Argued March 21, 1978 Decided June 15, 1978
An Iowa statute prescribes a so-called single-factor sales formula for apportioning an interstate corporation's income for state income tax purposes. Under this formula, the part of income from such a corporation's sale of tangible personal property attributable to business within the State and hence subject to the state income tax is deemed to be in that proportion which the corporation's gross sales made within the State bear to its total gross sales. Appellant, an Illinois corporation that sells animal feed it manufactures in Illinois to Iowa customers through Iowa salesmen and warehouses, brought an action in an Iowa court challenging the constitutionality of the single-factor formula. The trial court held the formula invalid under the Due Process Clause of the Fourteenth Amendment and the Commerce Clause, but the Iowa Supreme Court reversed. Held:
1. Iowa's single-factor formula is not invalid under the Due Process Clause. Pp. 271-275.
(a) Any assumption that at least some portion of appellant's income from Iowa sales was generated by Illinois activities is too speculative to support a claim that Iowa in fact taxed profits not attributable to activities within the State. P. 272.
(b) An apportionment formula, such as the single-factor formula, that is necessarily employed as a rough approximation of a corporation's income reasonably related to the activities conducted within the taxing State will only be disturbed when the taxpayer has proved by "clear and cogent evidence" that the income attributed to the State is in fact "out of all reasonable proportion to the business transacted . . . in that State," Hans Rees' Sons v. North Carolina ex rel. Maxwell, 283 U.S. 123, 135, or has "led to a grossly distorted result," Norfolk Western R. Co. v. State Tax Comm'n, 390 U.S. 317, 326. Here, the Iowa statute afforded appellant an opportunity to demonstrate that the single-factor formula produced an arbitrary result in its case, but the record contains no such showing. Pp. 272-275.
2. Nor is Iowa's single-factor formula invalid under the Commerce Clause. Pp. 276-281.
(a) On this record, the existence of duplicative taxation as between Iowa and Illinois (which uses the so-called three-factor — property, payroll, and sales — formula) is speculative, but even assuming some overlap, appellant's argument that Iowa, rather than Illinois, was necessarily at fault in a constitutional sense cannot be accepted. Where 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. Pp. 276-277.
(b) The Commerce Clause itself, without implementing legislation by Congress, does not require, as appellant urges, that Iowa compute corporate net income under the Illinois three-factor formula. If the Constitution were read to mandate a prohibition against any overlap in the computation of taxable income by the States, the consequences would extend far beyond this particular case and would require extensive judicial lawmaking. Pp. 277-281.
254 N.W.2d 737, affirmed.
STEVENS, J., delivered the opinion of the Court, in which BURGER, C. J., and STEWART, WHITE, MARSHALL, and REHNQUIST, JJ., joined. BRENNAN, J., post, p. 281, and BLACKMUN, J., post, p. 282, filed dissenting opinions. POWELL, J., filed a dissenting opinion, in which BLACKMUN, J., joined, post, p. 283.
Donald K. Barnes argued the cause for appellant. With him on the briefs were Walter R. Brown, John V. Donnelly, Carl G. Schmiedeskamp, and Robert W. Cook.
Harry M. Griger, Assistant Attorney General of Iowa, argued the cause for appellee. With him on the brief was Richard C. Turner, Attorney General.
Ernest S. Christian, Jr., and Allan Abbot Tuttle filed a brief for the Committee on State Taxation of the Council of State Chambers of Commerce as amicus curiae urging reversal.
James L. Rogers, John R. Phillips, and Philip B. Kurland filed a brief for the Iowa Manufacturers Assn. et al. as amici curiae urging affirmance.
Willam D. Dexter, James A. Redden, Attorney General of Oregon, and Theodore W. deLooze, Assistant Attorney General, filed a brief for the Multistate Tax Comm'n et al. as amici curiae.
The question in this case is whether the single-factor sales formula employed by Iowa to apportion the income of an interstate business for income tax purposes is prohibited by the Federal Constitution.
Appellant, Moorman Manufacturing Co., is an Illinois corporation engaged in the manufacture and sale of animal feeds. Although the products it sells to Iowa customers are manufactured in Illinois, appellant has over 500 salesmen in Iowa and it owns six warehouses in the State from which deliveries are made to Iowa customers. Iowa sales account for about 20% of appellant's total sales.
Corporations, both foreign and domestic, doing business in Iowa are subject to the State's income tax. The taxable income for federal income tax purposes, with certain adjustments, is treated as the corporation's "net income" under the Iowa statute. If a corporation's business is not conducted entirely within Iowa, the statute imposes a tax only on the portion of its income "reasonably attributable" to the business within the State.
There are essentially two steps in computing the share of a corporation's income "reasonably attributable" to Iowa. First, certain income, "the geographical source of which is easily identifiable," is attributed entirely to a particular State. Second, if the remaining income is derived from the manufacture or sale of tangible personal property, "the part thereof attributable to business within the state shall be in that proportion which the gross sales made within the state bear to the total gross sales." This is the single-factor formula that appellant challenges in this case.
The statute provides:
"Interest, dividends, rents, and royalties (less related expenses) received in connection with business in the state, shall be allocated to the state, and where received in connection with business outside the state, shall be allocated outside of the state." Iowa Code § 422.33(1) (a) (1977).
In describing this section, the Iowa Supreme Court stated that "certain income, the geographical source of which is easily identifiable, is allocated to the appropriate state." 254 N.W.2d 737, 739. Thus, for example, rental income would be attributed to the State where the property was located. And in appellant's case, this section operated to exclude its investment income from the tax base.
Iowa Code § 422.33(1)(b) (1977).
If the taxpayer believes that application of this formula subjects it to taxation on a greater portion of its net income than is "reasonably attributable" to business within the State, it may file a statement of objections and submit an alternative method of apportionment. If the evidence submitted by the taxpayer persuades the Director of Revenue that the statute is "inapplicable and inequitable" as applied to it, he may recalculate the corporation's taxable income.
During the fiscal years 1949 through 1960, the State Tax Commission allowed appellant to compute its Iowa income on the basis of a formula consisting of three, equally weighted factors — property, payroll, and sales — rather than the formula prescribed by statute. For the fiscal years 1961 through 1964, appellant complied with a directive of the State Tax Commission to compute its income in accordance with the statutory formula. Since 1965, however, appellant has resorted to the three-factor formula without the consent of the commission.
The operation of the two formulas may be briefly described. The single-factor sales formula yields a percentage representing a ratio of gross sales in Iowa to total gross sales. The three-factor formula yields a percentage representing an average of three ratios: property within the State to total property, payroll within the State to total payroll, and sales within the State to total sales.
These percentages are multiplied by the adjusted total net income to arrive at Iowa taxable net income. This net income figure is then multiplied by the tax rate to compute the actual tax obligation of the taxpayer.
In 1974, the Iowa Director of Revenue revised appellant's tax assessment for the fiscal years 1968 through 1972. This assessment was based on the statutory formula, which produced a higher percentage of taxable income than appellant, using the three-factor formula, had reported on its return in each of the disputed years. The higher percentages, of course, produced a correspondingly greater tax obligation for those years.
For those years the two formulas resulted in the following percentages:
Fiscal Year Sales Factor Three-Factor Ended Percentage Percentage — --------- ------------ ------------ 3/31/68 21.8792% 14.1088% 3/31/69 21.2134% 14.3856% 3/31/70 19.9492% 14.0200% 3/31/71 18.9544% 13.2186% 3/31/72 18.6713% 12.2343%
For a description of how these percentages are computed, see n. 3, supra.
Thus, in 1968, for example, Moorman's three-factor computation resulted in a tax of $81,466, whereas the Director's single-factor computation resulted in a tax of $121,363.
After the Tax Commission had rejected Moorman's appeal from the revised assessment, appellant challenged the constitutionality of the single-factor formula in the Iowa District Court for Polk County. That court held the formula invalid under the Due Process Clause of the Fourteenth Amendment and the Commerce Clause. The Supreme Court of Iowa reversed, holding that an apportionment formula that is necessarily only a rough approximation of the income properly attributable to the taxing State is not subject to constitutional attack unless the taxpayer proves that the formula has produced an income attribution "out of all proportion to the business transacted" within the State. The court concluded that appellant had not made such a showing.
We noted probable jurisdiction of Moorman's appeal, 434 U.S. 953, and now affirm.
Appellant contends that Iowa's single-factor formula results in extraterritorial taxation in violation of the Due Process Clause. This argument rests on two premises: first, that appellant's Illinois operations were responsible for some of the profits generated by sales in Iowa; and, second, that a formula that reaches any income not in fact earned within the borders of the taxing State violates due process. The first premise is speculative and the second is foreclosed by prior decisions of this Court.
Appellant does not 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. But appellant contends that we should proceed on the assumption that at least some portion of the income from Iowa sales was generated by Illinois activities.
Whatever merit such an assumption might have from the standpoint of economic theory or legislative policy, it cannot support a claim in this litigation that Iowa in fact taxed profits not attributable to activities within the State during the years 1968 through 1972. For all this record reveals, appellant's manufacturing operations in Illinois were only marginally profitable during those years and the high-volume sales to Iowa customers from Iowa warehouses were responsible for the lion's share of the income generated by those sales. Indeed, a separate accounting analysis might have revealed that losses in Illinois operations prevented appellant from earning more income from exploitation of a highly favorable Iowa market. Yet even were we to assume that the Illinois activities made some contribution to the profitability of the Iowa sales, appellant's claim that the Constitution invalidates an apportionment formula whenever it may result in taxation of some income that did not have its source in the taxing State is incorrect.
The Due Process Clause places two restrictions on a State's power to tax income generated by the activities of an interstate business. First, no tax may be imposed unless there is some minimal connection between those activities and the taxing State. National Bellas Hess, Inc. v. Department of Revenue, 386 U.S. 753, 756. This requirement was plainly satisfied here. Second, the income attributed to the State for tax purposes must be rationally related to "values connected with the taxing State." Norfolk Western R. Co. v. State Tax Comm'n, 390 U.S. 317, 325.
Since 1934 Iowa has used the formula method of computing taxable income. This method, unlike separate accounting, does not purport to identify the precise geographical source of a corporation's profits; rather, it is employed as a rough approximation of a corporation's income that is reasonably related to the activities conducted within the taxing State. The single-factor formula used by Iowa, therefore, generally will not produce a figure that represents the actual profits earned within the State. But the same is true of the Illinois three-factor formula. Both will occasionally over-reflect or under-reflect income attributable to the taxing State. Yet despite this impression, the Court has refused to impose strict constitutional restraints on a State's selection of a particular formula.
Thus, we have repeatedly held that a single-factor formula is presumptively valid. In Underwood Typewriter Co. v. Chamberlain, 254 U.S. 113, for example, the taxpayer challenged Connecticut's use of such a formula to apportion its net income. Underwood's manufacturing operations were conducted entirely within Connecticut. Its main office, however, was in New York City and it had branch offices in many States where its typewriters were sold and repaired. Applying a single-factor property formula, Connecticut taxed 47% of the company's net income. Claiming that 97% of its profits were generated by transactions in tangible personal property outside Connecticut, Underwood contended that the formula taxed "income arising from business conducted beyond the boundaries of the State" in violation of the Due Process Clause. Id., at 120.
Rejecting this claim, the Court noted that Connecticut "adopted a method of apportionment which, for all that appears in this record, reached, and was meant to reach, only the profits earned within the State," id., at 121, and held that the taxpayer had failed to carry its burden of proving that "the method of apportionment adopted by the State was inherently arbitrary, or that its application to this corporation produced an unreasonable result." Ibid. (footnote omitted).
See also Bass, Ratcliff Gretton, Ltd. v. State Tax Comm'n, supra; Norfolk Western R. Co. v. North Carolina ex rel. Maxwell, 297 U.S. 682.
In individual cases, it is true, the Court has found that the application of a single-factor formula to a particular taxpayer violated due process. See Hans Rees' Sons, Inc. v. North Carolina ex rel. Maxwell, 283 U.S. 123; Norfolk Western R. Co. v. State Tax Comm'n, supra. In Hans Rees', for example, the Court concluded that proof that the formula produced a tax on 83% of the taxpayer's income when only 17% of that income actually had its source in the State would suffice to invalidate the assessment under the Due Process Clause. But in neither Hans Rees' nor Norfolk Western did the Court depart from the basic principles that the States have wide latitude in the selection of apportionment formulas and that a formula-produced assessment will only be disturbed when the taxpayer has proved 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," 283 U.S., at 135, or has "led to a grossly distorted result," 390 U.S., at 326.
General Motors Corp. v. District of Columbia, 380 U.S. 553, on which appellant relies, does not suggest a contrary result. In that case the Court held that a regulation prescribing a single-factor sales formula was not authorized by the District of Columbia Code. It concluded that the formula violated the statutory requirement that the net income of a corporation doing business both inside and outside the District must be deemed to arise from "sources" both inside and outside the District. But that statutory requirement has no counterpart in the Constitution, and the Court in General Motors made clear that it did "not mean to take any position on the constitutionality of a state income tax based on the sales factor alone." Id., at 561.
The Court, it is true, expressed doubts about the wisdom of the economic assumptions underlying the challenged formula and noted that its use in the context of the more prevalent three-factor formula would not advance the policies underlying the Commerce Clause. But these considerations were deemed relevant to the question of legislative intent, not constitutional interpretation.
The Iowa statute afforded appellant an opportunity to demonstrate that the single-factor formula produced an arbitrary result in its case. But this record contains no such showing and therefore the Director's assessment is not subject to challenge under the Due Process Clause.
In his concurring opinion, Justice McCormick of the Iowa Supreme Court made this point:
"In the present case, Moorman did not attempt to prove the amount of its actual net income from Iowa activities in the years involved. Therefore no basis was presented for comparison of the corporation's Iowa income and the income apportioned to Iowa under the formula. In this era of sophisticated accounting techniques, it should not be impossible for a unitary corporation to prove its actual income from activities in a particular state. However, Moorman showed only that its tax liability would be substantially less if Iowa employed a three-factor apportionment formula. We have no basis to assume that the three-factor formula produced a result equivalent to the corporation's actual income from Iowa activities. Having failed to establish a basis for comparison of its actual income in Iowa with the income apportioned to Iowa under the single-factor formula, Moorman did not demonstrate that the single-factor formula Page 276 produced a grossly unfair result. Thus it did not prove unconstitutionality of the formula as applied." 254 N.W.2d, at 757.
Appellant also contends that during the relevant years Iowa and Illinois imposed a tax on a portion of the income derived from the Iowa sales that was also taxed by the other State in violation of the Commerce Clause. Since most States use the three-factor formula that Illinois adopted in 1970, appellant argues that Iowa's longstanding single-factor formula must be held responsible for the alleged duplication and declared unconstitutional. We cannot agree.
Since Illinois did not adopt its income tax until 1970, there was no possibility of any overlap until that year. The alleged overlap in the three years following Illinois' enactment of an income tax was 34.38% in 1970, 34.51% in 1971, and 37.01% in 1972.
In the first place, this record does not establish the essential factual predicate for a claim of duplicative taxation. Appellant's net income during the years in question was approximately $9 million. Since appellant did not prove the portion derived from sales to Iowa customers, rather than sales to customers in other States, we do not know whether Illinois and Iowa together imposed a tax on more than 100% of the relevant net income. The income figure that appellant contends was subject to duplicative taxation was computed by comparing gross sales in Iowa to total gross sales. As already noted, however, this figure does not represent actual profits earned from Iowa sales. Obviously, all sales are not equally profitable. Sales in Iowa, although only 20% of gross sales, may have yielded a much higher percentage of appellant's profits. Thus, profits from Iowa sales may well have exceeded the $2.5 million figure that appellant contends was taxed by the two States. If so, there was no duplicative taxation of the net income generated by Iowa sales. In any event, on this record its existence is speculative.
Since there is no evidence in the record regarding the percentages of its total net income taxed in the other States in which it did business Page 277 during those years, any claim that appellant was taxed on more than 100% of its total net income would also be speculative.
Even assuming some overlap, we could not accept appellant's argument that Iowa, rather than Illinois, was necessarily at fault in a constitutional sense. It is, of course, true that if Iowa had used Illinois' three-factor formula, a risk of duplication in the figures computed by the two States might have been avoided. But the same would be true had Illinois used the Iowa formula. Since 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. Rather, the claim must be that even if the presumptively valid Iowa formula yielded no profits other than those properly attributable to appellant's activities within Iowa, the importance of avoiding any risk of duplication in the taxable income of an interstate concern justifies invalidation of the Iowa statute.
Appellant contends that, to the extent this overlap is permitted, the corporation that does business in more than one State shoulders a tax burden not shared by those operating entirely within a State. To alleviate the burden, appellant invites us to hold that the Commerce Clause itself, without implementing legislation by Congress, requires Iowa to compute corporate net income under the Illinois equally weighted, three-factor formula. For the reasons that follow, we hold that the Constitution does not require such a result.
Appellant also contends that the Iowa formula discriminates against interstate commerce in violation of the Commerce Clause and the Equal Protection Clause, because an Illinois corporation doing business in Iowa must pay tax on a greater portion of its income than a local Iowa company, and an Iowa company doing business in Illinois will pay tax on less of its income than an Illinois corporation doing business in Iowa. The simple answer, however, is that whatever disparity may have existed is not attributable to the Iowa statute. It treats both local and foreign concerns with an even hand; the alleged disparity can only be the consequence of the combined effect of the Iowa and Illinois statutes, and Iowa is not responsible for the latter.
Thus, appellant's "discrimination" claim is simply a way of describing the potential consequences of the use of different formulas by the two States. These consequences, however, could be avoided by the adoption of any uniform rule; the "discrimination" does not inhere in either State's formula.
The only conceivable constitutional basis for invalidating the Iowa statute would be that the Commerce Clause prohibits any overlap in the computation of taxable income by the States. If the Constitution were read to mandate such precision in interstate taxation, the consequences would extend far beyond this particular case. For some risk of duplicative taxation exists whenever the States in which a corporation does business do not follow identical rules for the division of income. Accepting appellant's view of the Constitution, therefore, would require extensive judicial lawmaking. Its logic is not limited to a prohibition on use of a single-factor apportionment formula. The asserted constitutional flaw in that formula is that it is different from that presently employed by a majority of States and that difference creates a risk of duplicative taxation. But a host of other division-of-income problems create precisely the same risk and would similarly rise to constitutional proportions.
Thus, it would be necessary for this Court to prescribe a uniform definition of each category in the three-factor formula. For if the States in which a corporation does business have different rules regarding where a "sale" takes place, and each includes the same sale in its three-factor computation of the corporation's income, there will be duplicative taxation despite the apparent identity of the formulas employed. A similar risk of multiple taxation is created by the diversity among the States in the attribution of "nonbusiness" income, generally defined as that portion of a taxpayer's income that does not arise from activities in the regular course of its business. Some States do not distinguish between business and nonbusiness income for apportionment purposes. Other States, however, have adopted special rules that attribute nonbusiness income to specific locations. Moreover, even among the latter, there is diversity in the definition of nonbusiness income and in the designation of the locations to which it is deemed attributable. The potential for attribution of the same income to more than one State is plain.
Thus, while some States such as Iowa assign sales by destination, "sales can be assigned to the state . . . of origin, the state in which the sales office is located, the state where an employee of the business making the sale carries on his activities or where the order is first accepted, or the state in which an interstate shipment is made." Note, State Taxation of Interstate Businesses and the Multistate Tax Compact: The Search for a Page 279 Delicate Uniformity, 11 Colum. J. Law Soc. Prob. 231, 237 n. 20 (1975) (citation omitted).
See, e.g., Uniform Division of Income for Tax Purposes Act § 1(a).
Thus, one State in which a corporation does business may consider a particular type of income business income and simply include it in its apportionment formula; a second State may deem that same income nonbusiness income and attribute it to itself as the "commercial domicile" of the company; and a third State, though also considering it nonbusiness income, may attribute it to itself as the "legal domicile" of the company. See Note, supra n. 13, at 239.
The prevention of duplicative taxation, therefore, would require national uniform rules for the division of income. Although the adoption of a uniform code would undeniably advance the policies that underlie the Commerce Clause, it would require a policy decision based on political and economic considerations that vary from State to State. The Constitution, however, is neutral with respect to the content of any uniform rule. If division-of-income problems were to be constitutionalized, therefore, they would have to be resolved in the manner suggested by appellant for resolution of formula diversity — the prevalent practice would be endorsed as the constitutional rule. This rule would at best be an amalgam of independent state decisions, based on considerations unique to each State. Of most importance, it could not reflect the national interest, because the interests of those States whose policies are subordinated in the quest for uniformity would be excluded from the calculation.
This process is especially unsettling if a longstanding tax policy of one State, such as Iowa's, becomes the object of constitutional attack simply because it is different from the recently adopted practice of its neighbor.
While the freedom of the States to formulate independent policy in this area may have to yield to an overriding national interest in uniformity, the content of any uniform rules to which they must subscribe should be determined only after due consideration is given to the interests of all affected States. It is clear that the legislative power granted to Congress by the Commerce Clause of the Constitution would amply justify the enactment of legislation requiring all States to adhere to uniform rules for the division of income. It is to that body, and not this Court, that the Constitution has committed such policy decisions.
Finally, it would be an exercise in formalism to declare appellant's income tax assessment unconstitutional based on speculative concerns with multiple taxation. For it is evident that appellant would have had no basis for complaint if, instead of an income tax, Iowa had imposed a more burdensome gross-receipts tax on the gross receipts from sales to Iowa customers. In Standard Pressed Steel Co. v. Washington Revenue Dept., 419 U.S. 560, the Court sustained a tax on the entire gross receipts from sales made by the taxpayer into Washington State. Because receipts from sales made to States other than Washington were not included in Standard Pressed Steel's taxable gross receipts, the Court concluded that the tax was "`apportioned exactly to the activities taxed.'" Id., at 564.
In this case appellant's actual income tax obligation was the rough equivalent of a 1% tax on the entire gross receipts from its Iowa sales. Thus, the actual burden on interstate commerce would have been the same had Iowa imposed a plainly valid gross-receipts tax instead of the challenged income tax. Of more significance, the gross-receipts tax sustained in Standard Pressed Steel and General Motors Corp. v. Washington, 377 U.S. 436, is inherently more burdensome than the Iowa income tax. It applies whether or not the interstate concern is profitable and its imposition may make the difference between profit and loss. In contrast, the income tax is only imposed on enterprises showing a profit and the tax obligation is not heavy unless the profits are high.
Accordingly, until Congress prescribes a different rule, Iowa is not constitutionally prohibited from requiring taxpayers to prove that application of the single-factor formula has produced arbitrary results in a particular case.
The judgment of the Iowa Supreme Court is affirmed.
I agree with the Court that, for purposes of constitutional review, there is no distinction between a corporate income tax and a gross-receipts tax. I do not agree, however, that Iowa's single-factor sales apportionment formula meets the Commerce Clause requirement that a State's taxation of interstate business must be "fairly apportioned to the commerce carried on within the taxing state." Western Live Stock v. Bureau of Revenue, 303 U.S. 250, 256 (1938). As I have previously explained:
"[Where a sale] exhibits significant contacts with more than one State . . . it is the commercial activity within the State, and not the sales volume, which determines the State's power to tax, and by which the tax must be apportioned. While the ratio of in-state to out-of-state sales is often taken into account as one factor among others in apportioning a firm's total net income, see, e.g., the description of the `Massachusetts Formula' in Note. 75 Harv. L. Rev. 953, 1011 (1962), it nevertheless remains true that if commercial activity in more than one State results in a sale in one of them, that State may not claim as all its own the gross receipts to which the activity within its borders has contributed only a part. Such a tax must be apportioned to reflect the business activity within the taxing State." General Motors Corp. v. Washington, 377 U.S. 436, 450-451 (1964) (dissenting opinion).
I would therefore reverse.
The unspoken, but obvious, premise of the majority opinion is the fear that a Commerce Clause invalidation of Iowa's single-factor sales formula will lead the Court into problems and difficulties in other cases yet to come. I reject that premise.
I agree generally with the content of MR. JUSTICE POWELL'S opinion in dissent. I join that opinion because I, too, feel that the Court has a duty to resolve, not to avoid, these problems of "delicate adjustment," Boston Stock Exchange v. State Tax Comm'n, 429 U.S. 318, 329 (1977), and because the opinion well demonstrates that Iowa's now anachronistic single-factor sales formula runs headlong into overriding Commerce Clause considerations and demands.
Today's decision is bound to be regressive. Single-factor formulas are relics of the early days of state income taxation. The three-factor formulas were inevitable improvements and, while not perfect, reflect more accurately the realities of the business and tax world. With their almost universal adoption by the States, the Iowa system's adverse and parochial impact on commerce comes vividly into focus. But with its single-factor formula now upheld by the Court, there is little reason why other States, perceiving or imagining a similar advantage to local interests, may not go back to the old ways. The end result, in any event, is to exacerbate what the Commerce Clause, absent governing congressional action, was devised to avoid.
Iowa is not a member of the Multistate Tax Commission. Tr. of Oral Arg. 33. See United States Steel Corp. v. Multistate Tax Comm'n, 434 U.S. 452 (1978).
Iowa's income tax was first adopted in 1934. 1933-1934 Iowa Acts, Ex. Sess., ch. 82; Tr. of Oral Arg. 29. Its single-factor sales formula was embraced in § 28 of that original Act.
It is the duty of this Court "to make the delicate adjustment between the national interest in free and open trade and the legitimate interest of the individual States in exercising their taxing powers." Boston Stock Exchange v. State Tax Comm'n, 429 U.S. 318, 329 (1977). This duty must be performed with careful attention to the settings of particular cases and consideration of their special facts. See Raymond Motor Transp., Inc. v. Rice, 434 U.S. 429, 447-448, n. 25 (1978). Consideration of all the circumstances of this case leads me to conclude that Iowa's use of a single-factor sales formula to apportion the net income of multistate corporations results in the imposition of "a tax which discriminates against interstate commerce . . . by providing a direct commercial advantage to local business." Northwestern States Portland Cement Co. v. Minnesota, 358 U.S. 450, 458 (1959). I therefore dissent.
Iowa's use of single-factor sales-apportionment formula — though facially neutral — operates as a tariff on goods manufactured in other States and as a subsidy to Iowa manufacturers selling their goods outside of Iowa. Because 44 of the 45 other States (including the District of Columbia) which impose corporate income taxes use a three-factor formula involving property, payroll, and sales, Iowa's practice insures that out-of-state businesses selling in Iowa will have higher total tax payments than local businesses. This result follows from the fact that Iowa attributes to itself all of the income derived from sales in Iowa, while other taxing States — using the three-factor formula — are also taxing some portion of the same income through attribution to property or payroll in those States.
Those 44 States are as follows: Alabama, Alaska, Arizona, Arkansas, California, Colorado, Connecticut, Delaware, District of Columbia, Florida, Georgia, Hawaii, Idaho, Illinois, Indiana, Kansas, Kentucky, Louisiana, Maine, Maryland, Massachusetts, Michigan, Minnesota, Mississippi, Missouri, Montana, Nebraska, New Hampshire, New Jersey, New Mexico, New York, North Carolina, North Dakota, Ohio, Oklahoma, Oregon, Pennsylvania, Rhode Island, South Carolina, Tennessee, Utah, Vermont, Virginia, and Wisconsin.
West Virginia, the 45th State, uses a two-factor formula which omits the sales component. Colorado also has a two-factor property and sales formula, and Missouri a one-factor sales formula, which are available to taxpayers at their option as alternatives to the three-factor formula.
This surcharge on Iowa sales increases to the extent that a business' plant and labor force are located outside Iowa. It can be avoided altogether only by locating all property and payroll in Iowa; an Iowa manufacturer selling only in Iowa will never have any portion of its income attributed to any other State. And to the extent that an Iowa manufacturer makes its sales in States other than Iowa, its overall state tax liability will be reduced. Assuming comparable tax rates, its liability to other States, in which sales constitute only one-third of the apportionment formula, will be far less than the amount it would have owed with a comparable volume of sales in Iowa, where sales are the exclusive mode of apportioning income. The effect of Iowa's formula, then, is to penalize out-of-state manufacturers for selling in Iowa and to subsidize Iowa manufacturers for selling in other States.
A simplified example demonstrates the economic effect of the Iowa formula on out-of-state corporations.
Iowa Corp. is domiciled in Iowa, and its total property and payroll are located there. Illinois Corp. is domiciled in Illinois, with all its property and payroll in that State. Both corporations have $1 million in Page 285 net income, and both make half their sales in Iowa and half in Illinois. A 5% corporate income tax is levied in both States.
If both States use a single-factor sales apportionment formula, both would go through the following calculation in determining the tax liability of both corporations:
Sales in States ________________ = 1/2; 1/2 X $1,000,000 X 0.05 = $25,000 Total Sales
The pattern of payments and receipts would be as follows:
Total Taxes Taxes Paid Taxes Paid Paid by each to Iowa to Illinois Corporation __________________________________________________________________________ Illinois Corp. $25,000 $25,000 $50,000 Iowa Corp. 25,000 25,000 50,000 ___________________________________________________________________________ TOTAL 50,000 50,000
If both Iowa and Illinois again levy the same 5% income tax but use the three-factor formula, which is:
Sales in Property in Payroll in State State State ___________ + ______________ + ______________ Total Sales Total Property Total Payroll _________________________________________________________,
then each corporation's payment to its state of domicile would be
0.5+1+1 _______ X $1,000,000 X 0.05 = $41,667, and 3
its payment to the state in which it is a foreign corporation would be
0.5+0+0 _______ X $1,000,000 X 0.05 = $8,333. 3
The pattern of tax payments and receipts would be as follows:
This appeal requires us to determine whether these economic effects of the Iowa apportionment formula violate either the Due Process Clause or the Commerce Clause. I now turn to those questions.
For the reasons given by the Court, ante, at 271-275, I agree that application of Iowa's formula does not violate the Due Process Clause. The decisions of this Court make it clear that arithmetical perfection is not to be expected from apportionment formulae. International Harvester Co. v. Evatt, 329 U.S. 416 (1947). It has been said that the "apportionment theory is a mongrel one, a cross between desire not to interfere with state taxation and desire at the same time not utterly to crush out interstate commerce." Northwest Airlines, Inc. v. Minnesota, 322 U.S. 292, 306 (1944) (Jackson, J., concurring). It owes its existence to the fact that with respect to a business earning income through a series of transactions beginning with manufacturing in one State and ending with a sale in another, a precise — or even wholly logical — determination of the State in which any specific portion of the income was earned is impossible. Underwood Typewriter Co. v. Chamberlain, 254 U.S. 113, 120-121 (1920).
Hence, the fact that a particular formula — like the one at issue here — may permit a State to tax some income actually "located" in another State is not in and of itself a basis for finding a due process violation. Were it otherwise, any formula deviating in the smallest detail from that used in other States would be invalid. Because there is no ideal means of "locating" any State's rightful share, such uniformity cannot be dictated by this Court. Hence, the decisions of this Court properly require the taxpayer claiming a due process violation to show that the apportionment is "out of all appropriate proportion to the business transacted." Hans Rees' Sons, Inc. v. North Carolina ex rel. Maxwell, 283 U.S. 123, 135 (1931). As appellant has failed to make any such showing, I agree with the Court that no due process violation has been made out here.
This does not mean, as the Court suggests, ante, at 277-280, that this Court is disabled from ever determining whether a particular apportionment formula imposes multiple burdens upon or discriminates against interstate commerce. See General Motors Corp. v. District of Columbia, 380 U.S. 553 (1965); Bass, Ratcliff Gretton, Ltd. v. State Tax Comm'n, 266 U.S. 271 (1924); Underwood Typewriter Co. v. Chamberlain, 254 U.S. 113 (1920). Regardless of which formula more accurately locates the State in which any particular segment of income is earned, it is a mathematical fact that the use of different formulae may result in taxation on more than 100% of the corporation's income under the State's own definitions, as well as in skewed tax effects. See n. 2, supra. When this result has a predictably burdensome or discriminatory effect, Commerce Clause scrutiny is triggered. See Part III, infra. The effects of the challenged formula upon the particular corporation's income is strictly related only to inquiry under the Due Process Clause, since Commerce Clause analysis focuses on the impact upon commerce in general.
This conclusion does not ipso facto mean that Commerce Clause strictures are satisfied as well. This Court's decisions dealing with state levies that discriminate against out-of-state business, as Iowa's formula does, compel a more detailed inquiry.
It is a basic principle of Commerce Clause jurisprudence that "[n]either the power to tax nor the police power may be used by the state of destination with the aim and effect of establishing an economic barrier against competition with the products of another state or the labor of the residents." Baldwin v. G. A. F. Seelig, Inc., 294 U.S. 511, 527 (1935); accord, H. P. Hood Sons v. Du Mond, 336 U.S. 525, 532 (1949); Boston Stock Exchange, 429 U.S., at 335-336, and n. 14. Those barriers would constitute "an unreasonable clog upon the mobility of commerce." Baldwin, supra, at 527.
One form of such unreasonable restrictions is "discriminating State legislation." Welton v. Missouri, 91 U.S. 275, 280 (1876). This Court consistently has struck down state and local taxes which unjustifiably benefit local businesses at the expense of out-of-state businesses. Ibid.; accord, Boston Stock Exchange; Halliburton Oil Well Co. v. Reily, 373 U.S. 64 (1963); Nippert v. Richmond, 327 U.S. 416 (1946); Hale v. Bimco Trading, Inc., 306 U.S. 375 (1939); I. M. Darnell Son v. Memphis, 208 U.S. 113 (1908); Guy v. Baltimore, 100 U.S. 434 (1880).
This ban applies not only to state levies that by their terms are limited to products of out-of-state business, or which explicitly tax out-of-state sellers at higher rates than local sellers. It also reaches those taxes that "in their practical operation [work] discriminatorily against interstate commerce to impose upon it a burden, either in fact or by the very threat of its incidence." Nippert v. Richmond, supra, at 425. For example, this Court has invalidated a facially neutral fixed-fee license tax collected from all local and out-of-state "drummers," where it appeared the tax fell far more heavily upon out-of-state businesses, since local businesses had little or no occasion to solicit sales in that manner. Robbins v. Shelby County Taxing Dist., 120 U.S. 489 (1887). See also West Point Wholesale Grocery Co. v. Opelika, 354 U.S. 390 (1957); Memphis Steam Laundry Cleaner, Inc. v. Stone, 342 U.S. 389 (1952); Best Co. v. Maxwell, 311 U.S. 454 (1940); Real Silk Hosiery Mills v. Portland, 268 U.S. 325 (1925); Corson v. Maryland, 120 U.S. 502 (1887). Thus, the constitutional inquiry relates not simply to the form of the particular tax, but to its effect on competition in the several States.
As indicated in Part I above, application of Iowa's single-factor sales-apportionment formula, in the context of general use of three-factor formulae, inevitably handicaps out-of-state businesses competing for sales in Iowa. The handicap will diminish to the extent that the corporation locates its plant and labor force in Iowa, but some competitive disadvantage will remain unless all of the corporate property and payroll are relocated in Iowa. In the absence of congressional action, the Commerce Clause constrains us to view the State's interest in retaining this particular levy as against the constitutional preference for an open economy. See, e.g., Raymond Motor Transp., Inc. v. Rice, 434 U.S., at 440-442; Pike v. Bruce Church, Inc., 397 U.S. 137, 142 (1970); DiSanto v. Pennsylvania, 273 U.S. 34, 44 (1927) (Stone, J., dissenting); Dowling, Interstate Commerce and State Power, 27 Va. L. Rev. 1, 14-15, and n. 20 (1940).
The clog on commerce present here is similar to the risk of imposing "multiple burdens" on interstate commerce against which the Court has warned in various decisions. See, e.g., Western Live Stock v. Bureau of Revenue, 303 U.S. 250, 255-256 (1938); J. D. Adams Mfg. Co. v. Storen, 304 U.S. 307, 311-312 (1938); Gwin, White Prince, Inc. v. Henneford, 305 U.S. 434, 439 (1939); Northwestern States Portland Cement Co. v. Minnesota, 358 U.S. 450, 458 (1959). Compare Evco v. Jones, 409 U.S. 91 (1972), with General Motors Corp. v. Washington, 377 U.S. 436 (1964). In this case, Iowa corporations will not risk additional burdens when they make out-of-state sales. Cf. Hunt v. Washington Apple Advertising Comm'n, 432 U.S. 333, 351 (1977). Indeed, to the extent that they shift sales out of Iowa, their overall state tax liability will decrease. Out-of-state corporations selling in Iowa, however, do face the prospect of multiple burdens. Hence, there is clear discrimination against out-of-state corporations, which is the consequence of the particular multiple burden imposed.
Iowa's interest in any particular level of tax revenues is not affected by the use of the single-factor sales formula. It cannot be predicted with certainty that its application will result in higher revenues than any other formula. If Iowa needs more revenue, it can adjust its tax rates. That adjustment would not have the discriminatory impact necessarily flowing from the choice of the single-factor sales formula. Hence, if Iowa's choice is to be sustained, it cannot be by virtue of the State's interest in protecting its fisc or its power to tax. No other justification is offered. If we are to uphold Iowa's apportionment formula, it must be because no consistent principle can be developed that could account for the invalidation of the Iowa formula, yet support application of other States' imprecise formulae.
For example, if Iowa switched to a three-factor formula and retained the same rates, revenues from out-of-state corporations would decrease, since Iowa would no longer be attributing to itself all of the income earned by Iowa sales of such corporations. Revenues from corporations located in Iowa, however, would increase, since Iowa would now be attributing to itself some portion of the income earned by those corporations' out-of-state sales. See also n. 2, supra.
Given the nearly infinite variety of taxes, rates, and apportionment formulae, it might be possible for Iowa to alter its entire tax structure to effect a similar discrimination, and perhaps to do it in a way that avoids Commerce Clause scrutiny. See Barrett, "Substance" vs. "Form" in the Application of the Commerce Clause to State Taxation, 101 U. Pa. L. Rev. 740, 748 (1953). That speculative possibility cannot deter us from striking down an obvious discrimination against interstate commerce when one is presented. The Court has never shrunk from that duty in the past. To do so would be to abandon any effort of applying Commerce Clause principles to state tax measures.
This is not to say that States are always forbidden to offer tax incentives to encourage local industry or to achieve other valid state goals. See, e.g., Hughes v. Alexandria Scrap Corp., 426 U.S. 794 (1976). Such programs, and the interests being served, must be considered on a case-by-case basis.
It is argued that since this Court on several occasions has upheld the use of single-factor formulae, Iowa's scheme cannot be regarded as suspect simply because it does not embody the prevalent three-factor theory. Consideration of the decisions dealing with single-factor formulae, however, reveals that each is distinguishable.
In Underwood Typewriter Co. v. Chamberlain, 254 U.S. 113 (1920), this Court upheld Connecticut's use of a single-factor property formula to apportion the net profits of a foreign corporation. Such a formula is not clearly discriminatory in Commerce Clause terms. The only competitive disadvantage inevitably resulting from it would attend a decision to locate a plant or office in the taxing State. The Commerce Clause does not concern itself with a State's decision to place local business at a disadvantage. Cf. Allied Stores of Ohio, Inc. v. Bowers, 358 U.S. 522, 528 (1959).
Bass, Ratcliff Gretton, Ltd. v. State Tax Comm'n, 266 U.S. 271 (1924), is similarly distinguishable. In Bass, New York apportioned the net income of foreign corporations using a single-factor property formula that comprised real and tangible personal property, bills and accounts receivable, and stock in other corporations. This Court upheld that formula, observing that plaintiff in error had not shown that "application of the statutory method of apportionment has produced an unreasonable result." Id., at 283. As in Underwood Typewriter, however, the single-factor property formula did not necessarily discriminate against businesses carried on out of State; indeed, its impact would tend to increase to the extent that corporate business was carried on within the State. Cf. National Leather Co. v. Massachusetts, 277 U.S. 413 (1928); accord, e.g., International Shoe Co. v. Shartel, 279 U.S. 429 (1929); New York v. Latrobe, 279 U.S. 421 (1929); Hump Hairpin Co. v. Emmerson, 258 U.S. 290 (1922); United States Glue Co. v. Oak Creek, 247 U.S. 321 (1918).
Somewhat more troublesome is Ford Motor Co. v. Beauchamp, 308 U.S. 331 (1939). In that case, the Court sustained Texas' use of a single-factor sales formula to apportion the outstanding capital stock, surplus, undivided profits, and long-term obligations of corporations subject to the state franchise tax. While this case may be seen as standing for the proposition that single-factor sales formulae are not per se illegal, it is not controlling in the present case. In Ford Motor Co., as in Underwood Typewriter and Bass, there was no showing of virtually universal use of a conflicting type of formula for determining the same tax. Thus, it could not be said that the Texas formula inevitably imposed a competitive disadvantage on out-of-state corporations. Discrimination not being shown, there was no basis for invalidating the Texas scheme under the Commerce Clause.
Although overruling Ford Motor Co. would not be necessary in this case, the time may be ripe for its reconsideration. See, e.g., J. Hellerstein, State and Local Taxation 324 (3d ed. 1969). As suggested in General Motors Corp. v. District of Columbia, 380 U.S. 553, 561 (1965), a sales-only formula is probably the most illogical of all apportionment methods, since "the geographic distribution of a corporation's sales is, by itself, of dubious significance in indicating the locus of either" a corporation's sources of income or the social costs it generates.
The Court's willingness to uphold the sales-only formula in Ford Motor Co. may have been the result of its view that it was dealing solely with the "measure" of the tax rather than its "subject." See 308 U.S., at 336. This Court no longer adheres to the use of those formalistic labels, looking instead to "economic realities" in determining the constitutionality of state taxing schemes. Complete Auto Transit, Inc. v. Brady, 430 U.S. 274, 279 (1977).
The opposite is true here. In the context of virtually universal use of the basic three-factor formula, Iowa's use of the single-factor sales formula necessarily discriminates against out-of-state manufacturers. The only remaining question, then, is whether Iowa's scheme may be saved by the fact that its discriminatory nature depends on context: If other States were not virtually unanimous in their use of an opposing formula, past decisions would make it difficult to single out Iowa's scheme as more offensive than any other.
On several occasions, this Court has compared a state statutory requirement against the practice in other States in determining the statute's validity under the Commerce Clause. In Southern Pacific Co. v. Arizona ex rel. Sullivan, 325 U.S. 761 (1945), the Court struck down a state statute limiting passenger trains to 14 cars and freight trains to 70 cars. Noting that only one State other than Arizona enforced a restriction on train lengths, the Southern Pacific Court specifically considered the Arizona law against the background of the activities in other States:
That State was Oklahoma. Southern Pacific Co. v. Arizona ex rel. Sullivan, 325 U.S., at 773-774, n. 3.
"Enforcement of the law in Arizona, while train lengths remain unregulated or are regulated by varying standards in other states, must inevitably result in an impairment of uniformity of efficient railroad operation because the railroads are subjected to regulation which is not uniform in its application. Compliance with a state statute limiting train lengths requires interstate trains of a length lawful in other states to be broken up and reconstituted as they enter each state according as it may impose varying limitations upon train lengths. The alternative is for the carrier to conform to the lowest train limit restriction of any of the states through which its trains pass, whose laws thus control the carriers' operations both within and without the regulating state." Id., at 773. (Emphasis added.)
The clear implication is that the Court's view of the Arizona length limit might have been different if practices in other States had been other than as the Court found them. Had other States adopted the Arizona rule, there might have been no basis for holding it unconstitutional. See also Morgan v. Virginia, 328 U.S. 373 (1946); Hall v. DeCuir, 95 U.S. 485 (1878).
The Court also looked to the practices of other States in holding unconstitutional Illinois' mudguard requirement in Bibb v. Navajo Freight Lines, Inc., 359 U.S. 520 (1959). The type of mudguard banned on trucks operating in Illinois was required in Arkansas and permitted in 45 other States. The Court pointed out the conflict between the Illinois and Arkansas regulations and went on to consider the relevance of other States' rules:
"A State which insists on a design out of line with the requirements of almost all the other States may sometimes place a great burden of delay and inconvenience on those interstate motor carriers entering or crossing its territory. Such a new safety device — out of line with the requirements of the other States — may be so compelling that the innovating State need not be the one to give way. But the present showing — balanced against the clear burden on commerce — is far too inconclusive to make this mudguard meet that test." Id., at 529-530.
It seems clear from the Bibb Court's discussion that the conflict between the Illinois regulation and that of Arkansas would not have led to the latter's invalidation had it been the one before the Court. The Arkansas regulation merely required what was permitted in nearly all the other States. After looking to that virtually uniform practice opposed to that of Illinois, the conclusion that the Illinois requirement was "out of line" was a relatively simple one. Since it was not justified by any interest in increased safety, it was held unconstitutional. See also Raymond Motor Transp., Inc. v. Rice, 434 U.S., at 444-446.
Most nearly in point is General Motors Corp. v. District of Columbia, 380 U.S. 553 (1965). In that case, this Court held unlawful the District's use of a single-factor sales apportionment formula under the District of Columbia Income and Franchise Tax Act of 1947. Although the decision turned on a question of statutory interpretation, the Court's analysis is equally applicable to a Commerce Clause inquiry:
"The great majority of States imposing corporate income taxes apportion the total income of a corporation by application of a three-factor formula which gives equal weight to the geographical distribution of plant, payroll, and sales. The use of an apportionment formula based wholly on the sales factor, in the context of general use of the three-factor approach, will ordinarily result in multiple taxation of corporate net income . . . . In any case, the sheer inconsistency of the District formula with that generally prevailing may tend to result in the unhealthy fragmentation of enterprise and an uneconomic pattern of plant location, and so presents an added reason why this Court must give proper meaning to the relevant provisions of the District Code." Id., at 559-560 (footnote omitted).
The General Motors Court, then, expressly evaluated the single-factor sales formula in the context of general use of the three-factor method and concluded that the former created dangers for interstate commerce.
These cases lead me to believe that it is not only proper but essential to determine the validity of the Iowa formula against the background of practices in the other States. If one State's regulatory or taxing statute is significantly "out of line" with other States' rules, Bibb, supra, at 530, and if by virtue of that departure from the general practice it burdens or discriminates against interstate commerce, Commerce Clause scrutiny is triggered, and this Court must invalidate it unless it is justified by a legitimate local purpose outweighing the harm to interstate commerce, Pike v. Bruce Church, Inc., 397 U.S., at 142; accord, Hughes v. Alexandria Scrap Corp., 426 U.S. 794, 804 (1976). There probably can be no fixed rule as to how nearly uniform the countervailing state policies must be; that is, there can be no rule of 26 States, of 35, or of 45. Commerce Clause inquiries generally do not run in such precise channels. The degree of conflict and its resulting impact on commerce must be weighed in the circumstances of each case. But the difficulty of engaging in that weighing process does not permit this Court to avoid its constitutional duty and allow an individual State to erect "an unreasonable clog upon the mobility of commerce," Baldwin v. G. A. F. Seelig, Inc., 294 U.S., at 527, by taking advantage of the other States' commendable trend toward uniformity.
Such is the case before us. Forty-four of the forty-five States (including the District of Columbia), other than Iowa, that impose a corporate income tax utilize a similar three-factor apportionment formula. The 45th State, West Virginia, uses a two-factor formula based on property and payroll. See n. 1, supra. Those formulae individually may be no more rational as means of apportioning the income of a multistate business than Iowa's single-factor sales formula. But see General Motors Corp. v. District of Columbia, supra, at 561. Past decisions upheld differing formulae because of this inability to determine that any of the various methods of apportionment in use was the best; so long as a State's choice was not shown to be grossly unfair, it would be upheld. Compare Underwood Typewriter with Hans Rees' Sons. The more recent trend toward uniformity, however, permits identification of Iowa's formula, like the mudguard requirement in Bibb, as "out of line," if not per se irrational. Since Iowa's formula inevitably discriminates against out-of-state sellers, and since it has not been justified on any fiscal or administrative basis, I would hold it invalid under the Commerce Clause.
There are differences in definitions of the three factors among the States that use a three-factor formula. See, e.g., J. Hellerstein, State and Local Taxation 309-310, and n. 7 (3d ed. 1969); Note, State Taxation of Interstate Businesses and the Multistate Tax Compact: The Search for a Delicate Uniformity, 11 Colum. J. of Law Soc. Prob. 231, 235-238 (1975). Such differences may tend in less dramatic fashion to impose burdens on out-of-state businesses not entirely dissimilar to the one presented here. It may be that any such effects do not work inevitably in one direction, as does the burden imposed here, or they may be de minimis in Commerce Clause terms. In any event, they are not presently before us. It suffices to dispose of this case that nearly all the other States use a basic three-factor formula, while Iowa clings to its sales-only method.