Fiscal Reform in the Digital Era: Implementation of OECD Pillar One

Task Force 1: Macroeconomics, Trade, and Livelihoods: Policy Coherence and International Coordination

Workstream: Monetary Synchronisation and Fiscal Space and Trade, Investment, and Supply Chain Resilience

Globalisation and digitalisation have boosted economic activity while simultaneously posing challenges to international taxation. To solve the tax challenges arising from digitalisation, OECD has initiated Pillar One, to reformulate profit allocation rules by taxing companies in jurisdictions where they generate profits. Pillar One is estimated to increase tax revenue global income by US$220 billion, which is relatively low compared to the legal impacts it brought to developing states. Continuous global coordination is necessary to ensure that the rules are swiftly implemented. Regular technical assistance by the G20 and OECD, as well as enhancement of regional cooperation, are crucial to develop a common understanding of Pillar One implementation, ease compliance burdens, and mitigate high administrative costs.

1. The Challenge

In its July 2022 form, OECD adopted a firm stance to implement Pillar One^[1] as an overlay to existing international tax regime. The multilateral convention has been designed to deviate from the profit allocation rules contained in tax treaties and transfer pricing regulations. Base Erosion and Profit Sharing (BEPS) Action 1 was conceived to deal with the inability of existing rules to include digital economy as a tax subject. Following the consequences of coordinated efforts in aligning information asymmetries and combating tax avoidance, OECD has proposed a multilateral agreement, the enactment of which would affect the enforcement of domestic and international tax laws.

According to OECD,^[2] Pillar One is expected to reallocate more than US$200 billion taxing rights to market jurisdictions annually. The impact is equal to around US$12-25 billion tax revenue gains per year (over the period 2017-2021). The impact estimation is higher than the prior calculation in 2020^[3] as a result of design changes and most recent data. Besides the higher value of impact, Pillar One implementation under the new procedural design documented in Amount A is also expected to provide more profit allocations to low- and middle-income jurisdictions rather than high-income jurisdictions as a share of existing CIT revenues. Low- and middle-income jurisdictions benefit from special nexus thresholds, tail-end revenue provisions, and de minimis rules (e.g., for elimination of double taxation (EoDT)).^[4]

Apart from the estimated positive economic impact, the implementation of Pillar One may face challenges from the administration of the rules. It will be difficult to achieve the purpose and objective of Pillar One rules without a set of procedural rules that respect the legal traditions and accounting principles of different states. Therefore, three documents containing Pillar One procedural rules have been proposed by OECD, namely: (i) the Administration Framework for Amount A; (ii) the Tax Certainty Framework for Amount A; and (iii) tax certainty for issues related to Amount A.^[5] Work is still in progress for the collection of Amount B as well as for the development of a Common Documentation Package.

The foremost issue with the proposed Pillar One procedural rules is that, from the perspective of market state jurisdictions, new administrative tax obligations will be imposed on non-residents. Besides, given a consensus reached in October 2021, Pillar One rules will not be exclusively imposed on digital services companies, but rather, on “large and highly profitable groups” (i.e., the in-scope group).^[6] This means that the rules will govern companies that are already physically present in a market jurisdiction as well as those that are not. The territoriality principle is, therefore, under revision.

Within substantive income tax law, territoriality is one of the overarching principles in international tax law and establishes that a state can only have a legitimate tax claim over income when the income recipient resides in its territory or when the income is sourced from activities within its territory, or both. Using this line of argument, a non-resident may be taxed by a state only when the income derived or received by such person is, by virtue of domestic source rules, sourced from that state. In such a case, the person will be taxed under the withholding tax regime of that state. Tax treaty rules may restrict tax rates but not the criteria for withholding taxes. For a century, non-residents have not been liable for tax filing obligations in source states unless they are permanently established there. Now, however, in direct contrast with this reality, the Amount A tax return neither requires residencies nor permanent activities in the state to impose tax payment obligations.

a. Non-resident taxpayers’ liability for Amount A tax return

Under the proposed administrative framework of Pillar One, an in-scope multinational company against which the new Amount A tax is imposed will have to appoint one or more of its group entities to act as the taxpayer (OECD accepts both the ‘single taxpayer’ and the ‘multi taxpayer’ approaches). The problem of the Amount A tax reform is that it presupposes a broadening of the meaning of the territoriality principle in taxation.

Conceptually, withholding taxes (performed by resident income payers) will replace tax return obligations. Usually, non-residents are not entitled to the net calculation of income and do not have to file tax returns. However, if the permanent representative of a foreign establishment exists, such a representative will be liable to submit tax returns on behalf of its non-resident carrier. In practice, the broadened meaning of territoriality means that there will be a physical presence of non-residents in the jurisdiction imposing tax return obligations.

Usually, tax returns are not final. Tax return documents serve as an entry-point to subsequent tax procedures, including tax audits, issuance of tax notice, tax penalties, tax litigation, and tax recovery. Enforcements of these procedures require the physical presence of the taxpayer in the jurisdiction. The territoriality principle does not warrant a state with pro justitia power over a person in another state. However, in the case of a tax return imposed on digital services companies, since there is no physical presence in the market state, the tax return will be final, which is unorthodox in taxation.

An Administrative Framework and a Tax Certainty Framework are aimed at preventing tax disputes. The calculation of Amount A will be coordinated by the tax official in the region where the ultimate parent company resides (i.e., Lead Tax Administration), and market states will be informed of the amount of residual profits to be taxed under their domestic laws. The Tax Certainty Framework further seeks to guarantee that this amount has been ‘approved’ by resident states which are responsible for providing double-tax relief from the double taxation arising from the calculation. In this regard, market states’ tax sovereignty is reduced. Their tax base is predetermined through the coordinated filing, while their power to issue tax notice and impose tax penalties are practically stripped.

b. Resident taxpayers’ liability for an ‘inclusive’ income tax return

The Administrative Framework provides market states with the flexibility to “tax Amount A income in any manner they deem appropriate.”^[7] States have the freedom to “self-determine either tax Amount A income under their current income tax regimes or determine a separately levied income tax.”^[8] These proportions will be established taking into account the fact that an in-scope multinational has member entities in the market state. However, a ‘streamlined compliance’ procedure presupposes that a secondary obligation may be imposed.

In the case where an in-scope multinational company has member entities in different market states, the streamlined compliance procedure would not be enforceable. Rather, each entity in each market jurisdiction must include the calculation of Amount A into their tax returns. Consequently, all domestic rules on income tax assessments (i.e., filings and penalties) may be imposed on those entities.

Prima facie, the procedures to be adopted would support efficiency on behalf of in-scope multinationals, as no additional tax return is imposed. As a baseline, the streamlined compliance procedure assumes that the Amount A tax return and Common Documentation Package are ready to be distributed to market jurisdictions within 15 months after the end of the fiscal period. Payments of taxes pursuant to Pillar One rules must be done within 18 months after the end of the period. Meanwhile, states applying the self-assessment system would require that payment obligations are fulfilled before or at the same time as the submission of tax returns. In Indonesia, for example, companies have 120 days after the end of the fiscal year to fulfil these obligations.

Given time constraints, there may be errors in calculation, resulting in underpayment of taxes. Non-streamlined procedures also grant market states’ jurisdictions with the Amount A tax return and Common Documentation Package from the Leading Tax Administration 15 months after the end of the fiscal period. This could be disadvantageous for taxpayers, as they might be liable to pay for the underpaid taxes, and the penalties therefrom, to the market states treasury.

Having said that, in the case of the ineligibility of streamlined compliance procedure, it is only logical to extend the filing deadline to a longer date. OECD has made it explicit that “the circumstances where [the streamlined compliance] does not apply should benefit taxpayers.”^[9] However, such extension would lead to procedural tax discrimination, as other taxpayers in the same jurisdiction would have to comply with the 120-day rule. However, the vertical equality argument is invalid, as there are not substantive differences between these taxpayers. All these companies are taxpayers of the same tax regime. Additional income tax liability shall not justify preferable procedural tax treatment.

c. Use of new proxies in Amount A tax collection

In addressing the above challenges, OECD has introduced proxies in tax collection. This is indicated by the conceptualisation of a Lead Tax Administration and the Single Taxpayer Approach, as well as the Multiple Taxpayer Approach. A Lead Tax Administration functions as a sanctioning authority to whom the Amount A tax return and Common Documentation Package are submitted by in-scope multinationals and distributed to market jurisdictions. Meanwhile, the latter approaches presuppose that the traditional single-entity approach in profit allocation system would bring about immense challenges to market jurisdictions in exercising their new taxing rights. Therefore, by appointing a taxpayer that would be liable to pay taxes and submit tax returns (hence the name ‘single taxpayer approach’), market jurisdictions would focus their tax pursuits on that taxpayer, with the benefits of pursuing all entities within an in-scope multinational. By the same token, in the multiple taxpayer approach, an agent entity will be the contact point for the market jurisdiction in enforcing Amount A compliance of an in-scope multinational in that jurisdiction. As a contact point, such entity does not have tax liability, nor does it have reporting obligations in that jurisdiction.

Prior to these new proxies proposed by OECD, the tax collection system in many states has relied on withholding tax agents. These agents are not only valuable in domestic transactions but also cross-border transactions. The withholding tax agent is always a resident of the imposing jurisdiction. It is this very feature that allows tax administration the ability to impose penalties and perform tax recovery procedures. A withholding agent may be held liable for underpayment of taxes and the penalties therefrom. If the agent fails to fulfil that obligation, a tax recovery procedure can be deployed by seizing its assets.

The physical presence requirement may not be altered by any unprincipled measure. The use of new proxies serves as a practical solution. In proposing the multiple taxpayer approach, OECD reinstates the consistency of the tax agent principle known worldwide. However, the lack of physical presence in the market jurisdiction means that such an agent can only perform government-related functions and not as an agent with its own liabilities. Meanwhile, as concerns the single taxpayer approach, OECD admitted that the approach requires “a novel approach to the elimination of double taxation that would deem an entitlement to relief for Group Entities that have no corresponding tax liability.”^[10] Again, physical presence is key to sound tax principles.

The most problematic proxy within the Pillar One Administrative Framework may be the Lead Tax Administration. The concept raises many questions. For example, which democratic power allows a state to predetermine the tax base of another state? What are the guarantees that the administration would perform its functions in a fair manner? Whom does a state seek for legal aid in the event of unfair calculation? In such a highly politicised regime as Pillar One, the appointment of an unwatched supranational tax administration is questionable because it may exacerbate the divergent interests among different states in discussing taxation of the digital economy. It is likely that the Lead Tax Administration proxy will be rejected; if states accept this concept, tax sovereignty will fade away, eventually disappearing as a central concept in taxation.

2. The G20’s Role

As an intergovernmental institution dealing with international taxation within the development of the digital economy, the G20 plays an important role in coordinating the related states to achieve a consensus on Pillar One and conclude the repeated delays in Pillar One implementation. The platform of the G20 can push the agenda of meeting the consensus to implement Pillar One among the Inclusive Framework jurisdictions. The Inclusive Framework was launched in June 2012, following a request by the G20 leaders to identify the key issues that lead to BEPS.^[11] As the initiator of the Inclusive Framework, the G20, led by OECD, can accelerate the next steps to implement Pillar One. Moreover, as it includes OECD and non-OECD members, the G20 will be key to guarantee that measures developed by OECD will have substantive relevance and be practical for non-OECD members.

Besides strengthening coordination among Inclusive Framework jurisdictions, the G20 must stress that the sovereignty of each state is preserved. This is not the case when imposing tax returns on non-residents without physical presence in a market state. Even when physical presence is established, it is not acceptable that a super-national law dictates the fashion in which a sovereign state imposes tax procedures on its residents, nor it is acceptable to appoint the tax officials of certain states as proxies in the tax calculation of other states.

3. Recommendations for the G20

While model rules are currently being fine-tuned by OECD, the enforcement of Pillar One procedural rules is just around the corner. OECD is expecting the Inclusive Framework jurisdictions to reach an agreement on the Multilateral Convention (MLC) in mid-2023 and for Pillar One to enter into force in 2024. Such expectations, however, may not materialise in every jurisdiction, particularly in those with their own Digital Service Tax (DST). All the Inclusive Framework jurisdictions that have contracted the MLC and want to gain their benefits from Pillar One are supposed to withdraw any DSTs and other relevant or similar measures to tax the digital activities conducted within their jurisdictions.^[12] Once they join the MLC, the jurisdictions will be unable to enact any new unilateral or multilateral measures to tax digital activity in the future.

The challenge is that there is no clear indication of how Pillar One implementation can be more effective than DSTs. Moreover, the delays in Pillar One implementation have raised concerns among jurisdictions seeking to safeguard their tax income generated by digital activity. In this case, the G20 can step into the Pillar One agenda to provide more specific and clearer next steps for the agreement.

If the agenda to implement Pillar One meets expectations, the G20 may push for coordination among Inclusive Framework jurisdictions to ensure that the Pillar One rules are swiftly implemented. Close coordination between the G20 and OECD is required to establish the final rules, following the comments and feedback received on the MLC draft published by OECD in December 2022. The adjustment of Pillar One rules is expected to address the issue of preservation of tax sovereignty of each state. Given the time constraints in enforcing the Pillar One rules, regular technical assistance by the G20 and OECD, as well as the optimisation of regional cooperation, are crucial, given that Pillar One implementation will substantially affect existing domestic tax laws in some jurisdictions. OECD should continue^[13] to bring the two-pillar solutions agenda in the G20 Finance Ministers and Central Bank Governors (FMCBG) under the Indian Presidency 2023. These efforts would help develop a common understanding of Pillar One implementation, ease compliance burdens, and mitigate high administrative costs.

On the other hand, in the most likely worst-case scenario, if they fail to reach an agreement on Pillar One implementation, jurisdictions may adopt their own unilateral or multilateral measures in DSTs to obtain their taxing rights over MNEs engaged in digital activities. In such circumstances, the objective of developing a unilateral and standardised form of DST under Pillar One will fail, since each state can apply its own measures. There will be no bilateral or multilateral agreement between states that undertake DSTs or other relevant measures where the affected MNEs are based, which may prove a hurdle in this case. Moreover, various objects will be taxed under the different DSTs implemented in each state, adding complexity to global MNEs. More worrisome is that the absence of coordination and bilateral or multilateral agreements will result in possible double taxation. To minimise the potential negative consequences of this scenario, under the finance track, the G20 should focus on formulating technical procedures and steps to perform less distortive DSTs or other related measures. Standardised capacity building is required for each state promoted by the G20. Lastly, the G20 may explore the potential role of regional cooperation to assist jurisdictions in developing their digital tax policy by establishing a platform or forum for providing information and knowledge related to DSTs and other relevant measures.

Attribution: Adrianto Dwi Nugroho et al., “Fiscal Reform in the Digital Era: Implementation of OECD Pillar One,” T20 Policy Brief, June 2023.