G-24 Working Paper: Tax certainty options in the context of BEPS 2.0 Natalia Quiñones * | March 2022 * Natalia Quiñones is the former Head of the International Tax Office at DIAN in Colombia, and a former member of the Steering Group at the Inclusive Framework. She teaches at Universidad de los Andes, has published several articles on dispute resolution and has over 15 years of experience in private practice in Colombia.

2 Introduction International tax dispute prevention and resolution mechanisms have traditionally relied on bilateral negotiations between competent authorities, in the form of the Mutual Agreement Procedure(MAP), and, more recently, in the form of Advanced Pricing Agreements(APAs). These mechanisms have remained mainly unchanged since they were first designed and implemented(at the beginning of the 21 st century in the case of the MAP). Despite the historical stability of the regime, much of the developing world does not have extensive experience in the application of these mechanisms, in some cases due to there being a limited or non-existent treaty network, and in other cases due to only limited use of the mechanism by taxpayers operating in their jurisdictions. In 2015, the BEPS Action 14 Report established a minimum standard for all the Inclusive Framework (IF) members in an effort to address ongoing criticisms based mainly on the lack of finality and the uncertain duration of the MAP. This standard is currently being enforced by a peer-review system created by the Forum on Tax Administration(FTA) MAP Forum, which allows for deferral of the review for non-members of the G20 and non-members of the Organisation for Economic Co-operation and Development(OECD) that have low levels of MAP disputes. However, the OECD narrative since 2007 has been pointing towards the need to complement the MAP with mandatory binding solutions, leading to the inclusion of article 25.5 in the OECD Model Tax Convention in 2017 and of Chapter VI in the Multilateral Convention to Implement Tax Treaty Related Measures to Prevent Base Erosion and Profit Shifting(MLI). Despite developing country opposition, this trend is being further advanced with the tax certainty block included in the solution proposed to address the tax challenges arising from digitalization of the economy. The agreements published in October 2021 by the G-7, G-20, and the Inclusive Framework on this topic include dispute prevention and resolution mechanisms for the first of the solutions(Pillar One), which limits source/market country taxing rights with regard to remote or digital activities in exchange for the right to tax a small portion of the excess profits made by some of the largest globalized taxpayers. The dispute prevention and resolution mechanisms will deal with potential conflicts for Amount A and“all issues related to Amount A(e.g., transfer pricing and business profits disputes), in a mandatory and binding manner.” 1 Further, it announces that“An elective binding dispute resolution mechanism will be available only for issues related to Amount A for developing economies that are eligible for deferral of their BEPS Action 14 peer review and have no or low levels of MAP disputes.” 2 Because of this last exclusion, and because developing countries have traditionally objected to mandatory binding arbitration, the design of the mechanisms agreed in October are more of a priority for the supporters of mandatory binding solutions, led by the United States and other developed economies. 1 OECD Inclusive Framework(8 October, 2021). Statement on a Two-Pillar Solution to Address the Tax Challenges Arising from the Digitalisation of the Economy. Available at https://www.oecd.org/tax/beps/statement-on-a-two-pillar-solution-to-addressthe-tax-challenges-arising-from-the-digitalisation-of-the-economy-october-2021.pdf P.2.(last accessed on 21.2.2022) 2 Ibidem. This opt-out mechanism is only available for non-OECD, non-G-20 countries.

3 However, in this scenario it is clear that many developing countries will be subject to mandatory and binding dispute resolution mechanisms by virtue of the announced agreement, which has been signed by 137 jurisdictions, out of which over 90 are non-OECD, non-G-20 jurisdictions. However, the only differential focus in dispute resolution proposed by the OECD for developing countries consists of the elective binding mechanism, which will probably be temporary in nature. Furthermore, the announcement states that the elective mechanism covers only issues related to Amount A, so that all developing countries in the agreement would be bound by the mandatory and binding mechanism for Amount A disputes. It is possible that political hurdles will delay the implementation of the agreement because of the difficulties in parliamentary approval of the multilateral instrument to implement these changes(the Multilateral Convention or MLC 3 ). The United States Senate Foreign Relations Committee, for example, has recently expressed concerns over the United States Treasury Department’s negotiation of a global tax deal without proper engagement of the members of that committee. 4 Developing countries may also face difficulties in parliamentary approval of the agreement given the uncertain and generally insignificant economic impact of the global solution of Pillar One in terms of revenue. 5 In spite of these hurdles, it is still foreseeable that in the near future mandatory binding dispute resolution will be adopted as an international standard in tax matters. Indeed, the presence of this chapter in the MLC may help to obtain Congressional approval of the instrument even in the United States, where the Senate has resisted joining other multilateral tax efforts, such as the MLI and the Multilateral Competent Authority Agreement(MCAA). Over the next few months, the executive will try to convince the United States Congress of the need to approve the MLC, and it will be backed by the United States Council for International Business(USCIB), mainly because approval will entail the removal of unilateral measures and the presence of mandatory binding dispute resolution, not just for the economically insignificant Amount A, but for all issues related to Amount A. Four G-24 members(Nigeria, Kenya, Pakistan, and Sri Lanka) have not endorsed the October 2021 agreement, while 16 other G-24 countries have endorsed the agreement(Argentina, Brazil, China, Colombia, Cote d’Ivoire, Democratic Republic of the Congo, Egypt, Gabon, Haiti, India, Mexico, Morocco, Peru, South Africa, Trinidad and Tobago). Countries that have so far endorsed the deal may still change their position and withhold their signature or parliamentary approval of the MLC until there is enough evidence that relevant jurisdictions like the US and China will ratify the instrument and continue to consider the potential trade-offs of implementing Pillar One instead of pursuing domestic unilateral measures. In particular, they should carefully examine the economic impact analysis of Pillar One, especially in the case of countries with a limited or non-existent treaty network. However, the 3 As per the Statement published in October, the MLC will be open for signatures in a high-level signing ceremony in mid-2022 and will enter into force“in 2023 once a critical mass of jurisdictions as defined by the MLC have ratified it”(Ibidem, p.6). 4 United States Senate(2021).“Ranking Members Warn Against Bypassing Treaty Process”(8 October, 2021). Available at https://www.finance.senate.gov/ranking-members-news/ranking-members-warn-against-bypassing-treaty-process(last accessed on 21.2.2022) 5 OXFAM(2021). The effect of the OECD’s Pillar One proposal on developing countries- An impact assessment. Available at https://www.oxfamireland.org/sites/default/files/pillar_1_impact_assessment_-_04.0.21_final.pdf.

4 OECD and the EU will be pressing for ratification as has been the case for the BEPS MLI. 6 This political pressure will probably affect developing countries that did not sign the Statement, especially if their audits feature high-profile digital companies under the Significant Economic Presence legislation or other so-called unilateral measures. The present working paper aims to provide G-24 policy makers with design elements that can address specific concerns raised in the context of the mandatory and binding mechanisms to be agreed in the Multilateral Convention for Amount A and the issues related to Amount A. The possibility of requesting design features that benefit developing countries is essential for producing a more stable solution that will capitalize on the opportunities and reduce any medium- and long-term negative impacts for those countries that will decide to join. This will also be important for countries that will not be immediately bound by the mechanism, but that may join later on without the possibility of modifying the design. The OECD proposal on dispute resolution so far 7 The OECD Secretariat has been working along with some developed country delegates to establish the general framework for tax certainty in Pillar One, maintaining the privileges of administering the procedure, as it did for the MLI.8 The Secretariat has envisioned two separate mechanisms: a mandatory binding dispute prevention and resolution mechanism for Amount A and a mandatory binding dispute resolution mechanism for issues related to Amount A. The Statement made no specific mention of the issue of costs for both types of mechanisms, and it seems that the issue is still undecided. However, G-24 countries must bear in mind that the default cost allocation mechanism in the MLI is that each country bears its own participation costs and that countries share common costs equally.9 Dispute prevention and resolution for Amount A The Pillar One Blueprint published in October 2020 contains the main design elements for the dispute prevention and resolution mechanism that will define the amount, the paying entities, and distribution of Amount A for every MNE within the scope that wishes to obtain early tax certainty.10 It is important 6 Oei, Shu-Yi, World Tax Policy in the World Tax Polity? An Event History Analysis of OECD/G20 BEPS Inclusive Framework Membership(September 6, 2021). Yale Journal of International Law, Vol. 47, 2021-22, Available at SSRN: https://ssrn.com/abstract=3918499(last accessed on 21.2.2022) 7 This section was prepared based on public information available, as well as information from two interviews conducted with two members of the OECD Secretariat during October 2021. 8 The power of the seat is reflected in procedural decisions. For example, under the MLI part VI rules, if arbitrators cannot agree on a Chair for the panel,“the appointment will be made by the highest-ranking official of the Centre for Tax Policy and Administration of the OECD that is not a national of either Contracting Jurisdiction”, that is, Mr. Pascal Saint Amans, unless France is involved in the dispute.(MLI Explanatory Statement, at 237). This choice has been criticized even by developed country authors(See Ault, Hugh(2021), Tax Treaty Arbitration: A Reassessment. In Kofler,G., Mason, R., and Rust, A.(Eds.) Thinker, Teacher, Traveler. Reimagining International Tax. Essays in Honor of H. David Rosenbloom. Amsterdam, IBFD. P.32.). 9 MLI Explanatory Statement, at 254. 10 OECD(2020), Tax Challenges Arising from Digitalisation – Report on Pillar One Blueprint: Inclusive Framework on BEPS, OECD/G20 Base Erosion and Profit Shifting Project, OECD Publishing, Paris, https://doi.org/10.1787/beba0634-en. Paragraphs 17-19.(last accessed on 21.2.2022)

5 to mention that the quantum of Amount A will not be very contentious given the formulaic nature of its calculation, at least in cases with no financial segmentation. In contrast, there may be challenges and potential disputes in the application of sourcing rules for the distribution of Amount A, as well as for the selection of the paying entities and the jurisdiction in charge of relieving double taxation. The general idea is that each Multinational Entity(MNE) covered by Amount A will file a Country by Country report(CBC) in the Ultimate Parent Entity(UPE) jurisdiction and will prepare an Amount A proposal for the UPE to review. This UPE will finalize the proposal for Amount A determination and distribution as well as the identification of paying entities and the amount of residual profit allocated to them, based on the taxpayer proposal, the information submitted in the CBC and any additional information provided by the taxpayer. This proposal may be reviewed by a panel composed by a limited number of Competent Authorities(CAs) representing relevant jurisdictions(the“Review Panel”). It is probable that jurisdictions that are not represented in the panel will still get an opportunity to present written objections/observations to the UPE proposal, which the Review Panel will examine before issuing a decision to accept the UPE proposal, or an alternative proposal. It is important to mention that the Secretariat envisions a review mechanism whenever the panel has been unable to reach an agreement on any of the features of Amount A or if an interested jurisdiction that is not represented on the panel disagrees with the outcome of the Review Panel. In this case, the competing proposals will be submitted to a second panel(the“Determination Panel”), which can bring together competent authorities and independent members. The panel composition and mechanisms for nomination are still to be decided. The decision of this panel is to be binding upon all jurisdictions interested in Amount A, even if they were not directly represented on the panel, although the probability of sending written objections and observations to the panel still exists. It is important to note that jurisdictions that believe they are entitled to receive a portion of Amount A, but are not included in the proposal approved by the Determination Panel, will still be bound by this decision. The presence of the MNE in this mechanism was also granted in the July agreement, but it is unclear if they can be represented by law firms before the panel or if they can just accompany the UPE jurisdiction to answer questions from the panel regarding the proposal. It is safe to assume, however, that MNEs will retain law firms and accounting firms for the preparation of their CBC documentation and the initial Amount A proposal. Given that in many cases the IRS will act as the UPE administration,11 and given the IRS MAP procedure, these resources will almost certainly be used by the IRS in the preparation of its position in the review and determination panels, even if the MNE is not called for questioning. There has not been a public statement on the publicity and transparency of the prevention panels. 11 See Eden, L.(2021). Taxing The Top 100—Part 1: Who’s In, Who’s Out?. Available at https://news.bloombergtax.com/dailytax-report/taxing-the-top-100-part-1-whos-in-whos-out.(last accessed on 21.2.2022)

6 Dispute resolution for issues related to Amount A The scope of the issues that may be submitted to the dispute resolution mechanism was explained in an internal paper delivered to members of the Inclusive Framework that has not been made public as of November 5th, 2021. An explanatory parenthesis in the Statement lists transfer pricing and business profits disputes as examples of what may constitute issues related to Amount A.12 In practice, what this means is that unilateral transfer pricing adjustments and profit attribution audits performed by G24 tax administrations are likely to be considered“issues related to Amount A,” and are therefore subject to the mechanisms envisioned for this type of dispute. In other words, if a tax administration were to determine that a large MNE should attribute greater profits(or smaller losses) to the local office because certain expenses had not been borne by that specific permanent establishment, then the MNE could request a determination that the adjustment be considered an“issue related to Amount A” in order to remove it from the jurisdiction of domestic courts. However, given the formulary nature of Amount A, it seems logical that disputes can only be related to Amount A if they affect financial profits before taxes, which is the point of departure for the calculation of Amount A. As per the Statement, the determination of whether a dispute affects pre-tax profits will be made in the mandatory binding mechanism for OECD and G-20 countries, and electively for the other members of the IF that signed the agreement and that fulfill both conditions, namely, that they obtained deferral for the Action 14 Peer Review and that they have a low MAP inventory. Below is a list of G-24 member countries and observers with their current status in relation to the Pillar One certainty mechanisms: Table 1: G-24 Member country status as of February 2022 G-24 Member Country *G-24 country observers Angola* Argentina Brazil China* Colombia Cote d’Ivoire Democratic Republic of Congo Pillar One dispute resolution status if MLC is ratified Elective binding mechanism applies(0 cases for 2020) Mandatory binding mechanism applies(G20) Mandatory binding mechanism applies(G20) Mandatory binding mechanism applies(G20) Mandatory binding mechanism applies(OECD) Elective binding mechanism applies(NA) Elective binding mechanism applies(NA) 12 OECD Inclusive Framework(8.10.2021). Statement on a Two-Pillar Solution to Address the Tax Challenges Arising from the Digitalisation of the Economy. Available at https://www.oecd.org/tax/beps/statement-on-a-two-pillar-solution-to-address-thetax-challenges-arising-from-the-digitalisation-of-the-economy-october-2021.pdf p.2.(last accessed on 21.2.2022)

7 G-24 Member Country *G-24 country observers Egypt Gabon Haiti India Indonesia* Kenya Mexico Morocco Nigeria Pakistan Peru Saudi Arabia* South Africa Sri Lanka Trinidad and Tobago UAE* Algeria Ecuador Ethiopia Ghana Guatemala Iran Lebanon Philippines Syria Venezuela Pillar One dispute resolution status if MLC is ratified Elective binding mechanism applies(NA) Elective binding mechanism applies(NA) Elective binding mechanism applies(NA) Mandatory binding mechanism applies(G20) Mandatory binding mechanism applies(G20) Opted out of October agreement Mandatory binding mechanism applies(G20) Elective binding mechanism applies(5 cases for 2020, may be considered a high MAP caseload) Opted out of October agreement Opted out of October agreement Elective binding mechanism applies(1 case for 2020, will enjoy elective status until OECD membership confirmed) Mandatory binding mechanism applies(G20) Mandatory binding mechanism applies(G20) Opted out of October agreement Elective binding mechanism applies(1 case for 2020) Elective binding mechanism applies(0 cases for 2020) Not a member of the IF Not a member of the IF Not a member of the IF Not a member of the IF Not a member of the IF Not a member of the IF Not a member of the IF Not a member of the IF Not a member of the IF Not a member of the IF

8 The Secretariat is considering baseball arbitration as the basic mechanism for the design of the dispute resolution modality for all issues related to Amount A. If the design is to be inspired by MLI final-offer arbitration as per articles 23(1) and 23(2), the competent authorities of each jurisdiction involved in the dispute must submit a final offer to the arbitration panel, which is responsible for choosing only one of the final offers presented. The composition of the panel, as is the case for the Amount A dispute prevention mechanism, has not yet been defined. Again, it is likely that they will propose a mix of CAs and independent members on the panel. This form of mandatory binding dispute resolution would not be imposed on developing countries outside the OECD and the G-20 in the first stages, which is why countries who are eligible for deferral in their MAP peer review13 can opt out at this stage as long as they demonstrate that they have a low MAP inventory. Nonetheless, G-24 countries belonging to this group must realize that this exclusion is most likely to be temporary in nature, and that arbitration will probably become a minimum standard in the short to medium term. It is important to note, in any case, that it is likely that the first disputes will involve more developed jurisdictions, given that the adjustments on residual profits will most probably be conducted in the countries where MNEs conduct research and development, or maintain their intellectual property. As to the enforceability of the decision made by the resolution panel, the Secretariat is thinking of allowing countries participating in a dispute to adopt the decision through a MAP, where they are able to deviate from the decision if they so agree.14 Of course, if there is no agreement on the deviation, because the mechanism is binding, the Competent Authorities have an obligation to adopt the decision promptly via the MAP. Hence, the enforceability of the dispute resolution will follow the same procedure as for any other MAP decision, which will put additional pressure on countries to fulfill the minimum standard requirements for the implementation of MAP decisions, notwithstanding domestic time limitations. Depending on domestic regulations and constitutional derivations of the right to go to court, taxpayers may challenge the result of the MAP agreement that implements the resolution decision.15 In this latter case, the result is no longer binding on either of the countries involved, and the resources spent on the mechanism will be wasted. 13 Albania, Antigua and Barbuda, Angola, Anguilla, Armenia, Belize, Benin, Bosnia and Herzegovina, Botswana, Burkina Faso, Cabo Verde, Cameroon, Congo, Cook Islands, Costa Rica, Côte d’Ivoire, Democratic Republic of Congo, Djibouti, Dominica, Dominican Republic, Egypt, Eswatini, Gabon, Georgia, Grenada, Haiti, Honduras, Jamaica, Jordan, Kenya, Liberia, Malaysia, Maldives, Mauritius, Mongolia, Montenegro, Montserrat, Namibia, Republic of North Macedonia, Nigeria, Pakistan, Panama, Papua New Guinea, Paraguay, Peru, Saint Lucia, Saint Vincent and the Grenadines, Senegal, Seychelles, Sierra Leone, Sri Lanka, Turks& Caicos Islands, Ukraine, Uruguay, Zambia, as per the latest public OECD information available at https://www.oecd.org/tax/beps/beps-action-14-peer-review-assessment-schedule.pdf.(last accessed on 21.2.2022) Additionally, countries must demonstrate a low MAP case-load. 14 This would resemble Alternative A of article 25 in the UN Model, which also contemplates the possibility of agreeing to a MAP that deviates from a prior panel decision. 15 This is the case, for example, in the United States of America.

9 Within this framework, there is an option of an international appeals procedure, which has not yet been decided. The appeals mechanism would probably follow the same procedural rules as the resolution panel. Without the appeals, the only further possibility of challenging a resolution panel decision will be through a negotiation under the MAP that adopts the said decision. Furthermore, it is still unclear if Amount B will follow the same mechanism and grant access to taxpayers. It would seem logical that Amount B disputes will be treated as“issues related to Amount A” if the dispute involves any of the MNEs within the scope of Pillar One. Design features to address concerns about the mandatory binding dispute resolution Although the discussion on the design features of the prevention and resolution mechanisms may seem less of a priority for G-24 countries, given the elective mechanism announced in the Statement, it is highly recommended that developing countries participate in the discussions in order to prevent undesirable situations such as having to join the system later on, when participating countries will have more experience and will have written the rules according to their own interests and capabilities. Making proposals for the design of the dispute prevention and resolution mechanisms does not exclude other strategic alternatives that G-24 countries may explore to reduce the impact of the mandatory binding dispute resolution for Pillar One, as mentioned in the introduction. Possible complementary strategies for G-24 countries ★ Withholding ratification of the MLC until the United States ratifies the instrument. o This decision is consistent with political commitments acquired with the signature of the IF Statement. o This will limit the country’s possibility of levying Amount A, which is likely to be very small in economic terms compared to the amounts that may be levied with domestic legislation designed to tax remote and digital businesses. ★ Re-shaping transfer pricing and business profits audits performed on covered MNEs in order to focus more on deductibility requirements rather than arm’s length principles in order to escape the“issues related to Amount A” qualification, while taking care to respect the terms of article 24 of existing treaties. Costs Dispute prevention and resolution mechanisms require extensive investments in human resources, document production, translations, travel, and telecommunications. In some cases, access to expert witnesses and transfer pricing databases is also important for effective participation in a dispute prevention or resolution mechanism. These costs have traditionally been one of the main reasons for

10 developing country opposition to international tax arbitration.16 Furthermore, there is a perceived disadvantage for developing countries because of the taxpayer financial resources that usually contribute to the parent entity position rather than the market entity position. Some of the costs can be reduced by the implementation of efficiency measures, such as cutting inperson meetings to the minimum necessary and taking advantage of virtual communications technology. In the case of the dispute resolution mechanism, if there is a plural number of cases in a single year involving the same parties, an accumulation of cases with similar issues should be allowed in order to reduce costs. The cost allocation mechanism contemplated in article 25 in Part VI of the MLI creates a high degree of inequality when the costs are expressed as a percentage of GDP, especially for smaller and developing countries. Besides, the costs for legal fees are known to be the highest,17 and therefore developing countries will be at a disadvantage, if each country is to pay for the panel member that they nominate. This inequality will result in developed countries nominating costly experts while developing countries stick to competent authorities because of the costs of hiring an independent expert who could successfully counter the views of the expert retained by the developed country. This may have serious consequences in practice, since it is highly likely that the independent expert voice will dominate the panel discussion and will steer the decision in one direction. On the other hand, a full exemption from all costs would be strategically unrealistic, as it would reduce developing country rights to participate in stakeholder decision processes, and it could even encourage extreme audit positions if there are no costs associated with defending this position. One possible solution to reduce the impact of costs for G-24 countries and to introduce a differential focus for developing countries is to establish total cost procedures for each determination, review, or resolution panel in which a developing country is to participate, and then split the costs proportionately to GDP amongst all participating countries. These costs would have to include every expense related to the procedure, including panel member fees, expert witness fees, translations, travel costs, and any administrative fee charged by the administering body. This would in practice result in the majority of costs being borne by the developed countries, which is a fair outcome given the fact that mandatory binding dispute resolution mostly benefits companies who are residents of those jurisdictions. Otherwise, G-24 countries could request a cap to the costs charged to developing countries, expressed as a percentage of GDP. 16 Arias, Isaac and Calderoni, Anarella.(7.9.2021)“Analyzing Mandatory Binding Arbitration and the MAP”. Available at https://www.ciat.org/ciatblog-analyzing-mandatory-binding-arbitration-and-themap/?lang=en&utm_source=WebBlog&utm_medium=CartaAutor&utm_campaign=analisis-del-arbitraje-vinculante-obligatorioy-del-pam&utm_content=Septiembre2021.(last accessed on 21.2.2022) See also 17 International Chamber of Commerce(ICC)(2018). Techniques for controlling time and costs in arbitration: Report from the ICC Commission on Arbitration. Available at https://iccwbo.org/content/uploads/sites/3/2018/03/icc-arbitration-commissionreport-on-techniques-for-controlling-time-and-costs-in-arbitration-english-version.pdf(last accessed on 21.2.2022)

11 Another(less ideal) alternative can be drawn from WTO experience, where developed countries regularly contribute to the Global Trust Fund and the Aid for Trade fund.18 A central developing country defense fund could be used to finance the nomination of independent panel members or expert witnesses. A degressive percentage of the amounts in question(i.e., a smaller percentage for larger amounts) could be posted as a mandatory contribution to this fund by the levying jurisdiction. Alternatively, at the initial stage, the fund could include contributions from all countries participating in the mandatory binding resolution mechanisms proportionate to GDP. Some G-24 countries have national agencies dedicated to the legal defense of the country, especially in cases in which the country has faced investment arbitration claims. These agencies usually have ample resources that can be used in practice to fund legal and expert witness expenses. Finally, G-24 countries should find a balance between the possibilities of appealing against an unfair decision and limiting the scenarios for appeals before the determination panel, which are likely to increase the costs of the mechanism. A simple majority decision could be enough at the Review Panel level, and the Determination Panel need only meet if there is a disagreement in the application of a principle. In all cases, the reasons for establishing a Determination Panel must be clearly set out in the rules. Cost-related strategies for G-24 countries ★ Propose virtual meetings and case accumulation for cost efficiency. ★ Propose common funding of total costs split proportionate to the GDP of participating jurisdictions in each mechanism. ★ Propose ceilings to the costs borne by developing countries expressed as a percentage of GDP. ★ Propose the creation of a developing country fund to be financed by developed countries and by a percentage of the amounts in discussion for each panel, to be contributed by the levying country. ★ Take advantage of national legal defense agencies to use the resources to retain legal or accounting advice, or to bring an expert witness to the procedure. ★ Demand clarity in the rules that allow for a Determination Panel or any other mechanism for appeals. Panel composition Although the preliminary Review Panel for Amount A will decide on the proposal of the UPE jurisdiction, it is less likely that developing countries will bear an economic loss as a result of the decision of this panel. Contentious issues for the Review Panel to decide are limited to the sourcing rules and the determination of which entity within the group is responsible for the residual profits earned by the group. These issues may affect G-24 countries in the allocation of Amount A and in the alleviation of double taxation. However, in principle, it is possible that the UPE proposal will be acceptable to all 18 See https://www.wto.org/english/tratop_e/devel_e/teccop_e/financing_trta_e.htm.(last accessed on 21.2.2022)

12 countries in most cases, so in more than 90 per cent of cases the Review Panel will be enough to decide, and the issue will not go to the Determination Panel. In order to reduce costs associated with this procedure, G-24 countries could support the inclusion of CAs only in the Review Panel. This requires, of course, a balanced mix of panel members who can properly represent the views of developing countries, especially if a sourcing rule with various possible interpretations is involved. In all cases, G24 countries must ensure that developing country views will be defended by at least two members of the panel, depending on the final number of members to be included in the Review Panel. Lastly, it is crucial that all interested countries(regardless of whether they appear as members of the panel) are given the opportunity to present written objections/commentaries to the UPE proposal, or even to present an alternative proposal altogether for the Review Panel to evaluate. The Determination Panel and the Resolution Panel, in contrast, may require the presence of independent experts, as the failure of the MAP indicates that tax authorities have irreconcilable positions. The benefit of having independent experts lies in the possibility of unlocking persistent positions entrenched in high-income country CAs, especially when they adhere to an outdated OECD standard(as is the case in most permanent establishment(PE) rules). Independent experts are more likely to accept an interpretation of the standard that can more closely reflect the new ways of doing business(for example, admitting that an infrastructure construction project should give rise to source country taxation even if performed with 3D printers well below the UN’s six-month construction PE threshold). Besides, if the CAs in the Review Panel have failed to reach an agreement, it is probable that a new panel comprising only CAs will reflect the same divided opinion, so it is advisable to include experts who can shine a new light on how to resolve pending issues. However, if both tax authorities agree that other tax authorities from similar backgrounds may reach a different conclusion, then it is best to exhaust an independent CA stage before the independent experts are brought in. In this case, expertise should be the main criterion for choosing a panel member, as the latter is the most likely voice to steer the decision of the Determination Panel, especially when there is a need to conciliate tax sourcing rules by taking account of the financial nature of the Amount A determination. While it is true that it is difficult to find developing country nationals who are sufficiently independent and have much expertise in transfer pricing and sourcing rules, it is also true that there are many experts from other nationalities who support the views and interpretations of developing countries on important points such as holding an expansive view of the activities that generate a nexus in market countries, or excluding deductibility adjustments from the scope of“issues related to Amount A.” Moreover, in order to prevent the lack of diversity in the appointment of independent panel members, it might be useful to propose limitations on the nomination process so that the same panel members are not always chosen. It is thus suggested that non-CA panel members can only sit for a maximum of one panel per year and cannot sit on a panel for the same taxpayer until five years have passed. In order to find experts who are aligned with developing country views and to provide more transparency to the appointment of independent panel members, G-24 countries must promote the inclusion of strict disclosure rules for non-CA panel members. Within this framework, every expert must disclose any public view that they have defended orally or in writing concerning the issues at stake.

13 This disclosure obligation facilitates the identification of conflicts of interest and allows for greater transparency for the entire mechanism. In all cases, G-24 countries will be encouraged to adopt strong criteria for standards of impartiality and independence of every expert with respect to any given taxpayer to exclude biased decision makers from the panels, especially since not all countries that have an interest in the dispute will be represented in the prevention panels. It is certainly useful to start putting together a several lists of experts/panel members who either come from G-24/developing countries or else show a deep understanding of low- and middle-income country positions, in order to facilitate adequate representation of developing country positions in the prevention and resolution panels. Panel Composition recommendations for G-24 countries ★ Propose sole participation of CAs for the Review Panel. ★ Propose an intermediate independent CA stage where CAs from similar countries are called to see if a solution may be jointly proposed by them. This stage would operate before independent experts are brought in, but only if both CAs consider that a set of independent CAs may reach an agreement. ★ Accept independent members in the Review and Resolution Panel, as long as they comply with strict disclosure and conflict of interest rules. ★ Promote diversity in independent panel members by limiting non-CA panel members to a maximum of one panel per fiscal year and prevent the same panel member from sitting in a procedure involving the same taxpayer in a period of five years. ★ Put together lists of experts/panel members that either come from G-24/developing countries or fully understand their views. Baseball vs. Independent Opinion Final offer or“baseball” arbitration was invented in the United States for sports cases and is still used by the Court of Arbitration for Sport(CAS) in the case of international anti-doping disputes, inspired by the need to have a rapid determination in the framework of a live sports competition. The original purpose of its adoption in tax matters, beyond the agility of the procedure, is closely related to the reluctance of states to renounce sovereignty in tax matters. On the one hand, it serves as a strong incentive for countries to agree in the MAP, thereby preserving sovereignty in the hands of the competent authorities. This fact has been verified in Europe, where the lack of arbitration cases in spite of the presence of the European Arbitration Convention shows that the“all or nothing” approach is effective in promoting negotiation and preventing extreme audit positions. On the other hand, even when the dispute advances into arbitration, the fact that the arbitrator cannot deviate from the competent authority proposals implies that there is substantial control by the CAs on the outcome of arbitration. These advantages are accompanied by reduced resolution times and reduced costs, as well as a system that is farther removed from the investment arbitration system that has raised significant concerns in the developing world.

14 Nevertheless, baseball arbitration as practiced in the US has two major disadvantages for developing countries: the difference in experience increases the likelihood that a developed-country final offer will be chosen over the developing-country position, and the opaque and confidential nature of the procedure and the award make it very hard to establish a learning curve. To overcome both disadvantages, G-24 countries may insist on the need to include a reasoned decision so that one position may be chosen over another position based on the prevalence given to the principle behind the chosen offer, and not because of the experienced form of presentation of the case. The publication of a redacted brief, as discussed below, will not increase costs or reduce the speed of the decisionmaking process, and it will still protect taxpayer confidentiality by removing any sensitive details from the published brief. Reasoned final offer dispute resolution is an option that can be interpreted as being available under the current MLI Part VI rules,19 especially when the MLI itself has a provision for reasoned awards under the modality of independent opinion arbitration(article 23(2)(c)).20 It is highly likely that in these cases(when arbitration is actually triggered) there is a difference over a specific principle rather than a difference in the taxable amount within the arm’s length range. Unless the decision is reasoned and published, it will not be possible for developing countries to become involved and establish a learning curve that will not only be reflected in future arbitrations but in audits and MAP proceedings as well. Besides, having the possibility of reasoned decisions also generates a necessary space for adjusting international tax standards to apply them to the realities of the developing world. In sum, the issue of transparency is key to ensuring that there will be accountability for the system and that future disputes may be prevented based on the principles contained in the published decisions. Mechanism design recommendations for G-24 countries ★ Accept final offer as the default mechanism for the Determination and Resolution Panels in order to preserve sovereignty and reduce costs. ★ Require reasoned decisions in order to increase transparency and establish a learning curve. ★ Demand the publication of redacted briefs for each Determination and Resolution Panel. Administrative body Although it may be easier for the OECD to act as exclusive administering body, given its more extensive personnel and funding, it is important to understand that the administrative body may have a significant influence in procedural decisions, including the appointment of the panels. The general philosophy of the organization may be reflected in these procedural decisions. This exclusivity may not be beneficial to developing countries, given the general alignment that the OECD Secretariat has shown 19 The Explanatory Statement to Article 23(1) of the MLI clarifies that paragraph 1“permits the competent authorities of the Contracting Jurisdictions to mutually agree on different rules, which may apply to all cases or to a particular case”. 20 Art. 23(2)(c): The arbitration decision shall be delivered to the competent authorities of the Contracting Jurisdictions in writing and shall indicate the sources of law relied upon and the reasoning which led to its result. The arbitration decision shall be adopted by a simple majority of the panel members. The arbitration decision shall have no precedential value.

15 with US and G7 views. It is important to leave open the possibility of nominating alternative administrative bodies in the future, such as the United Nations, especially for the Resolution Panels. In the meantime, G-24 countries should promote default procedural rules that are neutral and inclusive of developing country interests. Administering body recommendations for G-24 countries ★ Allow for the appointment of additional alternative administrating bodies, such as the United Nations, especially for the Resolution panels, even if presently they may lack capacity in terms of funding or personnel. ★ Ensure that procedural decisions are all established in the rules, and that any lack of agreement between potential parties will be resolved in a neutral way. Transparency and Publicity As a matter of principle, given the impact and the binding nature of the decision upon developing countries, it is necessary to establish an obligation to publish the assessment and decision made by each panel, including at least a brief description of the facts, the proposal made by the UPE, any opposition or dissent by members of the panel, and the rationale behind any decision made. In the rationale portion of the brief, G-24 countries could request a reference to how the developmental stage of the parties was taken into account in the procedure if the initial audit or adjustment was made by a developing country and it was deemed to be an“issue related to Amount A.” In cases where competing proposals were introduced by the parties, the principles supporting each proposal should be made public in the redacted brief, together with the reasons for choosing the prevalence of one principle over another in the specific context of the relevant facts and circumstances. It is important to note that at this point it is very likely that there will be a difference in the principle to be applied, or in the interpretation of the main principles that underly the arm’s length standard and the attribution of profits to permanent establishments. Naturally, it is in the interest of developing countries to understand how those conflicts are resolved and how the principles are actualized to conform to current ways of doing business. It might even be the case that, faced with a specific fact pattern, the panel will accept the developing country’s interpretation, as it is more likely to reflect the need to update the principle to the new ways of doing business. To protect taxpayer rights, the affected taxpayers must have full access to all documents produced in the procedure except when there are duly motivated reasons to limit habeas data. This may also reduce the number of instances in which taxpayers will decide to challenge the decision made by the determination panel and also the number of cases in which the challenge will prevail in a domestic court. The publication of redacted briefs is also important in view of the potential experience gap that developing countries may encounter given the limited number of MAP cases involving transfer pricing in many jurisdictions. Having an opportunity to study these documents will enhance the competent

16 authority’s participation in future panels and even in specific bilateral or multilateral transfer pricing MAP procedures. Having a transparent procedure and outcome facilitates training and capacity building, which are always at the core of G-24 countries’ interests. It is important to note that if no publication is provided, the independent members may sell their insights to taxpayers and countries that can afford their services, which may widen the knowledge gap between developing and developed countries. The publications concerned should at least include versions in Spanish and French as official languages of the Inclusive Framework. Publication is already provided for in the MLI Sample Arbitration Agreement at paragraph 6, as is clear from par. 32 of the Commentary to the Sample Arbitration Agreement. Of course, publication should be made in a redacted form in order to protect the identity of the taxpayer and any other sensitive commercial or industrial information. The decision to publish the panel outcome not only brings further transparency to the process, but also facilitates training and learning by tax administrations that have not had extensive MAP experience. Transparency recommendations for G-24 countries ★ Request the publication of redacted briefs for each panel decision including the facts, competing proposals, issues or principles involved, and the reason for choosing one or the other proposal. ★ Taxpayers must be given access to procedural information and should be allowed to request the redaction of specific information that may be sensitive from a business point of view. Enforcement, annulment, and interaction with domestic remedies Enforceability and domestic annulment of mandatory binding dispute resolution decisions constitute an area that G-24 countries must explore in depth, especially when constitutional rights may create a variety of scenarios in each country once a decision has been published and adopted by CAs via the MAP. The enforcement of MAP decisions is different in each G-24 country, and the only commonality is that all countries that have gone through an Action 14 Peer Review must have legal faculties to implement MAP decisions regardless of domestic time constraints. In some countries, including the United States, the taxpayer has a right to challenge any MAP decision in domestic courts as part of their constitutional right to access courts. In these cases, the decision will no longer be binding upon the countries that signed the MAP, but the resources spent in the Resolution Panel will be at least partially wasted. Naturally, a domestic judicial decision issued in a foreign country is not likely to be enforced in G-24 countries, so in these situations developing countries should be free to enforce the original adjustment issued by their tax authorities even after a diverging decision had been reached. For this reason, it is recommended to clarify that the Resolution Decision is only binding if the taxpayer does not challenge it domestically in either country. Furthermore, countries may be bound by prior judicial decisions that can prevent them from agreeing to a MAP that provides a solution that deviates from the said precedent. This is especially important in the cases where a decision has been made in respect of other taxpayers that may create a precedent

17 applicable to the case of the taxpayer in question. Because of this possibility, it seems necessary to request that parties to a Resolution Panel disclose any binding or influential judicial decisions before the panel is initiated. Otherwise, the resources spent in the dispute resolution mechanism may be wasted if one of the countries cannot agree to the MAP that adopts the decision. This disclosure also aids in the construction of a transparent dispute resolution mechanism that does not waste resources. Finally, if an international appeals mechanism is included, G-24 countries must ensure that the parties entitled to appeal and the circumstances for appeals are clearly defined in a way that balances the potential for each country to insist on a particular interpretation of the standards on the one hand with the costs and the time involved in participating in an international appeals mechanism on the other. Enforcement recommendations for G-24 countries ★ Clarify that countries are only bound by the Resolution Panel decision(or any panel decision, for that matter) if the taxpayer does not challenge the MAP that adopts the decision in either country. ★ Request that countries disclose any binding judicial decisions before a Resolution Panel is initiated. ★ Create rules for an international appeals mechanism that balances costs and opportunities to insist in a particular position. Capacity building The impending reality of mandatory binding dispute resolution in OECD and G20 countries must be acknowledged by G-24 members who are members of the BEPS Inclusive Framework. As the developed world starts to gain more experience, there will be increasing pressure to adopt mandatory binding dispute resolution as a minimum standard for all IF members21 regardless of their developmental stage. This means, in practice, that the elective mechanism and the withholding of the signature of the MLC may be temporary strategies to delay the implementation of the mechanism. The wisest strategy, then, is to take advantage of this time to prepare and build local capacity, starting with MAP. This, of course, is unnecessary for those G-24 members who are not members of the IF, have a limited treaty network, and have a long-term goal to remain that way. For G-24 countries that are in the IF, one of the obvious requests to make at the Inclusive Framework discussions is to create a capacity-building program for developing countries on both the MAP and on mandatory binding dispute resolution, that should include tax inspectors from other countries who have extensive experience in the negotiation of MAPs and arbitration. Both the UN and the South Center have training programs geared at improving developing country capacities in the MAP. The South Center program(still in its early stages) also provides South-to-South mentoring, so that G-24 21 See, for example, the replay of the public consultation meeting on the 2020 review of the Action 14 Minimum Standard, available at https://www.oecd.org/tax/beps/public-consultation-meeting-2020-review-beps-action-14.htm.(last accessed on 21.2.2022) Specifically, see the video at 2:20:40 and the following minutes.

18 Competent Authorities may about the issues discussed and the difficulties encountered when facing tax administrations with greater resources and more extensive experience. The Centro Interamericano de Administraciones Tributarias(Interamerican Tax Administration Center, or CIAT), on the other hand, is looking to establish a training program in tax dispute resolution, including mandatory binding mechanisms and a dispute resolution toolkit. Capacity building recommendations for G-24 countries ★ Request a specific capacity building program on mandatory binding dispute resolution for developing countries. ★ Train CAs in the MAP while the elective mechanism is still elective or if the country desires to expand its treaty network in the long term. Conclusions The tax certainty discussions happening at the BEPS Inclusive Framework point towards an advance in the trend to propose mandatory and binding solutions to international tax disputes, which stems from an increasing ambition to achieve uniformity in the substantive(and ever more complex) international tax standards approved in the framework of the BEPS Project. Ten of the G-24 member and observer countries have not joined the BEPS IF, while 23 are active members of the framework, therefore experiencing the increased pressure for enhanced international tax dispute resolution in spite of the mind-blowing complexity of the new rules. However, the difference now is that they have an opportunity to request the inclusion of design features that may contribute to building a system that takes into account the needs and circumstances of developing countries, regardless of whether they choose to join the MLC in the middle of 2022 or decide instead to pursue domestic measures to tax remote and digital businesses. As mentioned above, actively contributing to shaping the system does not imply a political compromise to join, as has been exemplified by the United States. Below is a list of some of the design contributions that the author believes may make the system more inclusive and even provide opportunities for G-24 countries to shape the system through reasoned decisions in ways that are not possible given the political constraints and consensus requirements at the IF: Cost-related strategies for G-24 countries ★ Propose virtual meetings and case accumulation for cost efficiency. ★ Propose common funding of total costs split proportionately to the GDP of participating jurisdictions in each mechanism. ★ Propose ceilings to the costs borne by developing countries expressed as a percentage of GDP. ★ Propose the creation of a developing country fund to be financed by developed countries and by a percentage of the amounts in discussion for each panel to be contributed by the levying country. ★ Take advantage of national legal defense agencies to use the resources to retain legal or accounting advice, or to bring an expert witness to the procedure.

19 ★ Demand clarity in the rules that allow for a Determination Panel or any other mechanism for appeals. Panel Composition recommendations for G-24 countries ★ Propose sole participation of CAs for the Review Panel. ★ Propose an intermediate independent CA stage where CAs from similar countries are called in to see if a solution may be jointly proposed by them. This stage would operate before independent experts are brought in, but only if both CAs consider that a set of independent CAs may reach an agreement ★ Accept independent members in the Review and Resolution Panel, as long as they comply with strict disclosure and conflict of interest rules. ★ Promote diversity in independent panel members by limiting non-CA panel members to a maximum of one panel per fiscal year and prevent the same panel member from sitting in a procedure involving the same taxpayer in a period of five years. ★ Put together lists of experts/panel members that either come from G-24/developing countries or fully understand their views. Mechanism design recommendations for G-24 countries ★ Accept final offer as the default mechanism for the Determination and Resolution Panels in order to preserve sovereignty and reduce costs. ★ Require reasoned decisions in order to increase transparency and establish a learning curve. ★ Demand the publication of redacted briefs for each Determination and Resolution Panel. Administering body recommendations for G-24 countries ★ Allow for the appointment of additional alternative administrating bodies, such as the United Nations, especially for the Resolution panels, even if presently they may lack capacity in terms of funding or personnel. ★ Ensure that the procedural decisions are all established in the rules, and that any lack of agreement between potential parties will be resolved in a neutral way. Transparency recommendations for G-24 countries ★ Request the publication of redacted briefs for each panel decision including the facts, competing proposals, issues or principles involved, and the reason for choosing one or the other proposal. ★ Taxpayers must be given access to procedural information and should be allowed to request the redaction of specific information that may be sensitive from a business point of view. ★ Enforcement recommendations for G-24 countries ★ Clarify that countries are only bound by the Resolution Panel decision(or any panel decision, for that matter) if the taxpayer does not challenge the MAP that adopts the decision in either country. ★ Request that countries disclose any binding judicial decisions before a Resolution Panel is initiated.

20 ★ Create rules for an international appeals mechanism that balances costs with the opportunity to insist on a particular position. Capacity building recommendations for G-24 countries ★ Request a specific capacity-building program on mandatory binding dispute resolution for developing countries. ★ Train CAs in MAP while the elective mechanism is still elective, for those countries who are members of the IF or if the country desires to expand its treaty network in the long term. The implementation of these options makes the Pillar One dispute prevention and resolution mechanisms more viable for G-24 countries, which is crucial given the trend to include similar mechanisms as a minimum standard in the framework of BEPS Action 14. Regardless of the political outcomes of the negotiations on Pillar One, the contribution of G-24 countries to the shaping of the international tax dispute prevention and resolution system is necessary to come closer to a fair and inclusive international tax system.

21 About G-24 The Intergovernmental Group of Twenty-Four on International Monetary Affairs and Development(G24) coordinates the position of developing countries on monetary and development issues in the deliberations and decisions of the Bretton Woods Institutions(BWI). In particular, the G-24 focuses on issues on the agendas of the International Monetary and Financial Committee(IMFC) and the Development Committee(DC) as well as in other relevant international fora. About FES The Friedrich-Ebert-Stiftung(FES) is a non-profit German foundation funded by the Government of the Federal Republic of Germany and headquartered in Bonn and Berlin. It was founded in 1925 and is named after Germany’s first democratically elected President, Friedrich Ebert. FES is committed to the advancement of both socio-political and economic development in the spirit of social democracy, through civic education, research, and international cooperation. Friedrich-Ebert-Stiftung is the oldest political foundation in Germany. Acknowledgments The author is grateful to Stef van Weeghel, Pasquale Pistone, Tommaso Faccio, Chetan Rao, and Carlos Protto, for their very useful comments to an earlier version of this paper. The author also wishes to thank the OECD Secretariat for the generous exchanges conducted in October, as well as Marilou Uy, David Rosenbloom, Juan Felipe Merizalde, Hans Mooij, Abdul Choudhury, Lisandro Junco, and Claudia Vargas for their valuable exchange of ideas. Any error or omission is mine alone. Cover photo: glass facades of modern office buildings and reflection of blue sky. © ahavelaar via Adobe Stock Imprint Friedrich-Ebert-Stiftung New York Office 747 Third Avenue, Suite 34D New York, NY 10017, USA +1-212-687-0208 info@fesny.org www.fesny.org Intergovernmental Group of Twenty-Four on International Monetary Affairs and Development(G-24) G-24 Secretariat 700 19th Street, NW HQ1 Mailstop 12-012 Washington, D.C. 20431, USA +1 202-623-6101| g24@g24.org www.g24.org Commercial use of this publication by the Friedrich-Ebert-Stiftung(FES) and the Intergovernmental Group of Twenty-Four on International Monetary Affairs and Development(G-24) is not permitted without the written consent of the FES and G-24.