Issues in the Design of Formulary Apportionment in the Context of NAFTATax Law Review (1995)
This article responds to Professor Paul McDaniel’s Formulary Taxation in the North American Free Trade Zone. In his article, Professor Paul McDaniel addresses whether the NAFTA countries should adopt a system of formulary apportionment taxation. My commentary serves to highlight, with an emphasis on the state tax experience, the unnecessary restrictions imposed on Professor McDaniel’s proposal, the issues inherent in any formulary system, and the NAFTA-specific problems that could arise.
Professor McDaniel adopted two unnecessary constraints in his proposal. The first, the “water’s edge” constraint, which applies formulary apportionment only to income sourced within NAFTA countries and uses only NAFTA factors in the apportionment formula. He defends this water’s edge approach on the grounds that it may be both a requirement of existing tax treaties and a necessary concession to foreign countries that may resist worldwide unitary measures, though no case law exists to support the former; the latter is a political judgment call. A water’s edge approach would require source rules and arm’s length pricing rules to determine NAFTA sourced income. Sourcing in general is incompatible with a formulary system, and to do so negates one of the inherent benefits of formulary apportionment: eliminating the need to monitor transfer prices.
Second, Professor McDaniel’s emphasis on the existence of a unitary business presumably borrows that concept from U.S. Supreme Court cases dealing with state taxation of multistate income. He does not define a unitary business or defend its use. In the state context, the Due Process and Commerce Clauses require the existence of a unitary business in the context of formulary apportionment. There is no similar concept in the federal system and no need to incorporate the concept should NAFTA adopt formulary apportionment.
There are technical issues in any system of formulary apportionment, three of which are particularly relevant here. First, states are permitted to impose combined reporting requirements only on members of a unitary business. These combined reports require the in-state business to consolidate its income and factors with the income and factors of its unitary subsidiaries, regardless of whether or not those subsidiaries have any direct nexus with the taxing state. Combined reporting essentially treats out-of-state unitary subsidiaries as branches or divisions. Contrary to Professor McDaniel, a NAFTA approach should be free of the state condition that the members of the combined report be part of a unitary business. The only constraint on what entities should be included in the combined group is the threshold matter of a minimum ownership requirement.
Combined reporting must resolve the level of stock ownership of entities that must be included in a combined report. There are two common thresholds: a minimum 50% ownership test, or a “more than 50%” ownership test. The latter is more subject to manipulation by taxpayers seeking to avoid a combined report. These taxpayers can reduce their level of ownership to 50% while still maintaining control over the subsidiary. Fifty-fifty joint ventures would not be subject to combined reporting under a “more than 50% test
Designing an effective apportionment formula will likely be a complicated objective. The state experience suggests that the traditional three-factor formula composed of sales, property, and payroll is inefficient for activities that don’t constitute traditional manufacturing or merchandising. The state response has been the adoption of specific formulas particularly tailored to individual activities, though this solution necessitates the correct characterization of the taxed activity and an emphasis on arm’s length transactions between related businesses subject to different formulas. Any apportionment formula must be built on the principles of administrative simplicity, neutrality, economic nexus, inclination to tax, and sovereign control of tax.
A final issue is that of nexus, that is, the application of jurisdictional threshold rules. Professor McDaniel supports the traditional permanent establishment approach, but under this approach the use of receipts in an apportionment formula could assign income to countries without the jurisdiction to tax it. A preferred approach might be a threshold based on a predetermined amount of gross receipts, regardless of a significant physical presence.
As long as the NAFTA countries maintain their dissimilar tax codes, formulary apportionment applied in a NAFTA context has the unique effect of substituting treaty shopping for “rule shopping.” Mexico, for example, lacks provisions equivalent to the U.S. subpart F treatment of controlled foreign corporations. A U.S. parent could merge with its Mexican subsidiary to form a Mexican entity, avoiding subpart F. Ideally, the NAFTA countries would adopt uniform or equivalent provisions in their tax codes to avoid this kind of rule shopping, while also ensuring their revenue administrations are highly competent to avoid potential administration shopping.
Finally, Professor McDaniel suggests that dividends paid between two NAFTA corporations that are not part of the same combined report should be excluded from taxation, treating the distribution as though it occurred between two entities in the same combined report. While I question the logic of a 100% exemption, it is clear that transitional and tracing rules would be required to decide the order in which profits from the pre- and post-formulary regimes are considered to be distributed.
Citation InformationRichard D. Pomp, Issues in the Design of Formulary Apportionment in the Context of NAFTA, 49 Tax L. Rev. 795 (1994).