It is always exciting to find a new international tax paper written by the famous cohort of authors—our learned Professors J. Clifton Fleming, Jr. (BYU), Robert J. Peroni (Texas), and Stephen E. Shay (Boston College). These authors can be trusted to provide insight into carefully selected topics relevant to current issues in international tax. In each paper, they demonstrate profound knowledge and experience in the chosen topic, and share thoughtful policy suggestions. The new book chapter, Is Unilateral Formulary Apportionment Better than the Status Quo?, in The Allocation of Multinational Business Income: Reassessing the Formulary Apportionment Option (Wolters Kluwer 2020), is not an exception. It provides a condensed analysis of the arm's length standard and the rise of formulary apportionment as an alternative. Additionally, the paper suggests criteria for the cost/benefit analysis of unilaterally adopted formulary apportionment in both territorial and worldwide system paradigms. Readers with advanced knowledge of international tax will find this chapter to be interesting, and, thanks to the authors’ mastery of the topic, the paper is also accessible to readers with only a basic knowledge of international tax. I highly recommend this paper to professors who are looking for reading material on transfer pricing.
Many commentators have criticized the arm's length approach because (1) the arm's length approach is highly dependent on the particular facts and context of each case, (2) it enables multinational enterprises (MNEs) to game the system where MNEs with greater resources can overwhelm tax authorities, (3) the complexity and imprecision of the arm's length approach creates expensive and time-consuming controversies, (4) it does not properly recognize the internal efficiencies of the MNEs, (5) the comparable transaction data may not be available in many cases, and (6) it has not prevented profit shifting. I think the above succinctly summarizes the central criticisms of the arm's length approach.
The paper continues by noting that such deficiencies of the arm's length approach have resulted in the rise of formulary apportionment that uses a mathematical, factor-based formula to allocate the worldwide income of an MNE among various countries. The original formulary system uses an equally weighted, three-factor formula (assets, payroll, and sales), but alternative formulae may instead place greater weight on one or two factors. If such variation on the formula exists among countries replacing the arm's length approach with formulary apportionment, double taxation and double non-taxation might occur, resulting in incentives to shift profits or specific factors to low-tax countries.
While some commentators argue that at least the countries with major economies should adopt formulary apportionment with a uniform allocation formula in a coordinated way, the authors expect that there is little chance for such a uniform, multilateral formulary apportionment regime and that the more likely scenario is for countries to adopt formulary apportionment unilaterally. Under such a scenario, the problem caused by the mismatch of the formulae would persist.
After explaining that formulary apportionment does not require the adoption of a territorial system and that it can also be used in a worldwide system to identify foreign-source income for foreign tax credit purposes, the paper examines the problems that unilateral formulary apportionment might cause in both territorial and worldwide system paradigms.
First, a territorial regime substituting the arm’s length approach with unilateral formulary apportionment would face transaction costs and incentives to shift profits or business activity to low-tax countries, creating risks of double taxation and double non-taxation. The transition costs, such as renegotiating tax treaties with respect to Article 9 (transfer pricing using the arm’s length approach) and Article 25 (mutual agreement procedure) of the OECD Model, would be significant as well. The substitution of formulary apportionment may produce gains from the simplified system, but the authors expect that the gains may not be great.
Second, a conventional worldwide regime, which allows deferral of residence country tax until repatriation and cross-crediting with inappropriate income attribution and foreign tax credit limitation rules, would experience the same problems as in the territorial regime if it replaces the arm’s length approach with unilateral formulary apportionment. However, the authors have introduced a properly designed worldwide system that eliminates both deferral and cross-crediting in their previous works, available here (among others). The so-called “expanded worldwide system” can minimize the problems of transfer pricing even under the arm’s length approach cum source/income attribution rules, because the incentive for profit shifting is largely eliminated. Nonetheless, nonresident taxpayers would still have the incentive for base erosion through deductible payments in the expanded worldwide system, where the debate over arm’s length with source rules versus formulary apportionment would be meaningful.
Based on the thorough analysis discussed above, the authors urge countries that consider adopting formulary apportionment unilaterally to conduct a rigorous cost/benefit analysis. The criteria suggested by the authors is useful to policymakers in evaluating whether a country would be better off if it adopts formulary apportionment unilaterally compared to the status quo. Some readers may read between the lines and notice that the authors are less enthusiastic about formulary apportionment. However, even for the advocates of a formulary approach, the cost/benefit analysis criteria suggested in the paper would serve as a great tool to critically evaluate a proposal they endorse.