Pillar 2 as a De Facto New Revenue Allocation Mechanism

Takayuki Nagato

Takayuki Nagato is a professor of law at Gakushuin University in Japan. The author thanks Allison Christians for valuable comments on the preliminary draft.

The research for this article was partly conducted during the author’s stay at the New York University School of Law as a Hauser Global Research Fellow in 2019-2020. This research was supported by JSPS KAKENHI Grant Number JP18K12634 and JP22K01162.

In this article, Nagato argues that the OECD’s pillar 2 is a de facto new mechanism for allocating taxing rights on the deemed excess returns of the largest multinational enterprises, and that pillar 2 without pillar 1 would unjustifiably provide low-tax investment hubs, inaccurately classified as “source” jurisdictions under transfer pricing rules with the opportunity to maintain a relatively competitive tax environment while raising additional tax revenue from international cooperation to set a corporate tax competition floor.

I. Introduction

Following the OECD/G-20 inclusive framework’s October 2021 agreement on a two-pillar solution within the base erosion and profit-shifting project,¹ BEPS 2.0 has entered its implementation phase. According to the conventional narrative, BEPS 2.0 pillar 1 intends to reallocate taxing rights between source and residence jurisdictions to “address the tax challenges arising from the digitalisation of the economy.” Pillar 2 intends to address the remaining BEPS issues.²

However even after the 2021 agreement, the likelihood of pillar 1 producing a multilateral tax convention remains questionable because of the expected lack of participation by the United States. U.S. ratification of the multilateral convention is expected to be a challenge because of the U.S. Senate’s negative attitude toward it.³ In contrast, pillar 2 is more feasible because it is implemented unilaterally through domestic legislation. There is already a series of national legislative movements in important jurisdictions, with the exception of the United States, to implement pillar 2.⁴

This study challenges the stated purposes of the two pillars. Each pillar creates allocation rules for taxing the returns deemed excessive of the largest multinational enterprises.⁵ The debate over pillar 2’s ordering priority rules reflects conflict among jurisdictions over taxing rights on deemed excess returns, undertaxed to date. Nevertheless, the project aims to avoid tampering with the traditional allocation of taxing rights.

The study also attempts to challenge the conventional academic narrative on the power balance between jurisdictions in the international tax policymaking arena. The literature focuses on the unfair relationship between developed and developing countries. This study not only focuses on this relationship, but also sheds light on the relationship between small and large developed countries. I argue that international cooperation on pillar 2 without implementing pillar 1 would unjustifiably provide low-tax investment hubs, inaccurately classified as source jurisdictions under today’s problematic transfer pricing rules, with the opportunity to maintain a relatively competitive tax environment while raising additional tax revenue from international cooperation to set a corporate tax competition floor.

To achieve this study’s goals, I will look back on the development of the two-pillar solution, from the BEPS action 1 final report published in October 2015 to the publication of administrative guidance in July.⁶ This study separates the development into three periods:

• before the Biden administration;

• from the start of the Biden administration until the inclusive framework agreement in October 2021; and

• after the 2021 agreement.

This classification reveals which group of jurisdictions is expected to enjoy favorable treatment in pillar 2.

The remainder of this article is organized as follows:

• Section II chronologically summarizes the development of the two-pillar solution, focusing on the distributive aspects among jurisdictions;

• Section III analyzes who will benefit from the GLOBE model rules and the administrative guidance to reveal that investment hubs are the winners of the deal, notwithstanding the continued criticism that pillar 2 will be the most favorable to the largest developed countries, home to many MNEs; and

• Section IV concludes.

II. Development of the Two-Pillar Solution

A. Before the Biden Administration

1. The Genesis of the Reallocation of Taxing Rights

The original BEPS project was not intended to change the framework for allocating taxing rights, which originated in the 1920s.⁷ This is despite the fact that from the beginning of the project the question of whether BEPS problems can be solved without getting into the more fundamental question of the appropriate allocation of tax bases had been raised.⁸

The Task Force on the Digital Economy did not make a specific recommendation to introduce a new tax nexus of significant economic presence for the digital economy in the BEPS action 1 final report published October 2015.⁹ Rather, it highlighted the difficulty of ring-fencing the digital economy for tax purposes.¹⁰ At the same time, it was agreed that the task force would continue its work on taxation of the digital economy to report by 2020.¹¹ BEPS 1.0 did not bring stability into the landscape of taxation of large MNEs partly because it did not prohibit countries from taking unilateral measures, provided they respected existing treaty obligations.¹²

Countries’ unilateral measures include the diverted profits tax in the United Kingdom, the multinational antiavoidance law, the diverted profits tax in Australia, and the equalization levy in India. In September 2017 the European Commission proposed taking collective action at the EU level to achieve fair taxation of large digital companies.¹³ This is important in terms of not only developing countries but also developed countries in the EU calling for reconsideration of the allocation of taxing rights, although the European Commission’s solution was intended to be confined to the digital economy.¹⁴ In March 2018 the European Commission proposed two directives, both of which were not enacted because of a failure to obtain a unanimous consensus. One is a proposal for a council directive to establish a taxable nexus for digital businesses as a long-term solution,¹⁵ and the other is a proposal for a council directive to introduce common digital services taxes in the EU single market as a short-term interim solution.¹⁶ While the European DSTs proposal was finally abandoned in March 2019,¹⁷ several EU member countries, including France, the United Kingdom, Spain, and Italy, unilaterally enacted national DSTs.

The OECD’s 2018 interim report, “Tax Challenges Arising From Digitalisation,” classifies inclusive framework members into three groups.¹⁸ The first sees a misalignment between the place of value creation and the place of taxation. It is limited to highly digitalized business models relying on user participation. This group believes that the need to reconsider the allocation of taxing rights is limited to digital businesses while supporting the existing allocation of taxing rights for other businesses.

In contrast, the second group sees the digitalization of the economy as posing a challenge to the effectiveness of the existing international taxation framework and looks to reconsider the criteria for allocating taxing rights more generally, not only for digitalized business models.

The third group finds that the BEPS action plan largely addresses double nontaxation, and does not need to reconsider the framework of international tax rules. Under this grouping, the EU’s major powers, as the first group of countries, have shown their support for reconsidering, even though it was limited to digital businesses, the allocation of taxing rights, leading to pillar 1.

2. The Genesis of the Global Minimum Taxation

The idea of global minimum taxation in pillar 2 originated from the declared positions of Germany and France on the minimum taxation of profits in the Common Corporate Tax Base Directive.¹⁹ The two countries proposed the idea of global minimum taxation at the G-7 meeting in June 2018. This new proposal was supported by the G-7 countries, except for the United Kingdom.²⁰ The United States welcomed the proposal because it had already introduced the global intangible low-taxed income rules and the base erosion and antiabuse tax rules in the Tax Cuts and Jobs Act.²¹ Then, at the urging of Germany and France,²² the idea of a global minimum tax was introduced to the G-20 in December 2018. Pascal Saint-Amans, then-director of the OECD Centre for Tax Policy and Administration, thought it was difficult to connect this to the OECD mandate of addressing tax challenges from the digitalization of the economy.²³ The result was a distinct new topic for the international tax reform project approved by the inclusive framework in January 2019.²⁴

3. Linkage of Two Pillars

The policy note published in January 2019 was the first public announcement of the two-pillar solution;²⁵ and the Programme of Work, published in May 2019 after the public consultation²⁶ process, detailed the architecture of the international tax reform plan.²⁷ It was explained that pillar 1 “addresses the broader challenges of the digitalised economy and focuses on the allocation of taxing rights,” and pillar 2 “addresses remaining BEPS issues.”²⁸ The proposal in pillar 2 was named the global anti-base-erosion (GLOBE) proposal.²⁹

The Programme of Work detailed three proposals under pillar 1:

• the user participation proposal;

• the marketing intangibles proposal; and

• the significant economic presence proposal.³⁰

The user participation proposal was suggested by the United Kingdom. The U.K. strategy is to extend its taxing rights against U.S. digital companies while resisting a reform proposal to allow a more general move toward destination-based taxation via the value creation principle in the BEPS project.³¹

The marketing intangibles proposal was from the United States to avoid ring-fencing digital businesses. It was epoch-making for the United States to propose a plan to reallocate taxing rights rather than just resist the wave of DSTs in other countries.³²

The significant economic presence proposal came from India and was supported by other developing countries whose influence in reforming the international tax regime is increasing in the G-20 and the inclusive framework.³³

Even though pillar 1 was revolutionary because its objective was to reconsider the allocation of taxing rights, the two-pillar solution was soon criticized as a diversion from the BEPS action 1 final and interim reports’ focus on expanding source taxation, to protecting residence taxation in pillar 2. The leap was seen as influenced by the richest developed OECD countries.³⁴

The OECD secretariat developed a unified approach under pillar 1 by reconciling three separate ideas. The secretariat then proposed it for public consultation in October 2019.³⁵ The unified approach:

• covers highly digital business models but goes wider — broadly focusing on consumer-facing businesses;

• creates a new nexus, not dependent on physical presence but largely based on sales;

• uses a new profit allocation rule that goes beyond the arm’s-length principal; and

• increases tax certainty.³⁶

Under the unified approach, a share of deemed residual profit (amount A) will be allocated to market jurisdictions as a new taxing right. According to the preliminary findings of an economic impact assessment:

on average, low and middle-income economies would gain from pillar 1, experiencing a higher rate of increase in revenues than high-income economies even though, larger market jurisdictions will benefit more in absolute.³⁷

On the other hand:

Investment hubs, where the analysis suggests that levels of residual profit are high, would experience significant losses in tax base.³⁸

Small European countries, including Ireland, were therefore dissatisfied with the unified approach,³⁹ even though they did not officially express opposition.⁴⁰

The unified approach had inherent difficulties from the outset: It would require a revision of tax treaties; and it was unlikely to find support among more than two-thirds of the U.S. Senate, even though Steven Mnuchin, the Trump administration Treasury secretary, was supportive. Just before the G-20 meeting in December 2019, the United States announced that it supported the unified approach as a “safe harbor,” optional for corporations, while it fully supports a pillar 2 solution.⁴¹

The U.S. safe harbor proposal raised doubts about the feasibility of pillar 1. There was fear that the process might increase tax revenue of large residence countries through pillar 2 without pillar 1 augmenting it for source developing countries.⁴² Concern was heightened by a June 12, 2020, letter from Mnuchin to the finance ministers of several European countries that already had DSTs. The letter suggested delaying work on pillar 1 until the end of 2020 because of the COVID-19 pandemic. It added that the United States would welcome continuation of work on pillar 2.⁴³

France, Italy, Spain, and the United Kingdom wrote a June 17, 2020, joint response letter, saying that “postponing our work (on digital taxation) and not addressing these challenges would constitute a collective failure.”⁴⁴ In a June 18, 2020, statement, OECD Secretary-General José Ángel Gurría urged members of the inclusive framework to remain engaged in negotiations and meet the goal of delivering a global solution by the end of the year.⁴⁵ However, it was clear that pillar 1 was already on the brink of collapse when the blueprints of both pillars were published in October 2020.⁴⁶

This was disappointing for developing countries anticipating better revenue allocation. According to the OECD’s economic impact assessment published in October 2020, on average, low-, middle-, and high-income jurisdiction groups would all benefit from modest revenue gains (global revenue gain is projected to be 0.2 to 0.5 percent of global corporate income tax revenues).⁴⁷ Investment hubs⁴⁸ would significantly lose tax revenues under pillar 1.⁴⁹

While it was estimated that pillar 2 would yield a significant increase in corporate income tax revenues (1.7 to 2.8 percent of global corporate tax revenues) across low-, middle-, and high-income jurisdictions, the revenue gain was estimated to be relatively larger among high-income jurisdictions because gains from the income inclusion rule were estimated to accrue to the jurisdiction of the MNE group’s ultimate parent, generally high-income jurisdictions.⁵⁰ Without pillar 1, the revenue gains from the global tax deal will accrue to powerful OECD countries, as Yariv Brauner has insisted since the outset of BEPS 2.0 negotiation. It is therefore argued that the international ramifications of the designs of the two pillars should be jointly addressed within the framework of a single multilateral convention.⁵¹ Against this background, developing countries have sought their own ways to tax digital companies by proposing the creation of new taxing rights via the addition of article 12B of the U.N. model tax convention. This allows source jurisdictions to impose withholding taxes on the gross amount of payments to automated digital service enterprises.⁵²

B. The Biden Administration: Pillar 2 Centric Proposal

The transition from the Trump administration to the Biden administration brought significant change in the prospects for consensus-making in the inclusive framework on the two-pillar solution. The Biden administration appointed Janet Yellen Treasury secretary and Itai Grinberg Treasury deputy assistant secretary for multilateral tax to assume the role of negotiators in the global tax deal.

Yellen withdrew Mnuchin’s safe harbor proposal in February 2021 and returned to the international negotiating table.⁵³ Next, the United States shifted priority between pillar 1 and pillar 2 by presenting its plan on the two-pillar solution to the inclusive framework steering group on April 8, 2021.⁵⁴ The U.S. plan put pillar 2 in the center of the global tax deal to end the race to the bottom over multinational corporate taxation.

Pillar 2 was useful in avoiding putting U.S. MNEs at a competitive disadvantage under the U.S. Made in America Tax Plan to increase corporate tax revenue by raising the statutory corporate tax rate to 28 percent and the GILTI effective tax rate to 21 percent.⁵⁵ Pillar 1 is positioned to stabilize a multilateral international tax architecture to address the proliferation of unilateral measures.⁵⁶ The United States proposed a simplification of pillar 1 because it was not expected that the complexities and wide scope of the pillar 1 blueprint would be approved by the Senate. The simplification plan limits the number of in-scope MNEs to no more than 100 without materially reducing the quantum of profits available for reallocation using quantitative criteria of revenues and profit margins.⁵⁷

Although the Biden administration revived the hope of reaching a consensus on the global tax deal, it was not what inclusive framework member countries, except large powerful developed countries, expected. Developing countries were disappointed with the U.S. pillar-2-centric proposal.⁵⁸ Eastern European countries, including Hungary and Poland, disliked the minimum tax because they used tax preferences to attract factories from Germany and France.⁵⁹ Ireland announced that it no longer objected to pillar 2, but would never tolerate the minimum tax rate set above 12.5 percent.⁶⁰

The United States, however, proposed on May 20, 2021, to the steering group of the inclusive framework that the global minimum tax rate should be at least 15 percent, notwithstanding the announcement of Ireland.⁶¹ The proposal was welcomed by Germany and France, which had been going to accept even 12.5 percent, but Eastern European countries and China⁶² were not happy about the rate being higher than their expectations.⁶³ Developing countries were not satisfied with the proposal because they thought the 15 percent minimum rate was too low and feared that the revenue would go to the rich home jurisdictions of the MNEs.⁶⁴ Also, developing countries were divided over the desirability of the influence of global minimum tax on their preferential tax regimes.⁶⁵ As for pillar 1, developing countries required at least 20 percent of the deemed residual profit of MNEs to be reallocated to consumer countries to increase revenue.⁶⁶

It is certain that the U.S. proposal in May paved the way for the G-7 agreement in June⁶⁷ and the inclusive framework and G-20 meeting agreements in July,⁶⁸ although the G-7 agreement was criticized for its minimum rate being too low and lacking consideration for developing countries.⁶⁹ Following negotiations to determine concrete numbers, the G-20 agreement was finally reached in October 2021, confirming that the global minimum tax rate is 15 percent and amount A is 25 percent of the deemed residual profit.⁷⁰ Ireland, which did not join the inclusive framework agreement in July, finally joined the G-20 agreement in October after careful deliberation to retain its headline corporate tax rate of 12.5 percent for companies outside the scope of pillar 2, while the 15 percent rate applies to in-scope MNEs.⁷¹

The pillar-2-centric global tax deal, accompanied by the reduced scope of pillar 1, was harshly criticized from the revenue allocation perspective.⁷² It is clear from this criticism that many jurisdictions, including investment hubs, Eastern European countries, and other developing countries, are dissatisfied with the revenue allocation pushed by the G-7 countries, home to large MNEs.

C. Source Jurisdictions Strike Back

1. QDMTT Late to the GLOBE Rules

The global tax deal story did not end with the 2021 agreement. The GLOBE model rules published in December 2021⁷³ surprised tax experts worldwide. The model rules contained a new tax instrument that had been only slightly discussed before⁷⁴: the qualified domestic minimum top-up tax (QDMTT) in article 10.1 of the model rules.⁷⁵ The QDMTT functions roughly as a domestic minimum tax that:

operates to increase domestic tax liability with respect to domestic Excess Profits to the Minimum Rate for the jurisdiction and Constituent Entities for a Fiscal Year.⁷⁶

The QDMTT is a domestic tax but is excluded from the definition of covered tax⁷⁷ in calculating jurisdictional ETR.⁷⁸ Instead, a QDMTT is creditable against the IIR and the UTPR in calculating the amount of jurisdictional top-up tax.⁷⁹ This means that the QDMTT has the highest priority in taxing excess profits in the GLOBE model rules. If jurisdictions introduce the QDMTT, they can soak up tax revenue from undertaxed excess profits ahead of the application of the IIR by the jurisdictions of ultimate parent entities (UPEs) and the UTPR in other jurisdictions. Therefore, the addition of the QDMTT has completely altered the rule order of pillar 2, and the QDMTT would greatly increase the revenue gains of source jurisdictions in pillar 2,⁸⁰ in contrast to the previous criticism of pillar 2’s revenue allocation mechanism, which would mainly benefit residence jurisdictions.

The background of the addition of the QDMTT to the GLOBE model rules is a complete mystery for outsiders.⁸¹ As Michael Devereux and John Vella astutely point out, if it was discussed and agreed upon before, then it is odd that the QDMTT was not mentioned in the inclusive framework statement in October 2021.⁸² On the other hand, if an agreement on the QDMTT was not reached before October 2021, its addition significantly alters pillar 2’s incentives and distributional consequences.⁸³ Considering the QDMTT’s significant revenue redistribution role to source jurisdictions,⁸⁴ it is possible to speculate that the QDMTT was inserted because the majority of the inclusive framework member countries other than the G-7 countries requested it and the G-7 countries conceded to those countries under the uncertain U.S. political situation.⁸⁵ If this speculation is true, the traditional problem that rich OECD countries are imposing standards that favor them in the name of inclusivity⁸⁶ was partly overcome, at least on this point, even though it may be insufficient.⁸⁷ Therefore, accusation of the unfairness of the global minimum tax without considering the effect of the QDMTT⁸⁸ is too simplistic.

2. QDMTT Ordering Rules and CFC Rules

The effect of the QDMTT on the ordering rules in the GLOBE model rules extends not only to the application of the IIR but also to the application of the controlled foreign corporation tax regime.⁸⁹ Under the model rules, covered tax includes the UPE jurisdiction’s CFC tax.⁹⁰ The CFC tax is allocated to the constituent entity to calculate its ETR in the subsidiary’s jurisdiction.⁹¹ Therefore, a QDMTT applies a top-up tax on a post-CFC tax regime basis to increase the domestic tax liability on excess profits in the jurisdiction to the minimum rate. To determine the exact amount of a QDMTT, it is reasonable to interpret that the QDMTT is not creditable to CFC tax.⁹² If the QDMTT is creditable, a circular issue is created when calculating the jurisdictional ETR.⁹³ Commentary to the GLOBE model rules published in March 2022 mentions “domestic tax regimes should not provide credit for any tax imposed under a Qualified UTPR or IIR” to avoid a circularity problem,⁹⁴ but remains silent on the creditability of a QDMTT. Accordingly, Heydon Wardell-Burrus argues for the contrary interpretation — that a QDMTT is creditable to grant source jurisdictions priority over UPE jurisdictions.⁹⁵ Brian Arnold also supports the idea that residence jurisdictions should grant foreign tax credit for source jurisdictions’ QDMTT.⁹⁶

The argument that a residence jurisdiction should grant credit for a QDMTT in applying CFC rules is understandable considering the political situation of the United States at that time. The Biden administration failed to enact the Build Back Better Act, which included provisions to reform the GILTI regime into a qualified IIR at the end of 2021. If the GILTI regime is not updated to a qualified IIR, it is considered a “controlled foreign company tax regime” in the model rules, superseding the priority given to the United States,⁹⁷ notwithstanding its broken promises.⁹⁸

Wardell-Burrus is afraid of creating unintended incentives for countries to expand CFC rules if a QDMTT is not creditable.⁹⁹ Also, he argues for the primacy of the QDMTT based on the formula for calculating the top-up tax amount in the model rules. The QDMTT is imposed on excess profits calculated by subtracting the substance-based income exclusion (SBIE) from net GLOBE income for the jurisdiction.¹⁰⁰ According to Wardell-Burrus, it is odd that “country representatives bargained for the SBIE to be excluded from the top-up tax amount, only to have this benefit taken away through CFC rules which would give the relevant revenue to another jurisdiction.”¹⁰¹

Moreover, Arnold characterizes the QDMTT as:

a simpler and more certain method for countries to increase their domestic taxes on the excess profit of an MNE group to avoid giving up potential tax revenue to other countries.¹⁰²

He justifies the priority of the QDMTT for source jurisdictions by the availability of nonqualified domestic minimum tax,¹⁰³ which is categorized as covered tax, creditable against a CFC tax.¹⁰⁴

The administrative guidance on the GLOBE model rules published in February officially used the QDMTT acronym for the qualified domestic minimum top-up tax and added an explanation of the ordering rule to the commentary of the model rules. This clarifies the priority of the QDMTT over the CFC rules, including the GILTI regime,¹⁰⁵ by excluding a CFC tax from the calculation of the ETR for QDMTT purposes,¹⁰⁶ with a lenient allocation mechanism regarding a blended CFC tax for regular GLOBE ETR calculation purposes to entice the United States to allow FTCs for a QDMTT.¹⁰⁷

III. Pillar 2 Winners and Losers

A. Pillar 2 Implementation in Each Jurisdiction

Following the publication of the GLOBE model rules in December 2021, each jurisdiction started implementation into its domestic tax law.¹⁰⁸ Early adopters were South Korea and Japan, although neither officially announced its stance on the introduction of a QDMTT.¹⁰⁹ Several emerging Southeast Asian countries, including Malaysia, Thailand, and Vietnam, which are assumed to have tax incentives to attract foreign MNEs, will implement pillar 2 with a QDMTT in 2024.¹¹⁰ All EU member countries are required to enact the GLOBE rules by the end of 2023 following the EU directive on pillar 2 adopted by the EU Council in December 2022.¹¹¹ The United Kingdom and Canada, which are the remaining G-7 countries except the United States, will implement pillar 2 with a QDMTT.

Investment hubs, including Singapore, Hong Kong, Ireland, Luxembourg, the Netherlands, and Switzerland, will institute a QDMTT.¹¹² Moreover, even the U.K. crown dependencies¹¹³ and Caribbean islands¹¹⁴ are moving forward to implement pillar 2 along with a QDMTT. On the other hand, countries in Latin America and Africa are late to implement pillar 2.¹¹⁵

The trend is for countries, except for a limited number of high-tax countries, such as Japan, to adopt pillar 2 along with a QDMTT to avoid other jurisdictions collecting taxes on excess profits in their jurisdictions.¹¹⁶

B. Revenue Estimates

Countries cannot be indifferent to the effect of pillar 2 on their tax revenue. According to the latest OECD economic impact assessment (in January), the estimated annual global revenue gains from pillar 2 amounted to $220 billion for 2018, while those from pillar 1 were estimated to be only $13 billion to $36 billion for 2021 or $12 billion to $25 billion per year on average over the period 2017-2021.¹¹⁷ However, official jurisdiction-level estimates are not yet available.

The EU Tax Observatory study group estimated the revenue effects of pillar 2 without ratifying pillar 1 for each jurisdiction.¹¹⁸ Its estimates are based on the OECD’s country-by-country report statistics. They indicate that global revenue gains are more modest (€139 billion to €165 billion) than the OECD estimates.¹¹⁹ As for the distribution of revenue gains among jurisdictions, most of the revenue gains accrued to investment hubs, including Bermuda, the Cayman Islands, Hong Kong, Luxembourg, the Netherlands, Singapore, and Switzerland in its host country scenario, in which all jurisdictions are assumed to introduce a QDMTT.¹²⁰ As the study group acknowledges, however, the estimates do not incorporate behavioral changes of either MNEs or jurisdictions in response to pillar 2, and they make a strong assumption that all jurisdictions introduce a QDMTT.¹²¹ Considering the potential for MNE behavioral changes, investment hubs might lose a fair amount of cash inflow and tax base.¹²² However, the global minimum tax rate of 15 percent is too low to make it happen.¹²³

The U.S. Joint Committee on Taxation factored in behavioral changes of MNEs in estimating the revenue effect of pillar 2 on the United States.¹²⁴ The JCT’s upper boundary projection is a $224.2 billion revenue increase for the United States during the 10-year budget window of 2023-2033, based on the assumption that in-scope U.S. MNEs shift up to 75 percent of their low-taxed profits to the United States, compared with the hypothetical baseline in which no jurisdiction has enacted or plans to enact pillar 2. The lower boundary projection is a revenue decrease of $174.5 billion during the 2023-2033 period, based on the assumption that up to 75 percent of the low-taxed profits are shifted to jurisdictions with a QDMTT.¹²⁵ Because there will be significant heterogeneity in responses across U.S. MNEs, the JCT assumes a modified baseline between the upper and lower boundary scenarios.

In scenario 1 (the rest of the world¹²⁶ enacts pillar 2 in 2025, but the United States does not), the revenue decrease is projected to be $122 billion during 2023-2033. On the other hand, in scenario 2, in which both the rest of the world and the United States enact pillar 2, the revenue decrease is projected to be $56.5 billion.¹²⁷ In either scenario, the United States will lose revenue because it assumes that other jurisdictions’ QDMTTs will be credited against U.S. taxes, including the GILTI tax and the corporate alternative minimum tax on foreign income.¹²⁸

U.S. revenue would increase if the rest of the world does not enact pillar 2, while the United States enacts it (the JCT’s scenarios 4 and 5), which Kimberly Clausing, who was the lead economist for the Office of Tax Analysis at the U.S. Treasury under the Biden administration in 2021-2022, thinks is realistic.¹²⁹ But these scenarios are not realistic under the trend of widespread introduction of QDMTTs.¹³⁰ From a theoretical perspective, the fact that the vast majority of the profits of in-scope MNEs with parents in the G-7 countries can be attributed to MNEs that also have a presence in other G-7 countries,¹³¹ and that G-7 countries other than the United States have already adopted or will adopt pillar 2, provides the rest of the world with a strong incentive to adopt a QDMTT.¹³²

Although Clausing stresses that the estimates are heavily dependent on the assumptions of behavioral changes regarding profit shifting and that more low-taxed profits will shift to the United States,¹³³ it is inevitable that the widespread introduction of QDMTTs will decrease U.S. tax revenue. A German estimate indicates that the host country’s increase in the ETR greatly offsets the tax revenue gains of the UPE jurisdictions from the reduction in profit shifting.¹³⁴ The latest German revenue estimate expects that the annual revenue increase from pillar 2, including the revenue increase from the reduction in profit shifting, will be in the range of only €3.4 billion to €3.9 billion in 2026, which is equivalent to only 0.4 percent of the total annual tax revenue in 2021.¹³⁵ Moreover, the report of the estimate states that the projected revenue increase from the QDMTT, with a range of €2 billion to €2.3 billion, is likely to be overestimated.¹³⁶

C. Normative Analysis

1. The Primacy of QDMTT

Pillar 2’s promise for the global economy is that the GLOBE rules would set a floor for corporate tax competition, a goal that requires collective action to carry out.¹³⁷ Through cooperation, an MNE’s home jurisdiction can introduce minimum taxes without the fear that only MNEs headquartered in its jurisdiction will suffer from a competitive disadvantage. The home jurisdiction also increases tax revenue if its MNEs’ profit shifting is curbed by the minimum tax. Once a floor is set, host jurisdictions no longer have to provide tax incentives to attract foreign MNEs, raising tax revenue to the minimum tax level.

It seems a win-win situation if the level of corporate income tax revenue is now suboptimal worldwide. Some commentators welcome the development of pillar 2 as an accomplishment of “the Single Tax Principle”¹³⁸ or “full taxation”¹³⁹ in the global tax regime. Regarding this development, MNEs’ home jurisdictions may see the benefit of pillar 2 as ensuring an equal competitive environment for their MNEs, even though pillar 2 itself does not lead to an immediate direct increase in their tax revenue.¹⁴⁰

However, Leopoldo Parada convincingly points out the lack of a clear distributive rule in the concept of full taxation.¹⁴¹ Ruth Mason appropriately acknowledges that it is unrealistic to expect jurisdictions to pursue full taxation regardless of which country will receive tax revenue.¹⁴² In this regard, not only pillar 1 but also pillar 2 are revenue allocation mechanisms.¹⁴³ From this perspective, both early proponents and opponents of the global minimum tax clearly envisioned that tax revenue from it is collected by MNEs’ home jurisdictions.¹⁴⁴ Proponents then defend the imposition of the minimum tax by a home jurisdiction as following from its subsidiary taxing rights,¹⁴⁵ or its respect for source jurisdictions’ taxing rights carried out through FTCs.¹⁴⁶ Allison Christians and Laurens van Apeldoorn propose expansion of passthrough-type residence taxation paradoxically with carefully tailored FTCs to ensure source jurisdictions can impose source taxes without worrying about tax competition.¹⁴⁷ However, it is unclear whether these proponents predicted the emergence of the QDMTT in the GLOBE model rules.

Moreover, Joachim Englisch and Johannes Becker go further, arguing that some of the excess returns subject to the minimum tax originate from valuable intellectual property created in the UPE jurisdictions. The UPE jurisdiction’s minimum taxation therefore reinforces the principle of value creation.¹⁴⁸

Some prominent commentators defend the primacy of the QDMTT against the IIR and even against CFC taxes¹⁴⁹ to protect source countries’ primary taxing rights.¹⁵⁰ One of these commentators justifies the QDMTT as a tool for developing countries to help save legislative costs by providing a tax instrument for setting their jurisdictional ETRs at exactly 15 percent.¹⁵¹

2. Case Against the Primacy of the QDMTT

However, there are three reasons why the primacy of the QDMTT is not always defensible by the traditional dichotomy of source and residence.

First, the terminology of source is so loose that the jurisdiction categorized as source does not necessarily mean a source jurisdiction in a traditional sense in the discussion of international taxation.¹⁵² A source jurisdiction in the GLOBE context can be understood as just a jurisdiction other than the UPE jurisdiction in which MNEs’ income is booked. Even if commentators argue for the primacy of the QDMTT for the benefit of developing countries, it is highly likely that these countries cannot raise revenue from the QDMTT, the BEPS Monitoring Group and the South Centre contend.¹⁵³ This is because their tax base is built on existing transfer pricing rules,¹⁵⁴ which have numerous problems, including profit shifting.¹⁵⁵ It sounds even hypocritical to argue for the primacy of the QDMTT for developing countries¹⁵⁶ despite recognizing that the GLOBE model rules are intended to apply where profits are allocated based on the vague concept of source or value creation.¹⁵⁷ The concepts’ indeterminacy has been criticized by many commentators.¹⁵⁸ Accordingly, the plausible idea of expanding passthrough-type residence taxation as a backstop for source jurisdictions to increase source taxes without worrying about tax competition¹⁵⁹ will not be achieved by the GLOBE rules.

Not only developing countries that host foreign MNEs but also investment hubs are included in source jurisdictions in the GLOBE context. Many of them, however, are rich jurisdictions with sufficient legislative and administrative capacity to tweak and implement their own domestic tax rules. The BEPS Monitoring Group persuasively points out that the QDMTT will benefit conduit jurisdictions that have adopted tax incentives to encourage MNEs to declare high profits there, but will not benefit most developing countries considering the low global minimum ETR of 15 percent compared with the world weighted average headline corporate tax rate of 25 percent.¹⁶⁰ Moreover, despite that the QDMTT was invented for developing countries to help save legislative costs to raise the jurisdictional ETR to exactly 15 percent, there is still an argument that it may be difficult to implement in many developing countries.¹⁶¹ It is possible that only investment hubs, not developing countries in general, can enjoy tax revenue increases from QDMTT without losing foreign investment.¹⁶² Accordingly, QDMTT is not a good alternative tax instrument to pillar 1, even if its sudden insertion in the GLOBE model rules was implicitly intended to reallocate taxing rights on the excess returns of MNEs.

Second, the compliance costs of pillar 2 are borne by in-scope MNEs,¹⁶³ including those that are not engaged in aggressive tax planning such as Japanese MNEs.¹⁶⁴ In-scope MNEs are required to calculate ETR for QDMTT and other GLOBE rules separately because of the administrative guidance published in February, which aimed to calculate the amount of QDMTT before CFC tax. It is unclear how effectively the QDMTT safe harbor allowed in the administrative guidance published in July¹⁶⁵ will alleviate the compliance burden of in-scope MNEs. It is unfair that these in-scope MNEs are forced to incur a heavy compliance burden for revenue increases in other countries.

Third, QDMTT should not be prioritized over CFC rules in certain cases. It is understandable that the U.S. GILTI regime, which the Biden administration failed to reform to a qualified IIR, should not be prioritized over the QDMTT.¹⁶⁶ However, the CFC rules in many jurisdictions are anti-tax-avoidance measures to counter tax avoidance achieved by routing investments through low-tax jurisdictions.¹⁶⁷ If the QDMTT in low-tax jurisdictions always trumps the CFC rules of residence jurisdictions, the tax base of the UPE jurisdictions is materially eroded¹⁶⁸ by the global tax deal. This consequence contradicts the OECD’s traditional position that CFC rules are legitimate instruments to protect the domestic tax base and do not conflict with tax conventions.¹⁶⁹ Further, after the adoption of the anti-tax-avoidance directive (ATAD) (2016/1164) in the EU, it is even possible to say that a CFC tax, at least as a supplement to transfer pricing rules,¹⁷⁰ is now considered to be an essential element of the legislative arsenal used to combat tax avoidance in the OECD, the EU, and other related jurisdictions.¹⁷¹

IV. Consequence of International Cooperation

Ironically, pillar 2, originally purposed to address remaining BEPS issues, will result in keeping the profit-shifting incentive intact because of its low minimum rate. Moreover, pillar 2, as the second-best alternative to pillar 1 for reallocating taxing rights on the excess returns of MNEs, cannot achieve its goal because of its dependence on flawed transfer pricing rules. Instead, it provides rich low-tax investment hubs with a great opportunity to raise additional corporate tax revenue¹⁷² without losing their competitive tax environment. Investment hubs that attract substantial economic activities are unlikely to lose their competitive tax environment because of the SBIE¹⁷³ and qualified refundable tax credits¹⁷⁴ in the GLOBE model rules.

This is a good deal for investment hubs considering the economic effects of both pillars. Originally, it was even possible in their worst scenario that they would lose a lot of tax revenue from pillar 1,¹⁷⁵ as well as their competitive tax advantage via a higher global minimum tax. After much twisting and turning of the negotiations, pillar 1 will go away thanks to the United States, and pillar 2 will increase these hubs’ tax revenue by curtailing tax competition without adding burdening costs,¹⁷⁶ although they have long been the greatest beneficiaries of international tax competition, colluding with tax-aggressive MNEs. Is this a project worth pursuing globally, incurring significant negotiation, legislative, administrative, and compliance costs?