Kane & Kern: Progressive Formulary Apportionment — The Case For ‘Amount D’

Mitchell Kane (NYU) & Adam Kern (Covington & Burling, Washington, D.C.), Progressive Formulary Apportionment: The Case for ‘Amount D’, 171 Tax Notes Fed. 1713 (June 14, 2021):

In this report, Kane and Kern demonstrate the advantages of progressive formulary apportionment over other possible international tax reforms, and they show how it could be implemented seamlessly within the architecture of pillar 1.

Conclusion
In this report, we have introduced progressive formulary apportionment and urged that it be implemented within pillar 1 by establishing a new taxing right, amount D, in favor of developing countries. As we have acknowledged, our proposal is radical in important ways. Despite decades of debates about formulary apportionment, we are aware of no prior proposal in the tax literature or in the policy space that urges the adoption of apportionment factors that would use state-level macroeconomic or development indicators in addition to company-level factors. But in other ways, our proposal is simply a natural extension of the trajectory that we are observing in pillar 1.

Pillar 1 and amount A are, of course, about demarcating the rights of states to tax nonresident companies. (Although there may be pragmatic reasons, such as inversion risk, to curtail the taxation of resident companies, the absence of the right to tax those taxpayers under international norms is not thought to be a limiting factor.) Historically, the taxation of nonresidents has required at least some transactional connection between the state and the company that the state seeks to tax. Conceptually, one way to understand the shift from one-step to two- or three-step formulary apportionment — and the division between routine profit and some residual profit that this shift entails — is that it is an acknowledgment that our standard transactional tools run out at some point given the hyperconnectivity and integration of the global economy and of the large multinational corporations that play such an important role in it. The standard mechanisms (arm’s-length transfer pricing and separate entity accounting) and the standard logic (search for some nexus) work well enough for cases in which we can still readily determine physical presence and ascribe some ordinary return to the functions physically taking place within a jurisdiction. When the standard tools run out, however, we need a different approach — thus, the need to apportion residuals by formula.

In our view, the pillar 1 blueprint is misguided in its attempt to cling to historic demands for nexus when undertaking this task. The pillar 1 blueprint does so by tying some delimited business sectors to the amorphous concept of value creation. We think this effort is bound to fail because it is at odds with the very indeterminacy of the question about linking profit to jurisdictions that drives the need for reform in the first place. The recent call by the United States to replace the activity tests with comprehensive scoping has amplified this point.

We have not offered anything like a comprehensive normative defense of how best to allocate a portion of the international tax base in the absence of historic concepts of nexus. We have indicated the need for that project and begun its undertaking elsewhere. Here, we simply observe that it is difficult to see how the current contours of pillar 1 can be squared with a justifiable approach to this basic problem.

If we accept that any sound justification would be based on some mix of efficiency, distributional concerns, and administrative concerns, it is hard to see why market jurisdictions should have the exclusive rights to tax nonresidents on their residual profit. In terms of distribution, the only argument that would seem to be on the table to justify that position is that the market jurisdictions continue to contribute to company profit (even absent any reflection of historic nexus standards) because they play a crucial role as the ending point of a company’s supply chain. But we have questioned that contributory argument in this report. Further, even if one were wedded to the idea that nexus in the future will just mean participating in the company’s supply chain, many other jurisdictions touch global supply chains, such as jurisdictions in which a company is not resident but sources labor supply, which the pillar 1 blueprint ignores. In terms of efficiency, apportioning residual profit based on final sales may well be efficient as compared with apportionment that tracks more mobile production factors. But, as we have argued, that apportionment is even less efficient than an approach that does not look to company transactional factors at all.

We are rapidly approaching an inflection point in the design of the international tax system in which nexus will not play its historic role in allocating rights to tax large portions of taxable profit of multinational companies to nonresidence jurisdictions. The time is thus ripe to think in new ways about how to allocate taxing rights in that world. In this respect, progressive formulary apportionment, implemented through amount D, has highly desirable distributional characteristics, is efficient, and could easily be administered through the emerging global international tax architecture.