OECD Races Toward Completing Final Report on Digital Economy
But two fundamental “pillars” present structural challenges

print this article
illustration of people building a bridge

The Organisation for Economic Co-operation and Development (OECD) is racing toward meeting an ambitious target by the end of the year that could radically change the way all multinational enterprises (MNEs) are taxed. The target—a final report—was set forth in January 2019 under the cover of addressing the digital economy.1

However, the two fundamental “pillars” of this project are far broader than that. As the OECD’s base erosion and profit shifting (BEPS) project concluded that it is impossible to ring-fence the digital economy, the OECD is now setting the foundation of the new international tax system that reaches far beyond the digital economy with a global anti-base erosion (GloBE) framework, known as Pillar Two.2 So, what will happen at the end of this sprint to the finish? It may be meaningful to look at the current state of Pillar Two, and the developments under Pillar One, before looking at our future international tax system.

Because much of the attention in 2020 has focused on Pillar One, the pillar that initially addresses digital taxes, many MNEs have taken the view that this 2020 work stream is irrelevant to their business. This view may seem understandable if one’s business is not primarily digital or consumer facing. However, it misses a critical point: that key elements of both Pillar One and the global minimum tax framework/GloBE will affect all MNEs, and further input will be key to whatever consensus-based solution will be produced for Pillar Two.

The Sprint

During its May 2019 meeting, the Inclusive Framework adopted a Programme of Work, which set forth the design of two pillars for completion by the end of 2020.3 These two pillars were designed to provide a consensus solution to address the challenges of taxation in the digital economy under Pillar One and to address what the Inclusive Framework called the challenges in the remaining issues under BEPS under Pillar Two.4 The proposals in Pillar One address nexus and transfer pricing, whereas the proposal under Pillar Two is an exploration of “strengthening the ability of jurisdictions to tax profits where the other jurisdiction with taxing rights applies a low effective rate of tax to those profits.”5

In the fall of 2019, the OECD and the Inclusive Framework published more detailed consultation documents regarding Pillars One and Two in their proposal for a “unified approach” under Pillar One6 and GloBE under Pillar Two.7 Commentators had an opportunity to provide written and oral comments at the consultations, which amounted to two days of consultation for Pillar One and one day for Pillar Two. In January 2020, the OECD released a statement on its progress to date that contained additional details, mostly about Pillar One, as well as a timeline to achieve key policy features of a consensus-based solution to Pillar One by July 2020 and to produce a final report on the technical details of the Inclusive Framework’s consensus-based solution by the end of 2020.8 The progress note on Pillar Two was meaningfully limited to the last annex, which briefly discussed the key design work streams on the constituent pieces that build up the GloBE proposal.

State of Pillar Two

Overview of GloBE

On January 31, 2020, the OECD conducted Tax Talks #14 to discuss the Inclusive Framework’s statement concerning the status of progress toward implementing the unified approach for Pillar One and the status of GloBE under Pillar Two. In this statement, the members of the Inclusive Framework affirmed their commitment to reach an agreement on a consensus-based solution by the end of 2020, with a focus on improving tax certainty. Furthermore, the statement detailed that significant work remains in each of the four sections of GloBE.

As for the effort needed to develop the two pillars, GloBE is surprisingly underdeveloped in light of the OECD’s 2020 target. GloBE is a major sea change in the approach that governments will need to take when contemplating corporate income taxation. Unlike Pillar One, which is clearly tied to the first BEPS Action Item,9 Pillar Two exists as a wholly separate approach that fundamentally creates a new highly interconnected international minimum corporate income taxation system. The four components of the GloBE proposal are:10

  1. an income-inclusion rule that would tax the income of a foreign branch or a controlled entity if that income was subject to tax at an effective rate that is below a minimum rate;
  2. an undertaxed-payments rule that would operate by way of a denial of a deduction or imposition of source-based taxation (including withholding tax) for a payment to a related party if that payment was not subject to tax at or above a minimum rate;
  3. a switchover rule to be introduced into tax treaties that would permit a residence jurisdiction to switch from an exemption to a credit method where the profits attributable to a permanent establishment (PE) or derived from immovable property (which is not part of a PE) are subject to an effective rate below the minimum rate; and
  4. a subject-to-tax rule that would complement the undertaxed-payments rule by subjecting a payment to withholding or other taxes at the source and adjusting eligibility for treaty benefits on certain items of income where the payment is not subject to tax at a minimum rate.

OECD Progress Report—January 31, 2020

During the OECD’s January 31, 2020, Tax Talks, Pascal Saint-Amans, director of the Centre for Tax Policy and Administration at the OECD, admitted the timeline was highly ambitious.11 Annex Two in the statement indicated that, whereas Pillar One had numerous competing proposals with respect to the elements needed to be completed in the unified approach, the four individual elements under Pillar Two had already been identified and agreed on in the Programme of Work. Saint-Amans reaffirmed the OECD’s end-of-2020 commitment to have a final report on the technical details of Pillar Two, with several key policy features to be discussed at the OECD plenary meeting on July 1 and 2.

Annex Two briefly summarized the progress on the elements of GloBE with a focus on the first of the elements as it related to the income-inclusion rule. The approach for this element pulls from controlled foreign company (CFC) rules to require shareholders to account for a proportionate share of a corporation’s income at a minimum top-up rate. The statement indicates that “extensive work” is underway to evaluate the use of financial accounts as the basis for this determination, including exploring the use of principle-based criteria to identify needed adjustments to financial accounts, the appropriateness of blending low- and high-tax jurisdictions, and carve-outs.12

The remaining three elements rely, in large part, on the design of the income inclusion rule, though simplicity is a clear priority. The second element listed in Annex Two covered the switchover rule; the third element covered the undertaxed-payments rule; and the fourth covered the subject-to-tax rule. The statement noted that several proposals for the undertaxed-payments rule were under review with a focus on limiting complexity and the costs of compliance and administration.13 Looking at the statement through the lens of what was said during the Tax Talks, the OECD is clearly in the early design phase for GloBE and focusing on tax simplicity. Saint-Amans also emphasized that the OECD welcomes technical design elements for each of the four elements,14 and the statement mentioned that a special subgroup on financial accounts has been formed under Working Party 11 to bring together experts with technical tax and financial accounting expertise.15 Additionally, the statement highlighted that coordination with other international obligations was being evaluated, including European Union/European Economic Area law and tax treaty nondiscrimination obligations as well as the rules’ interaction with the broader international framework.16

OECD Economic Progress Report—February 13, 2020

During a webcast dated February 13, 2020, the OECD provided some promising yet very preliminary data in its economic progress report, showing estimates of up to $100 billion in additional tax revenues globally created by the combination of Pillars One and Two.17

In the presentation, David Bradbury, head of tax policy and statistics at the OECD’s Centre for Tax Policy and Administration, mentioned that some have critiqued the OECD for analyzing the impact of Pillars One and Two too late, whereas others have indicated that this impact analysis was done too early. To an extent, both views are right.

On the side of being too late, this analysis to inform members of key design decisions for Pillars One and Two was presented after the Inclusive Framework’s decision to proceed with the two-pillar approach as a way to reach a consensus-based solution in 2020. On the other hand, local country anti-avoidance measures, including anti-hybrid and economic substance legislation introduced during the BEPS project, were already changing the way MNEs operate, with results not yet fully captured in the economic impact assessment data, which establishes only a snapshot of post-BEPS midstream changes rather than a baseline.

With all of these changes, the economic progress report is also arguably too early. On top of digital services taxes, diverted-profits taxes and Actions 8 through 10 concerning the treatment of intangibles have also encouraged companies to reconsider their value chains. For companies with substantial U.S. operations, the 2017 U.S. Tax Cuts and Jobs Act (TCJA) created a global system of taxation that, with its foreign-derived intangible income (FDII) regime, has made the United States more attractive as a product-export hub and, with its global intangible low-taxed income (GILTI) and base erosion and anti-abuse tax (BEAT) regimes, low-tax international countries relatively less attractive.

One concern is that the OECD might base its decisions on data from 2016 country-by-country (CbC) reporting, before numerous economic substance and anti-hybrid rules came into effect and before the TCJA went into effect. Various countries, including the United States, have already adopted many of these anti-abuse measures, including GILTI and the domestic base erosion anti-avoidance systems of BEAT. And companies have already reacted to these law changes by restructuring. Questions may remain as to whether the TCJA regulations will stand up to eventual World Trade Organization scrutiny, but it is clear that many MNEs have restructured since 2016. Because of that, it is also clear that the incremental taxes from GloBE will not be achieved as currently sketched out by the OECD in its economic impact analysis.

Building on Pillar One

Practically speaking, assuming no further input from taxpayers, the OECD will lean on those policies it has at hand to help design Pillar Two. This likely consists of learning points from the BEPS projects, specifically Pillar One, and potentially from other large countries with global taxing systems, such as the United States. The state of Pillar Two suggests much work remains to be done, including the resolution of difficult technical issues. Because both involve global allocation issues, in particular how income or tax is surrendered and the multilateral nature of resolution mechanisms, the work to date under Pillar One’s unified approach is a meaningful point to reference as we look toward the design of Pillar Two.

Overview of the Unified Approach

In principle, Pillar One is an effort to address perceived flaws in taxing nexus and the arm’s-length standard relating to the challenges in taxing digital services. The unified approach under Pillar One creates three amounts with different nexus requirements. Amount A is the “supra return,” or return on intellectual property. Amount B is the routine return for distribution. Amount C is for other services performed in a jurisdiction.

Amount A departs significantly from current law because it would impose a version of formulary apportionment on the supra returns and would expand nexus beyond physical presence. Amount A would apply to automated digital services and consumer-facing businesses. The definition of automated digital services is broad and includes online advertising, online platforms, and cloud services. The latest Inclusive Framework document includes a list of examples of highly digitized models that are in scope: online search engines, online gaming, social media, online intermediation platforms, digital content streaming, cloud computing services, and online advertising services.

Amount A will also apply to consumer-facing businesses, which are businesses that provide goods to consumers for personal use. Examples include personal computing products, clothes, toiletries, cosmetics, luxury goods, branded foods and refreshments, franchise models (e.g., hotels and restaurants), and automobiles. The provision of intermediary goods is not included.

The Inclusive Framework is contemplating carve-outs for financial services and extractive industries for Amount A. Also, small MNEs will be scoped out (e.g., the Inclusive Framework could apply the CbC threshold of €750 million).

Nexus for Amount A will be based on substantial and systemic interactions with the market. The Inclusive Framework notes that the nexus standard is subject to revision and additional factors. An MNE will have nexus if it has generated in-scope revenue over a period of years in a jurisdiction. For automated digitized businesses in scope, the only nexus rule will be the revenue threshold, which will vary by jurisdiction. For consumer-facing businesses other than automated digitized services, the Inclusive Framework will explore additional or “plus” factors, which could include targeted advertising directed at a jurisdiction. However, mere sales of tangible personal property will be insufficient if the MNE is not physically present or does not specifically target the jurisdiction from a foreign location.

The Inclusive Framework also notes the need for future work to create clear and administrable rules. Although automated digitized services will be subject to “revenue nexus,” sourcing may also look to the location of users and other revenue based on consumption.

What has not been decided in the Inclusive Framework is the amount of profit that will be allocated. If there is agreement, countries will have to agree on a profit level (e.g., a return in excess of ten percent) and then decide that an amount in excess of that return (e.g., twenty percent of nonroutine profits) will be allocated to those countries where the MNE has the requisite nexus.18 An allocation key would determine the allocation of Amount A to those countries, possibly based on sales revenues.

Amount B is the routine return for activities in market jurisdictions, in particular, but not limited to, distribution activities that would be based on the arm’s-length principle and permanent establishment rules under Article 7 (e.g., physical presence). Amount B will take into account industries, regions, and functions.19

Amount C is for additional activities performed and not captured by Amount B that might warrant a profit in excess of the fixed return contemplated in Amount B.20 This could include virtually all activities, such as manufacturing, research and development, procurement, marketing and distribution beyond baseline amounts, and regional activities or back-office services. Amount C will also be determined by current nexus standards.

Leaning on Pillar One?

What do we believe the path to Pillar Two will require after what we have seen under Pillar One? The Inclusive Framework has not addressed several key design questions, including how Pillars One and Two interact and which rules apply first. This strongly suggests that design challenges remain for Pillar Two.

Even for Pillar One, significant work remains on the implementation of each of Amounts A through C for several straightforward questions. For Amount A, which country is the surrender jurisdiction? Does the home country or the country that holds the intellectual property provide the nonroutine return? What if multiple locations hold the IP—is Amount A divided among those investment hubs? What is the mechanism for relieving double taxation in the surrender jurisdiction? Is a tax credit sufficient? In addition to the challenges to selecting a jurisdiction, the costs and efforts associated with being the tax surrender jurisdiction could be a material challenge. Being such a jurisdiction is a significant burden from the perspectives of collection, allocation, and cross-jurisdictional audits.

Since the OECD initiated the BEPS process in 2012, the global economy has experienced significant growth. At some point, perhaps sooner than we all would like to believe, that growth may stop, and MNEs will incur losses.21 Will those losses be allocated beyond the home country or the innovation hub countries? How can businesses separate nonroutine losses from routine losses, and how should such losses be allocated?

From a purely economic perspective, the Inclusive Framework must agree to principles or blending rules on how the global minimum tax is applied and how they could interact with domestic tax law. In the United States, inconsistent approaches to state tax apportionment disputes on income lead to friction. Even if a single consistent allocation approach is applied, issues of characterization and deductibility, which go to the heart of how countries design their fiscal investment incentives, will proliferate.

Resolving tax disputes between countries for Amount A will be very challenging. Although the Inclusive Framework notes that it will need to establish innovative procedures, bilateral solutions will be insufficient. For example, due to the change in nexus, several countries may have claims to Amount A. If there is disagreement about the calculation of Amount A for a line of business or the global group, several countries may assert an increase of Amount A allocable to those jurisdictions. Countries have struggled with bearing the additional costs of existing unilateral audits, let alone developing resources to handle transactions that they have not yet audited and coordinating among potentially multiple taxing authorities. Bilateral dispute resolution will not be able to address multiple changes. Countries will inevitably need to agree to expand our existing competent authority resolution mechanism that has to consider these combinations of domestic and international tax law as we are seeing in Pillar One, and undeniably have impacts from a customs and value-added tax perspective for income under Amount B and, in some cases, Amount C. A chief OECD aim—preventing double taxation—will be sorely tested in this environment, and even in a perfect world this challenge will take years to resolve. The Inclusive Framework will need to create a process for these complicated disputes, and binding arbitration should be at the heart of it to ensure quick resolution.

Additionally, Pillar One will require a new multilateral instrument (MLI) that will apply uniformly to all participating countries. Unlike BEPS 1.0, BEPS 2.0’s MLI will not be a lengthy menu with options for each course. Everyone will need to jump into the pool at the same time.

Other Tax Frameworks?

February’s economic assessment provides clues to the design elements that the OECD may contemplate. Since simplicity is a key design element emphasized by the statement and in recent Tax Talks, the design elements will likely rely on elements of CbC documentation and existing taxing approaches of other member states.22 For example, an assumed minimum tax rate of 12.5 percent was applied for purposes of the impact assessment, which is similar to the Irish corporate income tax rate and rates targeted in Section 250 (FDII) of the United States’ TCJA. Income and taxes were combined per jurisdiction, similar to the approaches taken under CbC reports,23 and the income inclusion rule was assumed to apply before the BEAT-like undertaxed-payments rule.

In CbC reports, tax residency is already an item reported for each legal entity, and the master files already are the responsibility of the ultimate parent company’s taxing jurisdiction. Without further taxpayer guidance, one could envision the OECD’s CbC guidelines as pointing to a default tax surrender location. Similarly, for multilateral preaudits or rulings, a handful of countries already have preaudit processes in place. In some instances, these arrangements are multilateral, so the structure of these departments could be used as a model for future dispute resolution.

It may be tempting for the OECD to draw from the depth of local country experiences to design GloBE, but if Pillar One is any indication, the underlying complexity can be daunting. For example, any taxpayer who has a mix of jurisdictions with losses and profit, or who has tax credits or unused net operating losses, can attest to the need to coordinate local regulations and streamline these resolution processes.24 This “default” approach may appear simple and practical to the OECD, but taxpayers who have operated under these systems might find it better to provide input during this design phase than to simply wait for what is published.

Will the United States Implement the Inclusive Framework?

Even though its stated objectives were to strengthen the ability of countries to tax profit in low-tax jurisdictions and to continue to set their own rates,25 Pillar Two appears to pit at least two of the BEPS project’s stated goals against each other.26 In particular, the stated goal of Pillar Two—to develop rules for income inclusion and taxation of base-erosion payments—might enhance tax fairness, but at the cost of limiting tax sovereignty.

For example, a significant question for any taxpayer in the United States is whether Congress will support Pillars One and Two and, if so, will Congress ratify a multilateral treaty and pass legislation?

Unlike certain promulgations of the OECD, the pillars are not unilateral, because the United States must sign off on them to have the intended effect. The tax-writing committees are supportive of the OECD Inclusive Framework process and have written to Treasury and the European Commission expressing their support. But this support is for the process and not necessarily the result. Congress expressed its support, because it would like other countries to stop implementing digital services taxes that primarily affect U.S. MNEs. Congress is devoting resources to this issue—for instance, staff from the Joint Committee on Taxation attended the Pillars One and Two consultations.

Secretary of the Treasury Steven Mnuchin indicated in correspondence to the OECD that Congress will support a deal only if there is broad support and voiced “serious concerns” regarding mandatory departures from arm’s-length transfer pricing and taxable nexus standards.27 In addition, he proposed not only the treatment of Pillar One as a safe harbor but also a GILTI-like Pillar Two solution.

Another impediment is the multilateral treaty. The U.S. State Department’s current position is that it does not sign multilateral tax treaties. As discussed above, a bilateral treaty approach for Pillar One does not work. Moreover, the United States has a relatively anemic tax treaty network that would not prevent double taxation for Amount A. If U.S. treaties prevent the successful implementation of Amount A under Pillar One, we may still see a proliferation of unilateral measures. The United States will need to be flexible and consider a multilateral treaty for the pillars to operate as advertised.

Congress would also need to revisit the TCJA and make conforming changes to the BEAT and possibly GILTI. Perhaps those changes will ensure the BEAT will actually target base erosion, as opposed to creating a zombie-like alternative minimum tax. To date, Republicans generally have not advocated for changes to the international provisions of the TCJA, other than draft technical corrections legislation written by Representative Kevin Brady (R-Texas). Additionally, Democrats do not appear to be in a hurry to fix a partisan tax bill, even though certain features of the TCJA are undermining Democrats’ priorities. If Congress does not pass legislation and the rest of the Inclusive Framework proposal moves forward, then U.S. MNEs will be subject to multiple levels of tax, and the United States will lose revenue as a market jurisdiction. Unilateral action by Treasury could create an unstable system that would be subject to court challenge. In sum, the political parties will need to come together to support changing the TCJA to implement the Inclusive Framework.


In the sprint to the end of 2020, without taxpayer input at the ready, the OECD will very likely adopt existing countries’ approaches, such as elements of BEAT or existing preaudit facilities, which may seem simple but could lead to the global proliferation of poorly developed tax policy. A key takeaway here is that these pillars under the Inclusive Framework are not a limited-scope project to address the digital economy. Rather, Pillars One and Two are broad and affect all global businesses. The sweeping impact of Pillar Two will touch all aspects of an MNE’s tax structure, from tax credits to transfer prices to compliance burden and audit resolution.

The OECD is implicitly requesting taxpayer engagement on the design and implementation elements of both pillars. Business should continue to participate in the OECD wherever possible by filing comments and attending consultations. Engagement with countries is key, because the countries are in the room where everything happens. We are now in a sprint, but it would be better for the process to be more of a marathon. But, as any good runner knows, solid preparation is needed, the earlier the better. Taxpayers must prepare now to provide input, proposals, or design solutions to ensure the new global minimum tax system is grounded in practical considerations and to avoid a tax foot fault on a global scale.

Gene Tien and Joshua Odintz are partners at Baker McKenzie.


  1. “Addressing the Tax Challenges of the Digitalisation of the Economy—Policy Note,” OECD/G20 Base Erosion and Profit Shifting Project, OECD, January 23, 2019,
  2. The four components of GloBE include an income-inclusion rule, a switchover rule, an undertaxed-payments rule, and a subject-to-tax rule.
  3. “Programme of Work to Develop a Consensus Solution to the Tax Challenges Arising from the Digitalisation of the Economy,” OECD, May 31, 2019, www.oecd.org/tax/beps/programme-of-work-to-develop-a-consensus-solution-to-the-tax-challenges-arising-from-the-digitalisation-of-the-economy.htm.
  4. In 2012, the BEPS project set forth the start of an action plan focused on protecting the thirty-plus OECD countries’ tax base by strengthening international tax rules around tax revenues, tax sovereignty, and tax fairness. By 2020, over 135 nations have worked together to implement the BEPS project’s fifteen Action Items, and they are currently creating a new international tax framework by completing a comprehensive report covering the two “pillars” by the end of 2020.
  5. “Addressing the Tax Challenges of the Digitalisation of the Economy—Policy Note,” OECD/G20 Base Erosion and Profit Shifting Project, January 23, 2019.
  6. “Secretariat Proposal for a ‘Unified Approach’ under Pillar One,” OECD, October 9, 2019.
  7. “Global Anti-Base Erosion Proposal (GloBE)—Pillar Two,” OECD, November 8, 2019.
  8. “Statement by the OECD/G20 Inclusive Framework on BEPS on the Two-Pillar Approach to Address the Tax Challenges Arising from the Digitalisation of the Economy,” OECD/G20 Base Erosion and Profit Shifting Project, January 30, 2020.
  9. “Addressing the Tax Challenges of the Digital Economy—Action 1: 2015 Final Report,” OECD/G20 Base Erosion and Profit Shifting Project, 2015.
  10. “Global Anti-Base Erosion Proposal (GloBE)—Pillar Two,” OECD, November 8, 2019.
  11. OECD Tax Talks #14, OECD, January 31, 2020, www.oecd.org/ctp/tax-talks-webcasts.htm.
  12. “Statement by the OECD/G20 Inclusive Framework on BEPS on the Two-Pillar Approach to Address the Tax Challenges Arising from the Digitalisation of the Economy,” OECD/G20 Base Erosion and Profit Shifting Project, January 20, 2020, Section 2.1.
  13. Ibid., Section 2.3.
  14. OECD Tax Talks #14, OECD, January 31, 2020, www.oecd.org/ctp/tax-talks-webcasts.htm.
  15. “Statement by the OECD/G20 Inclusive Framework on BEPS on the Two-Pillar Approach to Address the Tax Challenges Arising from the Digitalisation of the Economy,” OECD/G20 Base Erosion and Profit Shifting Project, January 30, 2020, Section 2.
  16. Ibid., Section 2.5.
  17. “Webcast: Update on Economic Analysis & Impact Assessment,” OECD, February 13, 2020, www.oecd.org/tax/beps/webcast-economic-analysis-impact-assessment-february-2020.htm.
  18. These levels were applied in the OECD’s economic impact analysis presented on February 13, 2020.
  19. “Secretariat Proposal for a ‘Unified Approach’ Under Pillar One,” at paragraph 64, OECD, October 9, 2019.
  20. Ibid.
  21. It is too early to tell if the current coronavirus, COVID-19, will ultimately lead to a recession, but the OECD and Inclusive Framework must take this possibility into consideration.
  22. “Guidance on Transfer Pricing Documentation and Country-by-Country Reporting – Action 13: 2014 Deliverable,” OECD/G20 Base Erosion and Profit Shifting Project, 2014.
  23. Annex III to Chapter V, ibid.
  24. Examples include the United States’ Compliance Assurance Program or the United Kingdom’s Profit Diversion Compliance Facility.
  25. “Addressing the Tax Challenges of the Digitalisation of the Economy—Policy Note,” OECD/G20 Base Erosion and Profit Shifting Project, January 23, 2019.
  26. “Addressing Base-Erosion and Profit Shifting,” OECD, 2013, www.oecd.org/tax/addressing-base-erosion-and-profit-shifting-9789264192744-en.htm.
  27. U.S. Treasury Secretary Steven Mnuchin’s letter to OECD Secretary-General José Ángel Gurría dated December 3, 2019, www.orbitax.com/news/archive.php/U.S.-Treasury-Secretary-Sends–40283.

Leave a Reply

Your email address will not be published. Required fields are marked *

XHTML: You can use these tags <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>