that generate gross receipts that are several orders of magnitude larger than the net income they produce.7 Because those gross receipts from Microsoft’s treasury function “seriously distort[ed]” the ability of the sales factor to represent Microsoft’s business activities in the state, the court held that the state had met its burden to utilize alternative apportionment and to adopt an alternative sales factor in which only the net receipts from treasury operations were included.8 

Similarly, when the taxpayer in Twentieth Century Fox—a distributor of motion pictures—computed the property factor of its standard apportionment formula, it only included the nominal cost (hundreds of dollars) of its “positive prints” of film located in Oregon and excluded the much more costly value (millions of dollars) ascribed to the “film negatives.”9 The department contended that an alternative property factor which included the value of the negatives was necessary to reflect the true value of Fox’s property located in Oregon; the court agreed, upholding the department’s assertion of alternative apportionment on the basis that including the full production value of the film in the property factor would more accurately represent the value of Fox’s property in the state by reflecting the “economic reality of the distribution of motion pictures . . . .”10

The decision of the Illinois Court of Appeals in Miami Corp.11 further reinforces the thesis that a “reasonable” alternative apportionment method is one that is tied to the basis for deviating from the standard formula in the first place. Miami Corporation had an office in Chicago from which employees managed the sale of timber rights on land owned outside Illinois. It also had completely separate operations in Louisiana, from which an employee managed Louisiana real estate responsible for substantial oil and gas reserves that generated more than 80 percent of the taxpayer’s income.12 Because those oil and gas reserves were considered intangible property, they were excluded from Illinois’s standard property factor; however, the court held that the standard property factor’s failure to account for those substantial out-of-state intangibles caused it to be distortive of the taxpayer’s business activity in Illinois. Similarly, because Illinois’s standard payroll factor did not account for the taxpayer’s costs in engaging independent contractors in other states to manage the taxpayer’s holdings, that factor also did not fairly represent the taxpayer’s specific activities within Illinois. Id. Accordingly, the court upheld the taxpayer’s requested deviation from the standard formula to use a separate accounting formula that would isolate the income attributable to the taxpayer’s Illinois business activities.13

Similarly, in GATX Corp.,14 the Ohio Court of Appeals held that a taxpayer was entitled to use alternative apportionment to eliminate the payroll factor from the standard three-factor formula, noting that the taxpayer had both manufacturing and leasing operations, but that the “bulk of appellant’s income” was derived from its leasing operations. However, because the taxpayer’s manufacturing operations in Ohio were so labor-intensive, they produced a payroll factor that was “thirteen times the property factor and twentyseven times the sales factor.”15 Thus, because the taxpayer’s payroll caused its standard three-factor formula to be “out of focus,” the taxpayer was entitled to remove the payroll factor from its standard apportionment formula.16 Finally, in Union Pacific,17 the Idaho Supreme Court reached a similar conclusion. In that case, the standard sales factor permitted the taxpayer to include both accounts receivable and money received from the sale of those accounts in the denominator of its sales factor.

Thus, where those two types of receipts included in the standard sales factor were largely duplicative of each other, the court upheld alternative apportionment using an alternative sales factor that eliminated one of the two types of receipts.18


Accordingly, based on the reasoning and holdings in Microsoft, Twentieth Century Fox, Miami Corporation, GATX, Union Pacific and other similar cases, a proposed alternative method is most likely to be accepted as “reasonable” when the party asserting alternative apportionment is able to show a close connection between the basis for deviating from the standard formula and the proposed alternative method. While the approach identified in this article has not been expressly stated by any court or other authority, taxing authorities or state courts could easily rely on the case law discussed in this article as persuasive authority to establish this new approach. Without a doubt, recognition of the need for a connection between a proposed alternative apportionment method and the basis for deviating from the standard method would be a significant improvement for alternative apportionment guidance and jurisprudence. Though the abuse of alternative apportionment provisions by taxing authorities stands as a present and future threat to multistate taxpayers, more precise guidance regarding what alternative methods will be accepted as “reasonable” could go a long way toward reining in those abuses.