Understanding uniformity and diversity in state corporate income taxes.

INTRODUCTION

Uniformity in taxation can yield substantial benefits. It can hold down costs of compliance and administration (1) and avoid gaps and overlaps in taxation, and, therefore, inequities and tax-induced distortions of economic decisions that are unrelated to differences in tax rates. Of course, uniformity can also entail costs, especially if it extends to tax rates. Harmonization of rates in effect creates a cartel and eliminates the beneficial effects of tax competition. (2) By comparison, if tax rates converge to levels required to reflect benefits of public services economic efficiency (but perhaps not equity as some would define it) would be enhanced.

Uniformity can have adverse effects, however, even if rates are not harmonized. In particular, taxes may converge or be harmonized in ways that make no sense. In the extreme case, the resulting system has perverse and undesirable effects. Sales-only apportionment and the nexus standard of P.L. 86-272, both discussed below, are good examples of this. More benignly, harmonization may produce a system that generally makes sense but is needlessly complex. The model retail sales tax produced by the Streamlined Sales Tax Project (SSTP), typifies this. (3) Finally, harmonization may fail to be sufficiently comprehensive, even if it is otherwise sensible. As explained below, the Uniform Division of Income for Tax Purposes Act (UDITPA) is clearly not sufficiently comprehensive.

This paper describes generic forces creating uniformity and diversity in the determination of state corporate tax bases. Then it examines several episodes in the history of state corporate income taxes, in order to ascertain how uniformity--or the lack thereof--came about. Looking forward, it briefly discusses whether UDITPA is likely to be revised to make it more rational and comprehensive. At no point does the paper discuss tax rates.

There is no need to start this survey a century ago, when the National Tax Association (NTA) was being formed, as harmonizing state corporate income taxes could not really exist as a practical problem until well after 1907. After all, Wisconsin enacted the nation's first modern state income tax only in 1911 and had been joined by only a handful of other states by 1920. (4) But it should be noted that even before the United States entered WWI, "It was foreseen that more States would turn to the income tax, and that the problems arising from the resulting diversity could be avoided only by adoption by the states of systems of taxation which were reasonably consistent with one another," and in 1916 the NTA created a committee to prepare a model system for taxing corporate income (U.S. House of Representatives, 1964, pp. 128-29).

FORCES CREATING UNIFORMITY AND DIVERSITY

State taxes can exhibit uniformity either because they converge, as if led by an invisible political hand, or because conscious efforts have been taken to harmonize them. Even if once exhibiting relative uniformity, taxes may diverge into diversity, often being led by quite visible political hands.

Forces Creating Uniformity

It is useful to distinguish three types of forces that may cause taxes levied by state governments to become more nearly uniform. In order to put skin--if not flesh--on the bones of these abstract forces, I will mention without elaboration several examples, some of which are explored further in what follows.

Convergence

When unilateral and uncoordinated actions of state governments produce uniformity, convergence, rather than harmonization, is at work. States may unilaterally conform their taxes to an objective external standard such as the Internal Revenue Code. Although the resulting uniformity is likely to be desirable, this is not guaranteed. Alternatively, states acting unilaterally may converge upon a system that exhibits substantial uniformity even when there is no objective external standard. This may be a system that makes sense or one that does not. Often convergence will be the result of beggar-thy-neighbor policies, in which case the result is not likely to be desirable. The recent trend to use only sales to apportion corporate income is the best example of this. (5)

Voluntary, Pre-emptive, and Reactive Coordination

States may undertake efforts to harmonize their taxes for at least three reasons: voluntarily, simply because it is the right thing to do; grudgingly, in an effort to ward off federal legislation or judicial decisions that would restrict state sovereignty over tax policy; or reactively, to meet the demands of federal mandates, incentives, or court decisions. It is, of course, sometimes difficult to know when states are acting voluntarily and when pre-emptively. Here, too, there may or may not be an objective external standard and the result of cooperation may or may not make sense.

Unless enshrined in a formal compact that is given the stamp of approval by the U.S. Congress, state attempts at both voluntary and pre-emptive harmonization generally provide individual states considerable discretion regarding whether to adopt the harmonized system. Such efforts might, for example, take the form of model laws or guidelines. If a model state tax law were to be adopted in its entirety by all sates, the result might be a highly harmonized system. But if not all states adopt the model law, and if those states that do adopt it significantly modify its provisions, harmonization will be less complete. This has been the experience under UDITPA, especially regarding the choice of apportionment formula. Though the jury remains out, it appears that experience under the SSTP may be better.

Federal Mandates, Incentives, and Court Decisions

Finally, the federal government may mandate harmonization or provide incentives for it, and the federal judiciary may also impose harmonization on unwilling states. Such federal actions are, of course, the well-spring of reactive harmonization. The International Fuel Tax Agreement (IFTA) is a clear example of this. As with convergence and harmonization, federal actions may produce results that make sense or that do not make sense. While IFTA brought order to the realm of fuel tax apportionment and appears to be a nearly unqualified success, P.L. 86-272 creates a nexus standard that makes no sense. (On IFTA, see McLure, Pitcher, and Turner, 2007, which draws heavily on Pitcher (2001) and Denison and Facer, 2005.)

Federal legislation, including incentives as well as mandates, can be either proscriptive or prescriptive. That is, it can say what states cannot do or would be penalized for doing, or it can say what they must do or would benefit from doing. In theory, court decisions are only proscriptive, as prescription inherently involves legislating. In fact, when proscriptive decisions take on the attributes of prescription the line between proscription and prescription becomes fuzzy. Thus the rulings in National Bellas Hess and Quill that a state cannot require out-of-state vendors that lack a physical presence in the state to collect use tax on sales made into the state come close to being prescriptive.

Unlike both voluntary and pre-emptive harmonization undertaken by states, both proscriptive and prescriptive harmonization, being legally binding on the states, may significantly restrict state choices. States may engage in political activity to prevent enactment of federal legislation that would have this effect, as in the case of the Willis Committee's proposals for uniform rules for the division of corporate income, to be discussed below. Proscriptive action is generally unlikely to produce a truly harmonized system, except in regard to the proscribed policy, as there may be many ways to react to a given prohibition. Prescriptive federal legislation may provide greater uniformity, but it may not provide the level of detail that is required for real harmonization.

Forces Creating Diversity

Uniformity is, of course, not inevitable. Taxes levied by state governments may not be uniform simply because the forces pushing for uniformity are non-existent or are not strong enough. (6) Where uniformity exists (or could exist, for example, if UDITPA were adopted by all states without change) it may unravel into diversity for similar reasons, because individual states reject rules that would deprive them of revenues, or because they depart from the previously accepted rules in an attempt to gain a competitive advantage, for example, by adopting sales-only apportionment to encourage economic development.

If enough individual states take unilateral action to gain a competitive advantage the system may eventually again display substantial uniformity, as apportionment formulas are now tending to do. But such beggar-thy-neighbor policies are not likely to produce a system that is also desirable.

Federal legislation and court decisions may contribute to the unraveling of uniformity, by explicitly or implicitly leaving decisions on certain aspects of tax matters to the states, rather than codifying or mandating continued adherence to the uniform system. The decision of the U.S. Supreme Court in Moorman (described below) arguably led to the wholesale abandonment of the equally weighted three-factor apportionment formula that previously had been used almost universally. Furthermore, changes in the Internal Revenue Code, particularly those that reduce the tax base, may also lead to unraveling, as when states "decouple" from generous federal provisions for depreciation.

FROM THE BEGINNING TO UDITPA