U.S. Supreme Court


Decided April 3, 1995

JUSTICE SOUTER delivered the opinion of the Court.

This case raises the question whether Oklahoma's sales tax on the full price of a ticket for bus travel from Oklahoma to another State is consistent with the Commerce Clause. We hold that it is.

Oklahoma taxes sales in the State of certain goods and services, including transportation for hire. The buyers of the taxable goods and services pay the taxes, which must be collected and remitted to the State by sellers.

Respondent Jefferson Lines, Inc., is a Minnesota corporation that provided bus services as a common carrier in Oklahoma from 1988 to 1990. Jefferson did not collect or remit the sales taxes for tickets it had sold in Oklahoma for bus travel from Oklahoma to other States, although it did collect and remit the taxes for all tickets it had sold in Oklahoma for travel that originated and terminated within that State.

After Jefferson filed for bankruptcy protection on October 27, 1989, petitioner, Oklahoma Tax Commission, filed proof of claims in Bankruptcy Court for the uncollected taxes for tickets for interstate travel sold by Jefferson.  Jefferson cited the Commerce Clause in objecting to the claims, and argued that the tax imposes an undue burden on interstate commerce by permitting Oklahoma to collect a percentage of the full purchase price of all tickets for interstate bus travel, even though some of that value derives from bus travel throu other States. The tax also presents the danger of multiple taxation, Jefferson claimed, because any other State through which a bus travels while providing the services sold in Oklahoma will be able to impose taxes of their own upon Jefferson or its passengers for use of the roads.

The Bankruptcy Court agreed with Jefferson, the District Court affirmed, and so did the United States Court of Appeals for the Eighth Circuit. We granted certiorari, and now reverse....

The taxpayer does not deny Oklahoma's substantial nexus to the in-state portion of the bus service, but rather argues that nexus to the State is insufficient as  to the portion of travel outside its borders.

The difficult question in this case is whether the tax is properly apportioned within the meaning of the second prong of Complete Auto's test, "the central purpose [of which] is to ensure that each State taxes only its fair share of an interstate transaction." This principle of fair share is the lineal descendant of [the] prohibition of multiple taxation, which is threatened whenever one State's act of overreaching combines with the possibility that another State will claim its fair share of the value taxed: the portion of value by which one State exceeded its fair share would be taxed again by a State properly laying claim to it.

For over a decade now, we have assessed any threat of malapportionment by asking whether the tax is "internally consistent" and, if so, whether it is "externally consistent" as well.  Internal consistency is preserved when the imposition of a tax identical to the one in question by every other State would add no burden to interstate commerce that intrastate commerce would not also bear. This test asks nothing about the degree of economic reality reflected by the tax, but simply looks to the structure of the tax at issue to see whether its identical application by every State in the Union would place interstate commerce at a disadvantage as compared with commerce intrastate. A failure of internal consistency shows as a matter of law that a State is attempting to take more than its fair share of taxes from the interstate transaction, since allowing such a tax in one State would place interstate commerce at the mercy of those remaining States that might impose an identical tax. There is no failure of it in this case, however. If every State were to impose a tax identical to Oklahoma's, that is, a tax on ticket sales within the State for travel originating there, no sale would be subject to more than one State's tax.

External consistency, on the other hand, looks not to the logical consequences of cloning, but to the economic justification for the State's claim upon the value taxed, to discover whether a State's tax reaches beyond that portion of value that is fairly attributable to economic activity within the taxing State. Here, the threat of real multiple taxation (though not by literally identical statutes) may indicate a State's impermissible overreaching. It is to this less tidy world of real taxation that we turn now, and at length.

The very term "apportionment" tends to conjure up allocation by percentages, and where taxation of income from interstate business is in issue, apportionment disputes have often centered around specific formulas for slicing a taxable pie among several States in which the taxpayer's activities contributed to taxable value..... In Central Greyhound, we held that New York's taxation of an interstate busline's gross receipts was constitutionally limited to that portion reflecting miles traveled within the taxing jurisdiction.

In reviewing sales taxes for fair share, however, we have had to set a different course. A sale of goods is most readily viewed as a discrete event facilitated by the laws and amenities of the place of sale, and the transaction itself does not readily reveal the extent to which completed or anticipated interstate activity affects the value on which a buyer is taxed. We have therefore consistently approved taxation of sales without any division of the tax base among different States, and have instead held such taxes properly measurable by the gross charge for the purchase, regardless of any activity outside the taxing jurisdiction that might have preceded the sale or might occur in the future.

Such has been the rule even when the parties to a sales contract specifically contemplated interstate movement of the goods either immediately before, or after, the transfer of ownership. As we put it, a necessary condition for imposing the tax was the occurrence of "a local activity, delivery of goods within the State upon their purchase for consumption."  In other words, the very conception of the common sales tax on goods, operating on the transfer of ownership and possession at a particular time and place, insulated the buyer from any threat of further taxation of the transaction.

In deriving this rule covering taxation to a buyer on sales of goods we were not, of course, oblivious to the possibility of successive taxation of related events up and down the stream of commerce, and our cases are implicit with the understanding that the Commerce Clause does not forbid the actual assessment of a succession of taxes by different States on distinct events as the same tangible object flows along. Thus, it is a truism that a sales tax to the buyer does not preclude a tax to the seller upon the income earned from a sale, and there is no constitutional trouble inherent in the imposition of a sales tax in the State of delivery to the customer, even though the State of origin of the thing sold may have assessed a property or severance tax on it. In light of this settled treatment of taxes on sales of goods and other successive taxes related through the stream of commerce, it is fair to say that because the taxable event of the consummated sale of goods has been found to be properly treated as unique, an internally consistent, conventional sales tax has long been held to be externally consistent as well.

A sale of services can ordinarily be treated as a local state event just as readily as a sale of tangible goods can be located solely within the State of delivery.  Although our decisional law on sales of services is less developed than on sales of goods, one category of cases dealing with taxation of gross sales receipts in the hands of a seller of services supports the view that the taxable event is wholly local. Thus we have held that the entire gross receipts derived from sales of services to be performed wholly in one State are taxable by that State, notwithstanding that the contract for performance of the services has been entered into across state lines with customers who reside outside the taxing State. So, too, as we have already noted, even where interstate circulation contributes to the value of magazine advertising purchased by the customer, we have held that the Commerce Clause does not preclude a tax on its full value by the State of publication....

Cases on gross receipts from sales of services include one falling into quite a different category, however, and it is on this decision that the taxpayer relies for an analogy said to control the resolution of the case before us. In 1948, the Court decided Central Greyhound Lines, Inc. v. Mealey, striking down New York's gross receipts tax on transportation services imposed without further apportionment on the total receipts from New York sales of bus services, almost half of which were actually provided by carriage through neighboring New Jersey and Pennsylvania. The Court held the statute fatally flawed by the failure to apportion taxable receipts in the same proportions that miles traveled through the various States bore to the total. The similarity of Central Greyhound to this case is, of course, striking, and on the assumption that the economic significance of a gross receipts tax is indistinguishable from a tax on sales the Court of Appeals held that a similar mileage apportionment is required here, as the taxpayer now argues.

We, however, think that Central Greyhound provides the wrong analogy for answering the sales tax apportionment question here. To be sure, the two cases involve the identical services, and apportionment by mileage per State is equally feasible in each. But the two diverge crucially in the identity of the taxpayers and the consequent opportunities that are understood to exist for multiple taxation of the same taxpayer. Central Greyhound did not rest simply on the mathematical and administrative feasibility of a mileage apportionment, but on the Court's express understanding that the seller-taxpayer was exposed to taxation by New Jersey and Pennsylvania on portions of the same receipts that New York was taxing in their entirety. The Court thus understood the gross receipts tax to be simply a variety of tax on income, which was required to be apportioned to reflect the location of the various interstate activities by which it was earned.

Here, in contrast, the tax falls on the buyer of the services, who is no more subject to double taxation on the sale of these services than the buyer of goods would be. The taxable event comprises agreement, payment, and delivery of some of the services in the taxing State; no other State can claim to be the site of the same combination. The economic activity represented by the receipt of the ticket for "consumption" in the form of commencement and partial provision of the transportation thus closely resembles Berwind-White's "delivery of goods within the State upon their purchase for consumption," especially given that full "consumption" or "use" of the purchased goods within the taxing State has never been a condition for taxing a sale of those goods.....

In sum, the sales taxation here is not open to the double taxation analysis on which Central Greyhound turned, and that decision does not control. Before we classify the Oklahoma tax with standard taxes on sales of goods, and with the taxes on less complicated sales of services, however, two questions may helpfully be considered.

Although the sale with partial delivery cannot be duplicated as a taxable event in any other State, and multiple taxation under an identical tax is thus precluded, is there a possibility of successive taxation so closely related to the transaction as to indicate potential unfairness of Oklahoma's tax on the full amount of sale? And if the answer to that question is no, is the very possibility of apportioning by mileage a sufficient reason to conclude that the tax exceeds the fair share of the State of sale?

The taxpayer argues that anything but a Central Greyhound mileage apportionment by State will expose it to the same threat of multiple taxation assumed to exist in that case. But the claim does not hold up. The taxpayer has failed to raise any spectre of successive taxes that might require us to reconsider whether an internally consistent tax on sales of services could fail the external consistency test for lack of further apportionment (a result that no sales tax has ever suffered under our cases).

If, for example, in the face of Oklahoma's sales tax, Texas were to levy a sustainable, apportioned gross receipts tax on the Texas portion of travel from Oklahoma City to Dallas, interstate travel would not be exposed to multiple taxation in any sense different from coal for which the producer may be taxed first at point of severance by Montana and the customer may later be taxed upon its purchase in New York. The multiple taxation placed upon interstate commerce by such a confluence of taxes is not a structural evil that flows from either tax individually, but it is rather the "accidental incident of interstate commerce being subject to two different taxing jurisdictions."

Nor has the taxpayer made out a case that Oklahoma's sales tax exposes any buyer of a ticket in Oklahoma for travel into another State to multiple taxation from taxes imposed upon passengers by other States of passage.... True, it is not Oklahoma that has offered to provide a credit for related taxes paid elsewhere, but in taxing sales Oklahoma may rely upon use-taxing States to do so. This is merely a practical consequence of the structure of use taxes as generally based upon the primacy of taxes on sales, in that use of goods is taxed only to the extent that their prior sale has escaped taxation.

Finally, Jefferson points to the fact that in this case, unlike the telephone communication tax at issue in Goldberg, Oklahoma could feasibly apportion its sales tax on the basis of mileage as we required New York's gross receipts tax to do in Central Greyhound. Although Goldberg indeed noted that "[a]n apportionment formula based on mileage or some other geographic division of individual telephone calls would produce insurmountable administrative and technological barriers," and although we agree that no comparable barriers exist here, we nonetheless reject the idea that a particular apportionment formula must be used simply because it would be possible to use it....

Oklahoma's tax on the sale of transportation services does not contravene the Commerce Clause.

JUSTICE SCALIA, with whom JUSTICE THOMAS joins, concurring in the judgment.

I agree with the Court's conclusion that Oklahoma's sales tax does not facially discriminate against interstate commerce. That seems to me the most we can demand to certify compliance with the "negative Commerce Clause" - which is "negative" not only because it negates state regulation of commerce, but also because it does not appear in the Constitution.

I would not apply the remainder of the eminently unhelpful, so-called "four-part test" of Complete Auto Transit, Inc. v. Brady (1977). Under the real Commerce Clause, it is for Congress to make the judgment that interstate commerce must be immunized from certain sorts of nondiscriminatory state action - a judgment that may embrace such imponderables as how much "value [is] fairly attributable to economic activity within the taxing State," and what constitutes "fair relation between a tax and the benefits conferred upon the taxpayer by the State."  I look forward to the day when Complete Auto will take its rightful place among the other useless and discarded tools of our negative-Commerce-Clause jurisprudence.

JUSTICE BREYER, with whom JUSTICE O'CONNOR joins, dissenting.

Despite the Court's lucid and thorough discussion of the relevant law, I am unable to join its conclusion for one simple reason. Like the judges of the Court of Appeals, I believe the tax at issue here and the tax that this Court held unconstitutional in Central Greyhound Lines, Inc. v. Mealey (1948), are, for all relevant purposes, identical. Both cases involve taxes imposed upon interstate bus transportation. In neither case did the State apportion the tax to avoid taxing that portion of the interstate activity performed in other States. And, I find no other distinguishing features. Hence, I would hold that the tax before us violates the Constitution for the reasons this Court set forth in Central Greyhound.

Central Greyhound considered a tax imposed by the State of New York on utilities doing business in New York - a tax called "`[e]mergency tax on the furnishing of utility services.'" That tax was equal to "two per centum" of "gross income," defined to include "receipts received . . . by reason of any sale . . . made" in New York. The New York taxing authorities had applied the tax to gross receipts  from sales (in New York) of bus transportation between New York City and cities in upstate New York over routes that cut across New Jersey and Pennsylvania. The out-of-state portion of the trips accounted for just over 40 percent of total mileage.

Justice Frankfurter wrote for the Central Greyhound Court that "it is interstate commerce which the State is seeking to reach," that the "real question [is] whether what the State is exacting is a constitutionally fair demand . . . for that aspect of the interstate commerce to which the State bears a special relation," and that by "its very nature an unapportioned gross receipts tax makes interstate transportation bear more than `a fair share of the cost of the local government whose protection it enjoys.'" The Court essentially held that the tax lacked what it would later describe as "external consistency." That is to say, the New York law violated the Commerce Clause because it tried to tax significantly more than "that portion of the revenues from the  interstate activity which reasonably reflects the in-state component of the activity being taxed."

The tax before us bears an uncanny resemblance to the New York tax. The Oklahoma statute (as applied to "[t]ransportation . . . by common carriers") imposes an "excise tax" of 4% "the gross receipts or gross proceeds of each sale" made in Oklahoma. The New York statute imposed a 2% tax on the "receipts received . . . by reason of any sale . . . made" in New York. Oklahoma imposes its tax on the total value of trips of which a large portion may take place in other States. New York imposed its tax on the total value of trips of which a large portion took place in other States. New York made no effort to apportion the tax to reflect the comparative cost or value of the in-state and out-of-state portions of the trips. Neither does Oklahoma. Where, then, can one find a critical difference?...

To suggest that the tax here is constitutional simply because it lends itself to recharacterizing the taxable event as a "sale" is to ignore economic reality. Because the sales tax is framed as a percentage of the ticket price, it seems clear that the activity Oklahoma intends to tax is the transportation of passengers - not some other kind of conduct (like selling tickets).

In any event, the majority itself does not seem to believe that Oklahoma is taxing something other than bus transportation; it seems to acknowledge the risk of multiple taxation. The Court creates an ingenious set of constitutionally-based taxing rules designed to show that any other State that imposes, say, a gross receipts tax on its share of bus ticket sales would likely have to grant a credit for the Oklahoma sales tax (unless it forced its own citizens to pay both a sales tax and a gross receipts tax). But, one might have said the same in Central Greyhound.... The difficulties with this approach lie in its complexity and our own inability to foresee all the ways in which other States might effectively tax their own portion of the journey now (also) taxed by Oklahoma. Under the Court's footnote rules, is not a traveler who buys a ticket in Oklahoma still threatened with a duplicative tax by a State that does not impose a sales tax on transportation (and thus, would not have to offer a credit for the sales tax paid in Oklahoma)? Even if that were not so, the constitutional problem would remain, namely that Oklahoma is imposing an unapportioned tax on the portion of travel outside the State, just as did New York.

Ultimately, I may differ with the majority simply because I assess differently the comparative force of two competing analogies. The majority finds determinative this Court's case law concerning sales taxes applied to the sale of goods, which cases, for example, permit one  State to impose a severance tax and another a sales tax on the same physical item (say, coal). In my view, however, the analogy to sales taxes is not as strong as the analogy to the tax at issue in Central Greyhound. After all, the tax before us is not a tax imposed upon a product that was made in a different State or was consumed in a different State or is made up of ingredients that come from a different State or has itself moved in interstate commerce. Rather, it is a tax imposed upon interstate travel itself - the very essence of interstate commerce. And, it is a fairly obvious effort to tax more than "that portion" of the "interstate activity['s]" revenue "which reasonably reflects the in-state component."  I would reaffirm the Central Greyhound principle, even if doing so requires different treatment for the inherently interstate service of interstate transportation, and denies the possibility of having a single, formal constitutional rule for all self-described "sales taxes." The Court of Appeals wrote that this "is a classic instance of an unapportioned tax" upon interstate commerce.  In my view, that is right. I respectfully dissent.

Exploring Constitutional Conflicts