Ruth Mason (firstname.lastname@example.org) is the Edwin S. Cohen Distinguished Professor of Law and Taxation at the University of Virginia School of Law.
In this article, the author discusses reactions to the recent agreement on pillars 1 and 2 of the OECD/G-20 base erosion and profit-shifting project.
A year ago, I published an article arguing that the OECD/G-20 base erosion and profit-shifting project left international tax at a crossroads with a few obvious alternative courses — more multilateral cooperation, increased unilateralism that would threaten the gains of BEPS, or a reversion to a predominantly bilateral regime.1 The agreement in principle on pillars 1 and 2, announced July 1 in a statement by 130 of the inclusive framework countries,2 suggests that countries overwhelmingly prefer multilateralism over the other options.
This article offers initial reactions to the agreement and seeks to contextualize it into larger debates.
How We Got Here
The earliest precursor to the 2021 compromise was the “1920s compromise,”3 which generally allocated passive income to taxpayers’ residence states and active income to the source states. Suitable for a pre-digital era, tax nexus in the 1920s required physical presence. Devised primarily for use among relatively rich countries, the 1920s model evolved into the modern OECD model treaty, now used by rich and poor countries alike. Many view that model as fundamentally flawed when used between rich and poor countries because it systematically shifts revenue from the poor to the rich country.4
Countries that suffered revenue losses from treaties evidently regarded those losses as worth the gains that treaties brought.5 As the economy “ephemeralized,”6 however, the model gained new and powerful opponents. Western European countries, long accustomed to being on the winning side of tax treaties (at least in their bargains with less-developed countries) chafed at the rigidity of the physical presence nexus requirement. The internet, new business practices, and, yes, accommodating tax rules in both the United States and Europe enabled some of the world’s most profitable companies — nearly all of them U.S.-headquartered — to avoid tax in countries where they had extensive involvement. Most of the FAANGs — Facebook, Apple, Amazon, Netflix, and Google — easily avoided establishing tax nexus with Western European countries.
Although some countries, such as Ireland and Luxembourg, turned a tidy profit from U.S. multinationals’ tax avoidance activities, the most powerful countries in Europe — particularly France and Germany — saw the need for fundamental change. However, the possession of a veto in tax matters by each EU member ensured that fundamental reform wouldn’t be homegrown in Europe. At the same time, as countries that had long been disadvantaged by the residence-state bias in tax treaties grew in economic and political power, their interests became harder to ignore. India and Brazil were perhaps the most vocal, but many others were similarly situated.
The 2008 crisis and rising popular sentiment against corporate tax dodging spurred action from previously complacent governments. The G-20 tasked the OECD with finding solutions to the corporate tax avoidance problem, establishing the BEPS project. Contrary to the dismissive judgment of many academics, the BEPS project made meaningful changes that appear to be enduring.7 The addition of the G-20 on an equal footing with the OECD countries in international tax policymaking, along with the establishment of the inclusive framework — open to any country willing to commit to the BEPS minimum standards — constitute major changes and a needed acknowledgment of the inability of the OECD countries to make international tax policy without regard to the rest of the world.8
The pragmatic credo of BEPS was: Agree where you can, and otherwise move on. This pragmatism led to vague and hortatory pronouncements on some issues (such as controlled foreign corporations), but to highly detailed recommendations on others (such as interest barriers and hybrids).
The BEPS countries plugged some of the most notorious gaps in the international framework, thereby growing global tax revenue. But because states didn’t squarely address how to divide the resulting revenue, BEPS was unstable. Various unilateral measures, such as European digital taxes and the Indian equalization tax, reflected not only the instability of the BEPS compromises but also growing dissatisfaction with the larger 1920s compromise on tax treaties that shifted too much revenue from source states to residence states.
BEPS was a starting point: It repeatedly raised, but failed to answer, fundamental allocation questions. And it made little headway on the tax competition questions that concerned France and Germany but couldn’t be resolved solely within Europe. And while it didn’t fundamentally tackle source and residence or tax competition, BEPS built robust institutions of multilateral cooperation, including the cooperation between the OECD and the G-20 and the establishment of the inclusive framework. BEPS also marketed to the world the technical prowess of the OECD, which had the right experts and leadership for the moment. To the staff at the OECD, no issue within the BEPS project seemed too big or too complex, no deadline too close. The BEPS project built transnational relationships and the infrastructure for resolving the very distributional questions that it highlighted.
Pillar 1: Nexus on Demand
There was no pause between the delivery of the final reports of the BEPS project and the far more complicated and controversial work on the two pillars. Pillar 1 addresses reallocation; it picks up where BEPS action 1 on the digital economy left off. Unable to agree on how to deal with the digitalization of the economy and what it should mean for tax treaties, the BEPS countries essentially gave up on the problem of taxing the digital giants, offering the explanation that the digital sector couldn’t be ring-fenced from the rest of the economy. This was, of course, the U.S. view. The United States had no interest in assuring that other countries could tax U.S. tech companies, even if the United States didn’t intend to currently tax those companies.
Despite stalling by the United States, some countries were unwilling to live with the outcome of the physical presence nexus rule as it applied to Big Tech. Some devised digital taxes. These were turnover taxes on narrow income streams associated with tech companies, such as online ad sales and sales of user data. Because they applied only to very-high-turnover companies (usually to companies with more than €750 million in annual turnover) and to very narrow income streams, these new digital taxes disproportionately targeted U.S. companies while leaving domestic competitors untaxed.9 After a proposal for an EU-wide digital tax failed in the face of opposition by Ireland, Sweden, and Denmark, about half of the EU countries — including, most prominently, France — enacted or proposed unilateral digital taxes.10
The idea of pillar 1 was to repeal digital taxes in exchange for addressing dissatisfaction with the income allocation that resulted from tax treaties. The countries would agree on a new treaty nexus that didn’t require physical presence. The big questions for pillar 1 were: What would the new nexus be? And how would income be attributed to the new nexus?
The July 1 statement contains no real surprises for those who have followed the recent negotiations. It contemplates a nexus for companies of a specific size that derive a specific threshold of revenue from the market state. Following a recent U.S. proposal, pillar 1 applies to companies with global turnover greater than €20 billion11 that exceed a 10 percent profitability threshold. This addresses two complaints about digital taxes — that they apply to loss-making companies and that target a specific sector that traditionally has been dominated by U.S. companies. Although the revenue threshold in the July statement is high, it isn’t industry specific.12 According to an estimate by Martin Simmler and Michael P. Devereux, pillar 1 would affect 78 companies.13
Countries making requisite treaty modifications consistent with the proposal would be entitled to tax in-scope nonresident companies provided that the nonresident company derives at least a €1 million from the jurisdiction. Interestingly, this nexus threshold goes down to €250,000 for jurisdictions with a GDP less than €40 billion. To put this in perspective, in Wayfair, the Supreme Court approved a sales tax collection obligation for nonresident companies that was triggered by $200,000 of sales annually into South Dakota.14
The amount to be allocated (amount A) to this nexus is a range of 20 to 30 percent of residual profit (defined as pretax profits as a proportion of revenue determined using financial accounting standards).15 This amount will be split among the states that possess the new nexus according to rules still to be agreed on. Countries possessing the new nexus will also be entitled to tax “amount B,” which is an arm’s-length allocation to in-country marketing and distribution activities, with details to be worked out. Double tax will be relieved with details to come, but the July 1 statement is imprecise about these and other much-debated points, including segmentation.16
Again, nothing in the statement is a big surprise, and much remains to be determined. Commentators will debate the statement, but it is worth noting that pillar 1 represents a meaningful compromise. No country got all it wanted, but none seemed to go away empty-handed, either. Countries threatening digital taxes or that received too little source tax under the old model got a new nexus and a larger share of income to tax. The United States, in contrast, may have preferred no new nexus at all, but if there had to be a new nexus, it wanted European companies — not just U.S. digital giants — in scope. At €20 billion in revenue for in-scope companies, the proposal will allocate significant income while affecting only a small number of very large companies.
According to an early estimate by Simmler and Devereux, the proposal described on July 1 will tax more European companies, and tax more companies across more sectors, than prior proposals.17 The United States also got its safe harbor, but seemingly only in appropriate cases, and it got its long-sought commitment to binding arbitration, at least with countries that have developed mutual agreement procedure programs.
Pillar 2: Minimum Tax
Ever since the United States adopted CFC rules during the Kennedy administration, Treasury officials have tried to persuade other countries to do the same. The reasons are plain: Once a country has a minimum tax, it wants all the others to have one, too, to discourage companies from migrating to escape minimum taxation.18 By shoring up and expanding residual tax, pillar 2, the minimum tax proposal, is principally about preventing profit shifting and reducing tax competition. Therefore, it chiefly benefits rich, high-tax countries. Unlike pillar 1, which has been touted as not affecting real economic activity, the pillar 2 proposal would affect tangible investment as well as paper profits.19
Arguments have been made that minimum taxes will empower developing countries to tax. The argument goes like this: If a German-headquartered multinational will be taxed at 15 percent — no matter where in the world it earns its income — developing country A has no need to engage in destructive tax competition with developing country B to attract investment from German-headquartered multinationals. Instead, both A and B can tax German companies at 15 percent, secure in the knowledge that Germany’s minimum tax would ultimately impose that result anyway. The problem with this argument, of course, is that it removes agency from both A and B — neither can lower its taxes to compete with Germany or each other.
Germany’s sales pitch to developing countries A and B thus seems a little strained, and how much the minimum tax really constrains tax competition will depend in part on any substance carve-out from pillar 2.20 But developing countries A and B aren’t really what concerns Germany. Instead, Germany has in its sights Ireland and other low-tax EU states. The EU has spent the last 70 years liberalizing — it eliminated border tariffs and withholding taxes on interest and royalties; it wrote and strictly interpreted rules preventing discrimination against cross-border commerce; and it took various other steps to make the EU “open for business.”
In the process, however, the EU removed many of the transaction costs that inhibit tax planning. This lack of internal barriers (especially the lack of withholding on intra-EU base-eroding payments) combined with opportunism on the part of some small EU member states to create a perfect storm for within-Europe base erosion. The BEPS project and attendant political pressure helped a little. They addressed some particularly egregious practices, such as Irish tax residence rules that facilitated stateless companies and the Luxembourgish propensity for same-day transfer pricing rulings. But BEPS did not dare touch tax rates.
Nor could the EU enact minimum tax rates because Ireland (whose 12.5 percent rate is below the OECD’s 15 percent proposed minimum rate) and other low-tax EU states (such as Estonia and Hungary) possess a legislative veto on EU tax matters. The commission threatened to use obscure provisions in the Treaty on the Functioning of the European Union to counteract the tax veto, but that was an empty threat because the special provisions also require unanimity.21 Hence France and Germany turned to the OECD, hoping, as the United States had since the 1960s, that there would be safety in numbers on minimum taxation. If enough states enacted minimum taxes, there would be no safe exit for profitable multinationals; profit shifting and inversions wouldn’t help.
Finally, even if, within the EU, France and Germany could not exert sufficient pressure on Ireland to convince it to give up its veto, maybe the United States and other non-European high-tax countries could be called in to help. As one of only a handful of inclusive framework countries that has not signed on to the July 1 statement, Ireland apparently remains unpersuaded of the virtues of tax-rate harmonization.
But the big EU states don’t necessarily need the EU holdouts to capitulate. A benefit of a minimum tax is that it counteracts profit shifting and tax competition even if only one state adopts it. But more is more when it comes to fiscal fail-safes like minimum taxes.22 More states with minimum taxes, especially if they coordinate the rate, means fewer possibilities for multinationals to escape either the source-based or the residence-based minimum-tax rules.
Regarding pillar 2, the July 1 statement is equivocal. Pillar 2 is elective; it sets out guidelines matching those previously discussed publicly for minimum taxes at source and residence. Again, there are no surprises here compared with previously released information, although the countries confirmed that the residence state top-up tax would take precedence over the source state top-up, a priority rule that benefits rich countries. As expected, the rule would apply to very large companies (€750 million in revenue), though countries may apply minimum taxes to domestically headquartered companies that earn less. As expected, financial accounting rules would determine whether the minimum tax applies; the minimum rate would be at least 15 percent; there would be a de minimis exclusion and a substance carveout based on tangible assets and payroll; safe harbors would be announced; the top-up will apply on a per-jurisdiction basis (unlike the U.S. global intangible low-taxed income regime), and so on.
A Really Big Deal
Throughout this article, I have characterized details revealed in the statement as unsurprising because the various proposals were previously published and much debated. The true surprise is that the July 1 statement exists at all. That 130 countries — almost but not all the members of the inclusive framework23 — with diverse interests and conceptions of fairness managed somehow to agree on a direction for the future of international tax is surprising, and it shouldn’t be overlooked or taken for granted. Governments and officials sank enormous energy into the project. Although much remains to be done, and success is by no means assured, countries have displayed an impressive commitment to compromise and multilateralism.
Moreover, although some nongovernmental organizations, including Oxfam International, have criticized the statement, portraying it as the status quo,24 we should be clear-eyed about what this agreement in principle really does. Pillar 1 proposes a new nexus. That must be regarded as a significant change, even if the amount of revenue allocated to that nexus isn’t as high as some would prefer.25 Even more important, the nexus would depend on consumer-side activity, which is a radical departure from past practices.
Perhaps most significant, the countries reverse-engineered a distributive rule. They identified an amount of income for reallocation. This was an “amount” of income earned by companies with certain profit characteristics, not a “type” of income as in traditional treaties. Having agreed that this amount should be reallocated, the countries invented a nexus to enable taxation by the countries that it was agreed ought to be entitled to tax that amount. In this way, the countries bootstrapped nexus to a desire to reallocate. We typically think of allocation as following nexus, but in pillar 1, nexus followed allocation. This may not be the robust formulary apportionment that some advocated, but it is certainly a step down that road.
Another significant development is the agreement’s departure from the arm’s-length standard for the division of amount A among the countries with the new nexus. Even though we don’t know the precise form the departure will take, the July 1 statement reflects an acknowledgment that arm’s length isn’t fit for all purposes. This will no doubt disappoint proponents of the arm’s-length standard, who see pillar 1 as not only an affront to what they regard as a principled conception of income allocation but also a threat to the arm’s-length standard as it exists outside the pillar 1 agreement. By the same token, advocates of formulary apportionment will see important if limited progress in pillar 1.
Pillar 1’s two-tiered nexus threshold —which changes depending on the GDP of the market state — likewise represents an important acknowledgement that different thresholds may be appropriate for smaller economies.26 Although it is a large leap from this acknowledgement to, for example, asymmetrical withholding rates, it is nevertheless significant that 130 countries agreed, even if only in principle, to depart in this way from the formal reciprocity of tax treaties.
Oxfam also expressed disappointment with the distributional outcome of pillar 1,27 and although the changes to the concept of nexus are significant, they redistribute among rich countries more than they do between rich and developing countries. Because pillar 1 allocates to market countries, and many rich countries have large markets, a significant amount of the income described by pillar 1 will remain in the hands of rich countries.
Setting aside the difficult political and perhaps unresolvable moral considerations embedded in claims that the international tax system must redistribute from developed to developing states,28 we learned something strategic from pillar 1. When France and other countries imposed digital taxes — and didn’t back down in the face of threatened U.S. tariffs — they made a significant difference. Thus, a key lesson from the 2021 compromise may be that greater source taxation must be taken, not given.29
As for pillar 2, it was always a project principally by and for high-tax states. In pillar 2, we see the United States use the OECD for internal politics (much as France and Germany have used pillar 2 to pursue goals they could not achieve within the EU). The Biden administration would like to raise the U.S. corporate tax rate and the GILTI rate to fund its infrastructure plan. Pointing to (ideally a U.S.-led) agreement on high minimum tax rates abroad may help with domestic politics. Specifically, a global commitment to high minimum taxes would help the Biden administration persuade Congress that its rate proposals won’t harm U.S. companies abroad or cause them to flee.
Pillar 2, like BEPS before it, made use of fiscal fail-safes — rules under which if one state doesn’t tax, another automatically fills the void.30 Like the BEPS anti-hybrid rules, pillar 2 uses redundant fiscal fail-safes, with the income inclusion rule imposed by the residence state, and the undertaxed payment rule imposed by the source state if the residence state lacks an income inclusion rule. An admittedly simplistic way of thinking about the rules is as involving three types of states: (1) high-tax states, both rich and developing, that are pushing for minimum taxes to curb profit shifting and corporate tax competition; (2) rich, low-tax states that oppose the minimum tax because it will prevent them from (in the words of their neighbor-adversaries) poaching (mostly paper) profits; and (3) poor, low-tax states that oppose the proposal because it will prevent them from using low tax rates to attract real investment and activity.
Even if most people agree, and not all do, that the states in group 1 ought to prevail over the states in group 2, there is nothing approaching consensus that the states in group 1 ought to be able to cut the states in group 3 out of the competition for real business and investment. The pillar 2 proposal doesn’t distinguish between groups 2 and 3. The appeal for the states in groups 2 and 3 is supposed to be that they can stop competing with each other, and they can get at least 15 percent tax. Pillar 2 lumps groups 2 and 3 together, but the need to distinguish between the two groups suggests that the OECD may want to reexamine its policies on, for example, tax sparing and asymmetrical withholding rates.31
That pillar 2 originated as a plan to protect the tax bases of rich countries doesn’t mean that developing countries cannot benefit from it. Nigeria and Argentina have consistently warned that the 15 percent floor is too low to meaningfully combat profit shifting from source states. Here, too, the details remain to be negotiated. The July 1 statement merely states that the floor will be at least the 15 percent already agreed to by the G-7.
The countries released the statement on July 1 to meet a self-imposed deadline. The United States and France have been threatening each other with digital taxes and retaliatory tariffs. Despite leaving some issues open, the July 1 statement signals that the countries have not failed and won’t — at least for now — descend to trade war. But they have set for themselves another impossibly short deadline (the October G-20 meeting in Venice) to work out the details of the proposal.32 And implementation of the proposals described on July 1 will depend (at least from the U.S. perspective) on treaty partners dropping their digital taxes.
The U.S. Senate recently proved that it is still capable of ratifying uncontroversial treaty protocols, but who knows what the evenly divided Senate will make of nexus on demand. Finally, whatever details the countries eventually hammer out on pillar 2 will require a significant derogation to accommodate GILTI (and, if the Biden administration has its way, a SHIELD (stopping harmful inversions and ending low-tax developments) that differs from the OECD-contemplated undertaxed payments rule. Thus, the July 1 statement is an important marker, but significant issues remain outstanding.
1 Ruth Mason, “The Transformation of International Tax,” 114 Am. J. Int’l L. 353 (2020).
2 OECD/G-20, “Base Erosion and Profit Shifting Project: Statement on a Two-Pillar Solution to Address the Tax Challenges Arising From the Digitalisation of the Economy” (July 1, 2021).
3 Michael J. Graetz and Michael M. O’Hear, “The ‘Original Intent’ of U.S. International Taxation,” 46 Duke L.J. 1021 (1997).
4 Tsilly Dagan, “The Tax Treaties Myth,” 32 N.Y.U. J. Int’l L. & Pol. 939 (2000).
5 Honduras did not, however. In an example cited in every international tax policy class, Honduras unilaterally terminated its treaty with the United States upon concluding that the treaty didn’t bring enough new investment to justify its revenue cost.
6 Buckminster Fuller described ephemeralization as the ability to do “more and more with less and less until eventually you can do everything with nothing.”
7 For the argument that the academic reception to the BEPS project took an unnecessarily narrow view, see Mason, supra note 1.
8 For more background on the July 1 statement, see Mason, supra note 1; Rasmus Corlin Christensen and Martin Hearson, “The New Politics of Global Tax Governance: Taking Stock a Decade After the Financial Crisis,” 26 Rev. Int’l Pol. Econ. 1068 (2019); Allison Christians, “BEPS and the New International Tax Order,” 2016 BYU L. Rev. 1603 (2016); and Itai Grinberg, “The New International Tax Diplomacy,” 104 Geo. L.J. 1137 (2016).
9 Mason and Leopoldo Parada, “Digital Battlefront in the Tax Wars,” Tax Notes Int’l, Dec. 17, 2018, p. 1183.
10 Elke Asen, “What European OECD Countries Are Doing About Digital Services Taxes,” Tax Foundation (Mar. 25, 2021).
11 According to the statement, this threshold could go down to €10 billion after seven years.
12 There are carveouts for the financial sector and extractives, but proposals failed that would have applied pillar 1 only to automated digital services or consumer-facing businesses.
13 Simmler and Devereux, “Who Will Pay Amount A?” Oxford Centre for Business Taxation blog, July 5, 2021.
15 The income targeted by the pillar 1 proposal aims at so-called excess profits, including location-specific rents, that can be taxed without significantly distorting location decisions. See Joseph Bankman, Mitchell Kane, and Alan O. Sykes, “Collecting the Rent: The Global Battle to Capture MNE Profits,” 72 Tax L.R. 197 (2019); and Wei Cui, “The Digital Services Tax: A Conceptual Defense,” 73 Tax L.R. 69 (2019).
16 Segmentation would allow the proposal to apply to profit-making parts of loss-making businesses, such as the highly profitable Amazon cloud services segment of the insufficiently profitable Amazon.com Inc.
17 Simmler and Devereux note that “sectoral composition of companies subject to pillar 1 is strongly affected by the definition of profitability — pre-tax profits as a proportion of revenues. Among European firms with revenues above $20 billion, there are almost twice as many companies that have a return on equity above 10 percent compared to those that have a return on revenue above 10 percent.”
18 Likewise, any country with a minimum tax would want other countries also to have them in order to counteract the disincentive that minimum taxes create for new domestic incorporations.
19 To the extent that pillar 1 reaches location-specific rents, it should not discourage companies from, for example, entering new jurisdictions or soliciting users or consumers in a particular place. Minimum taxation, by contrast, would counteract both tax incentives to move tangible investments and tax incentives to move paper profits.
20 According to the July 1 statement, “The GloBE rules will provide for a formulaic substance carve-out that will exclude an amount of income that is at least 5 percent (in the transition period of 5 years, at least 7.5 percent) of the carrying value of tangible assets and payroll.”
21 Charlotte McFaddin, “Evaluating the Tax Veto in a Digital Age: Legislative Efficiency and National Sovereignty in the European Union,” 62 Va. J. Int’l L. 2022 (Mar. 2021).
22 Another reason to advocate for wide adoption is that commentators usually assume that the Court of Justice of the European Union will give more deference to EU-wide antiabuse rules or to OECD-agreed antiabuse rules than to unilateral antiabuse rules, which have faced stringent scrutiny under a standard that permits such rules to forbid only “wholly artificial arrangements.” Cadbury Schweppes PLC v. Commissioners of Inland Revenue, C-196/04 (CJEU 2006).
23 Inclusive framework members not signing on include Barbados, Estonia, Hungary, Ireland, Kenya, Nigeria, and Sri Lanka.
24 Oxfam International, “OECD Inclusive Framework Agrees Two-Pronged Tax Reform and 15 Percent Global Minimum Tax: Oxfam Reaction,” July 1, 2021 (describing the proposal as a “raw deal” for developing countries). The Global Alliance for Tax Justice criticized pillar 1 for keeping too much revenue in the hands of rich states and pillar 2 for suggesting a minimum tax rate that was too low. The Independent Commission for the Reform of International Corporate Taxation expressed its desire for transition to a purely formulary allocation approach. Independent Commission for the Reform of International Corporate Taxation, “G20/OECD Inclusive Framework Tax Deal: A Missed Opportunity,” July 1, 2021 (“This agreement only serves the interests of a handful of countries, the richest.”).
25 Simmler and Devereux estimate that amount A to be allocated by pillar 1 would be $87 billion, if set at the lowest proposed proportion (20 percent), supra note 13. Richard S. Collier and Devereux heralded the nexus rule as a significant change. Collier and Devereux, “On Why It Really Is Such a Big Deal,” Oxford Business blog, July 2, 2021 (“a new partial allocation of revenues by reference to the market or demand side (rather than as exclusively based on the supply or production side as under the existing system) are the first serious multilateral steps in a paradigm shift relating to the global income allocation system”).
26 The statement contemplates a two-tiered rule: If a country has a GDP of less than €40 billion, then a nonresident company will have pillar 1 nexus if it has €250,000 in revenue there. In contrast, if the country’s GDP is more than €40 billion, the nexus threshold is set to €1 million.
27 Simmler and Devereux estimate that amount A to be allocated by pillar 1 would $87 billion, if set at the lowest redistributive (20 percent) proportion. Of this, they estimated $56 billion would be generated by U.S.-headquartered companies, including $28 billion from the largest five U.S. technology U.S. companies: Apple, Microsoft, Alphabet, Intel, and Facebook. The share of the to-be-reallocated amount A associated with Chinese companies is about 10 percent (China is the second-most affected country after the United States). Id.
28 For more on these topics, see Mason, supra note 1, at 389-393 (arguing that an absence of agreed efficiency and justice norms in international tax made it likely that pillar 1 would be resolved politically, rather than by reference to high-minded ideals).
29 Even the CJEU got into the action, rendering what can almost be described as advisory opinions on the legality under EU law of digital taxes as part of two cases challenging an unrelated Hungarian turnover tax. Despite their obvious protectionism, the CJEU gave every indication that it would approve unilateral digital taxes. See Mason and Parada, “The Legality of Digital Services Taxes Under EU Law,” 40 Va. Tax Rev. 175 (2020) (discussing the controversy, analyzing the precedent relevant to company-size discrimination, and analyzing the CJEU’s decisions in Tesco Global and Vodafone).
30 For more on fiscal fail-safes, see Mason, supra note 1, at 376-380. For criticism, see id. at 385-388. See also Parada, “Full Taxation: The Single Tax Emperor’s New Clothes,” 24 Fla. Tax Rev. (forthcoming 2021) (criticizing full taxation as unprincipled).
31 Again, the substance carveout from pillar 2 could help assuage concerns of countries in group 3.
32 For arguments that the states would be well served by a multilateral implementing treaty, see Mary C. Bennett, “Contemplating a Multilateral Convention to Implement OECD Pillars 1 and 2,” Tax Notes Int’l, June 14, 2021, p. 1453.