INSIGHT: Global Tax Policy—What is the Best Alternative to Uncoordinated Unilateral Tax Measures?

In a recent Bloomberg Insight piece my friend and fellow laborer in the vineyards of international tax policy Will Morris makes the important point that the OECD/Inclusive Framework’s current work on global income taxation of multinational enterprises could have serious consequences for all multinationals, not just those considered “digital” (“The OECD Digitalization Project: It’s Not Just About Digital,” Daily Tax Report, July 16, 2019). He concludes his article with another point: that “the alternative to this project is chaos”—that is, a raft of uncoordinated unilateral tax measures aimed at nonresident multinationals. This echoes statements by U.S. Treasury officials involved in the work and by the leadership of the OECD Centre for Tax Policy and Administration.

The statement implies that chaos will be avoided if an agreement is reached. It’s worth testing the validity of that proposition, since it seems possible that a bad agreement could also result in chaos, or worse. The goal should be an agreement that will make things better.

Clearly, in the absence of multilateral agreement on how to tax nonresident multinationals, it is likely that there will be uncoordinated unilateral measures of the kind that are already appearing around the world. It does not follow, however, that agreement on the proposals being considered by the OECD’s Inclusive Framework will prevent such measures. Indeed, the proposals, if adopted without adequate grounding in neutral principles, could give rise to greater risks of double taxation than would otherwise have existed.

As many readers will know, the program of work on digitalization, which has been approved by the G-20, has two “pillars.” Each pillar has two parts. Pillar One would create new taxing rights through new rules on (1) tax nexus, allowing countries to impose income tax on remote sellers’ profits allocated to the country, regardless of physical presence, and (2) allocation of a multinational’s profits—giving additional taxing rights to market jurisdictions by reallocating profits toward the sales end of the value chain. Pillar Two involves a global minimum tax regime involving (1) residence-country taxation of low-taxed foreign subsidiaries and (2) source-country taxation of certain deductible outbound payments via either withholding tax or denial of deductions.

Both pillars have elements that depart radically from longstanding international tax standards that were recently restated by the OECD and G-20 in the final reports of the BEPS project, issued in October 2015. The allocation of profits among the members of a controlled group of companies constituting a multinational enterprise is currently based on the OECD Transfer Pricing Guidelines, which were revised under Actions 8-10 of the BEPS project and are followed by every country in the world except Brazil. (Notably, the OECD and Brazil recently announced that they are pursuing alignment of the Brazilian rules with the Guidelines.) The Pillar One work, however, would depart from the Guidelines in order to give more taxing rights to market countries, although there does not yet appear to be a principled basis for determining how much more those countries should get.

The program of work contemplates measures to address increased risks of double taxation and tax disputes between countries as a result of Pillar One and Pillar Two implementation. U.S. Treasury officials have said that a universal commitment to robust dispute resolution mechanisms is likely to be an important part of whatever is agreed. This would also be a radical departure (in a good way) from the current policy of the majority of countries around the world.

U.S. Treasury and other participants in this work have justified their support for the process on the basis that, if nothing is done, market countries will take matters into their own hands and unilaterally enact new taxes on nonresident businesses’ income from sales to local customers. This is undoubtedly true for some countries, at least with respect to the types of businesses and selling arrangements targeted by the unilateral measures taken, or threatened, in recent years. But it cannot be assumed that things will be better if some, but not all, countries sign up to follow the new Pillar One and Pillar Two proposals, especially if those proposals allow for flexibility in the details of what each country is agreeing to do.

Complete consensus among the 131 members of the Inclusive Framework will not be required in order for a proposed solution to be recommended to the G-20 next year. It seems likely that agreement among a significant cross-section of countries, including all of the G-7 and a majority of the G-20, plus a majority of the other members, will be sufficient. Thus, a “consensus solution” may ultimately be recommended over the objections of a number of countries. In addition, the recommended solution might permit individual countries to determine certain design features unilaterally. For example, it might be left to each country to state the minimum rate below which income would be considered “low-taxed” for the purposes of the Pillar Two regime in each country.

In addition, some countries, including the U.S., the U.K., France, and Germany, have enacted new tax measures in recent years that may conflict with whatever is agreed regarding Pillars One and Two. Given the vagaries of domestic politics, it cannot be said with confidence that all of those laws would be changed. An agreement could, however, include a list of the laws in question together with an explicit commitment by the relevant countries to use their best efforts to repeal them or make necessary amendments.

Assuming that the OECD process can produce a political agreement in 2019 or 2020 on how to allocate a multinational business’s profits among the various countries touched by that business, there is no guarantee that the agreement would hold in the future, especially if it is not grounded in neutral principles but is more in the nature of a payoff in an attempt to buy tax peace. When a future government in any given market country decides that it wants more tax revenue from nonresident multinationals, it will not be constrained in turning the dials of its allocation formula accordingly, if the numbers in the formula are just numbers based on countries’ relative bargaining power as of mid-2020.

Given that the OECD digitalization project has powerful political backing and is therefore likely to result in some sort of recommendation to the G-20 next year, it would be pointless to argue that it should simply be abandoned. It is important, however, to recognize that the Pillar One and Pillar Two proposals are fraught with risks that could be as bad, or worse, than the alternative chaos of uncoordinated unilateral measures. Perhaps there is a better way.

Taxing the net income of a multinational business in each of the various jurisdictions relevant to that business is inherently complex and difficult. The Inclusive Framework members should consider stepping back and looking at other possible ways of addressing the digitalization-related issues. For example, the recently published OECD “Going for Growth” report for 2019 recommends less reliance on income taxes and greater use of other types of taxes. Revenue from a globally agreed approach to imposing such taxes on large businesses might then be allocated among countries worldwide, alleviating the pressure in market countries to extract income tax revenue from nonresident multinationals.

This column does not necessarily reflect the opinion of The Bureau of National Affairs, Inc. or its owners.

Jeff VanderWolk is a partner at Squire Patton Boggs (US) LLP in Washington