Last October, the OECD/G20 announced a global deal that sought to remake the international tax system. Key to that change was the commitment of the many nations, including the U.S. and members of the E.U., to support, in principle, two pillars. Pillar One focused on eliminating physical presence requirements thereby allowing market countries to tax digital platforms. Pillar Two created a minimum tax backstop in residence countries. But as Reuven Avi-Yonah, Young Ran “Christine” Kim, and Karen Sam, show in their article A New Framework for Digital Taxation, forthcoming in the Harvard Journal of International Law, for Pillar One, the declaration of victory is premature. And so-called Digital Service Taxes (DSTs) may not be so terrible either.
Pillar One grew out of the failure of the original BEPS proposal and the rise of DSTs, which the U.S. opposed as discriminatory against its tech giants. In principle, it targeted the largest multinational entitles (MNEs) and developed Amount A, which eliminated the traditional physical presence nexus for source-based taxation and created some apportionment of this residual profit based on revenue earned in a particular market jurisdiction. Crucially, Pillar One is to replace the haphazard structure of DSTs, thus placating the U.S. Furthermore, the details regarding Amount A are to occur in a multilateral tax instrument.
But the article asserts that these two key points that pushed toward unity also sow the seeds for failure. With respect to DSTs, there is a bit of a chicken-and-egg problem. Many market jurisdictions, like the E.U. countries, are forging ahead with their DSTs. They will not want to drop DSTs until the U.S. adopts Pillar One and the multilateral instrument. But the U.S. seems unlikely to try to adopt a Pillar One instrument unless DSTs are dropped.
Furthermore, implementing the multilateral instrument not only requires complex negotiations. But it also requires adoption, which raises particular problems when it comes to the U.S. These problems stem from U.S. politics and the status of treaties in U.S. domestic law.
Currently, the U.S. has numerous bilateral tax treaties. But implementing Pillar One through a multilateral instrument would require overriding provisions of these treaties. The best practice is to have the Senate ratify the treaty with a 2/3 vote. But many Republican Senators have decried the Administration for not consulting and working with them through the negotiations around Pillar One. The need for other countries to repeal the DSTs also hampers the ability to receive a 2/3 vote in the Senate for Pillar One.
But U.S. law provides unique features for treaty override. One is through a Congressional-Executive agreement, whereby the President negotiates an agreement, and it is passed as legislation through both chambers of Congress. The other is through the normal legislative process of enacting a statute, because a domestic federal statute passed after a treaty can override treaty provisions. But these come with significant costs. First, tax treaties are treaties. Overriding their provisions through either of these means sets a precedent that tax treaty obligations can thus be overridden by normal legislative procedures rather than the 2/3 vote in the Senate to ratify a treaty. While legally nothing stops this type of action, it creates a dangerous norm. The article argues that such overrides can lead to whipsawing when new Congresses and Administrations change their mind. Second, even these means are not guarantees of success for implementing Pillar One. After all, given the slim majorities in the House and the Senate, using this means requires unity among the Democratic Caucus, and it is unclear that such unity can be ensured.
The article also argues that the DSTs themselves may not be discriminatory and threatened U.S. tariffs are a dangerous overaction. While the U.S. claimed that the French intentionally sought to discriminate against U.S. digital MNEs, when the statements cited are viewed in the frame that the French are tired of digital companies not paying their fair share of taxes to market jurisdictions, the notion of discriminatory intent loses its edge. Furthermore, when it comes to discriminatory effects of the DST, U.S. complaints also fail. Often U.S. complaints fail to note that the point of all DSTs, including the French one, is to tax purely digital platforms. A DST is unnecessary for targeting companies that already have a tax nexus in France. Yet, many of the examples the U.S. cites as unfairly escaping the DST are exactly those types of companies. Meanwhile, the U.S.’s tariffs on sparkling wine only angered the French and harmed U.S. wine importers and consumers. The article shows that little harm fell onto French winemakers.
The article closes with two alternatives to the Pillar One approach. The first is multilateral, found under the U.N.’s Model Tax Convention Article 12B. It eliminates the physical presence requirement and allows market countries to tax income from Automated Digital Services. Importantly, the Article 12B gives MNEs the option of gross or net taxation. It also uses a withholding tax mechanism, which is easier to administer and works on the current foundations of the international tax system. In creating a structure this way, Article 12B makes some radical changes while building on mechanisms in place, thereby simplifying the approach.
The other is a unilateral Data Excise Tax, as perhaps the next generation of DSTs. The paper makes a strong proposal to tax on the basis of data collected. Doing so avoids some of the problems of trying to figure out the exact value of the data collected. This new provision also eliminates old physical presence nexus. It avoids the questions of double taxation because it is not based on income or revenue. It is a pure excise tax.
The article raises important points for not only tax experts but also for international law and international relations experts to consider. The paper reminds one of the old proverb that perfect can be the enemy of the good. Perhaps it is better to have something that stops aggressive double non-taxation through some digital tax than seeking the perfectly negotiated solution.
The article also forces those of us in the U.S. tax sphere to think about how our actions affects broader international relations. The paper paints the U.S. as a bad actor in the international tax game. While national interests are always important, the article raises the point that the rules-based order and network of alliances serves U.S. interests in ways that companies like Apple and Google cannot. Indeed, recent events have shown the importance of these relations, especially with countries in the E.U. like France. We would do well then, in the tax sphere, to work in a way that buttress our allies, rather than getting involved in nasty DST fights to protect our national champion companies.