Understanding uniformity and diversity in state corporate income taxes.

Zodrow, George R. "Capital Mobility and Source-Based Taxation of Capital in Small Open Economies." International Tax and Public Finance 13 No. 2/3 (May, 2006): 269-4.

Charles E. McLure, Jr.

Hoover Institution, Stanford University, Stanford, CA 94305

(1) Gupta and Mills (2003) find that, as a percentage of revenue, costs of compliance with state corporate income taxes are roughly twice those of compliance with the federal corporate income tax. Of course, this reflects, in part, the fact that the federal tax rate is much higher than state rates. But if state taxes were substantially more uniform, compliance with them should not be costly. That, other things equal, costs increase with the number of states in which a corporation does business suggests to the authors (p. 370) that "tax compliance costs are largely driven by complexity and disconformity."

(2) This paper is not about tax competition and its effects, which will not be discussed further. There is, of course, a voluminous literature that attempts to determine whether, and under what conditions, tax competition is beneficial or harmful. See Wilson (1999), Zodrow (2003, 2006), Wilson and Wildasin (2004), and Wildasin (2006). Zodrow (2006, note 10), writes regarding models that disparage tax competition, "These models often suggest a 'race to the bottom' as tax competition eliminates capital income taxes, although this could just as easily be labeled a 'race to the top' as countries are forced to use efficient benefit taxes to finance their public services rather than inefficient non-benefit taxes on mobile capital."

(3) I have argued elsewhere that a sensible system of state sales taxation would reflect four rules: tax (virtually) all sales to consumers; exempt (virtually) all sales to business; subject to de minimis rules, treat sales by remote (out-of-state) vendors the same way as sales by local vendors; and simplify the system enough that implementing the third rule would not unreasonably burden interstate commerce. See McLure (2002) and references provided there. By comparison, the SSTP produced uniform definitions of broad categories of products, which states could either tax or exempt, and harmonizes some administrative procedures. As a result, the Uniform Sales and Use Tax Administration Act requires 15 pages and the Simplified Sales and Use Tax Agreement (SSUTA) another 70 plus pages. Moreover, Walter Hellerstein and John Swain (Hellerstein and Swain, 2007/2008 and Hellerstein and Hellerstein, 2007, Appendix 19A) devote more than 170 pages to explaining SSUTA.

(4) The Report of the Willis Committee (U.S. House of Representatives, 1964, pp. 95-136) provides a brief historical overview of the development of the tax up to the early 1960s, some of which is reproduced in Hellerstein and Hellerstein (2007), introduction to part IV. Hawaii enacted an income tax in 1901, but did not become a state until more than 50 years later.

(5) Some might argue that reducing corporate income tax rates and top rates on individual income are also beggar-thy- neighbor policies. Several observations are in order. First, neither the project on harmful tax practices of the Organisation for Economic Co-operation and Development or the European Union's code of conduct for business taxes considers low tax rates, per se, to constitute a harmful tax practice. Second, rate reductions may simply reduce rates to levels that more closely reflect benefits provided to the taxpayer, in which case they would be desirable on efficiency grounds. By comparison, it is hard to argue that sales--only apportionment produces taxation that is related to benefits. There is taxation when production is for the in-state market, no taxation to the extent that output is exported, but (assuming nexus in the state) full taxation when products made outside the state are imported. In this regard it might be noted that the General Agreement on Tariffs and Trade allows border tax adjustments (taxation of imports and rebate of taxes on export) only for indirect taxes such as the value added tax. Sales-only apportionment amounts to providing border tax adjustments for state corporate income taxes. See McLure and Hellerstein (2002).

(6) Hildreth, Murray, and Sjoquist (2005) present a formal game-theoretical discussion of the conditions under which it is expected that states would cooperate. They conclude (p. 583), "cooperation is difficult to achieve ... and that significant non-uniformity in state corporate tax systems exists should not be a surprise."

(7) There could, of course, be no federal model before 1913, when the Sixteenth Amendment to the U.S. Constitution was adopted, authorizing enactment of a non-apportioned federal income tax.

(8) Hellerstein and Hellerstein (2007, [paragraph] 7.02). Only Arkansas and Mississippi do not take federal taxable income as their starting point, and even those states' corporate tax bases conform substantially to the federal base. Despite this "broad conformity," state and federal tax bases exhibit important differences, to which Hellerstein and Hellerstein (2007, [paragraph] 7.03-17) devote 130 pages. The Willis Report (U.S. House of Representatives, 1964, pp. 128-29) indicates that in 1919 a committee of the NTA produced a study that suggested that the states enact a model statute based on the federal definition of income but concludes, "the model act seems not to have made much of an impression on the States since none adopted it."

(9) See U.S. House of Representatives (1964, pp. 115, 130) and Hellerstein and Hellerstein (2007, [paragraph] 8.03), which cites three well-known defects of separate accounting: expense, the need for transfer prices, and the economic interdependence of activities conducted in various states.

(10) Hellerstein and Hellerstein (2007, [paragraph] 8.06, note 175) note that the rationale behind the Massachusetts formula is discussed in the Proceedings of the NTA, 1950, p. 349 and following.

(11) T.S. Adams had done this in 1917; see U.S. House of Representatives (1964, p. 130).

(12) However, non-business income that arguably has it source in a given state, such as income from leasing a building and capital gains from the sale of a building, is allocated to that state. Non-business patent and copyright royalties are allocated to the various states on the basis of the licensees' use of the asset in the state.

(13) While this impression is difficult to demonstrate conclusively, it is corroborated by correspondence with Dan Bucks, former Executive Director of the Multistate Tax Commission for 17 years (now Montana Director of Revenue), Harley Duncan, Executive Director of the Federation of Tax Administrators, and Ben Miller, Counsel, Multistate Tax Affairs, for the California Franchise Tax Board. In 2004 corporations apportioning income (those with multistate operations) reported $1.12 billion in business income to California, compared to only $18.5 million in non-business income reported by apportioning corporations and $23 million in total income reported by corporations operating only in California; see California Franchise Tax Board (2005), Table C-2, p. 148. States that, unlike California, do not require combined reporting may show a higher proportion of non-business income, especially because they would likely treat substantial amounts of intercorporate dividends as non-business income to be allocated to states of commercial domicile (which would tax it as non-business income, if at all). On the other hand, following Mobil, states that do not require combined reporting may apportion dividends (or, for that matter, interest, royalties, and other intangible income) received from corporations with which the taxpayer has a unitary relationship.

(14) Warren (2005, p. 135) calls the rule for attribution of sales other than of tangible property in Section 17 of UDITPA "the weakest part of the act." He suggests that its drafters did not intend UDITPA to be applied to services and notes that the Multistate Tax Commission regulations do "a good job at fleshing out section 17." UDITPA's draftsman (Pierce, 1957, p. 781) notes, "Generally, it was felt that the provisions of Section 17 were the best that could be designed to cover the greater proportion of cases." He suggests (p. 780) that the relief provisions of Section 18 should be employed in unusual fact situations involving this kind of income.

(15) UDITPA's draftsman explicitly recognized the first two omissions; see Pierce (1957, p. 747).

(16) In Northwestern Portland Cement Co. v. Minnesota, 358 U.S. 450 (1959), the state asserted jurisdiction to tax the income of an out-of-state corporation that made 48 percent of its sales in the state. In Williams v. Stockham Valves and Fittings, Inc., 358 U.S. 450 (1959), the taxpayer's only activity in Georgia was the maintenance of a sales-service office with a sales representative and a secretary.

(17) Authors of the Willis Report may have been unwittingly prescient when they wrote (House of Representatives, 1964, p. 438):

[A] significant impact on revenues might be anticipated only for

those States which use destination-oriented sales factors in their

apportionment formulas.... This suggests that objection to the

Federal statute are closely related to the apportionment formulas

used by the States. But for the widespread and growing use of the

destination test, neither considerations of revenue nor

considerations of equity would create strong pressure in the

direction of sustaining jurisdiction to tax on the basis of

solicitation or similar activities. (Italics added.)