Decoding Pillar One Amount A: Some Insights And A Q&A With ChatGPT

The OECD released a new report in October 2023 with the aim of providing more clarity to companies around the world on how to implement Pillar One, the international framework proposed by the OECD and signed by 138 countries, for how to allocate profits of companies, particularly digital companies across countries in which they operate. The idea is to create a uniform formula for calculating taxes that can be applied to make the process simple for companies to navigate. The Multilateral Convention is over 200 pages long, and the explanatory text supporting it, is over 600 pages. There are still many issues to resolve before countries sign onto it.

The revenue gains from this system as estimated by the OECD Economic Impact Analysis range between $10 to $30 billion, which is a tiny share of global corporate taxes. The US corporate tax revenues, for instance, in 2022 were $425 billion. A fundamental overhaul of the international system with the promise of raising maybe $10 to $20 billion of additional revenues makes one wonder whether the benefits outweigh the costs. In addition, as several people have pointed out, for the United States to ratify this, two-thirds of the Senate would have to approve it, and that may be hard given the current political dynamic in the US.

To understand the new rules, I asked ChatGPT the following questions using the Multilateral Convention document as the source [1]. I also reviewed the Economic Impact Assessment report to supplement the information in the MC based on the actual OECD analysis. It is clear that the rules are riddled with exceptions, don’t assume any behavioral responses from companies in response to these changes, and with the lack of simplicity, may potentially create an equally, if not more, complex system over time.

My Q&A With ChatGPT

Aparna Mathur (AM): Who is in scope?

ChatGPT: The ‘Multilateral Convention to Implement Amount A of Pillar One’ applies to multinational enterprises (MNEs) that meet certain revenue and profitability thresholds. Specifically, the Amount A rules apply to MNEs with global turnover above €20 billion and profitability above 10 per cent. These MNEs are referred to as "In-Scope MNE Groups" or "Covered Groups".

There are exceptions to who is in scope for Amount A. For example, the Amount A rules do not apply to MNEs that are regulated financial institutions, such as banks and insurance companies, unless they are part of a group that also includes non-regulated entities. Additionally, MNEs that are part of a "qualifying extractives group" are also excluded from the scope of Amount A. Finally, there are certain conditions under which a Group that was previously subject to a review by a scope review panel and was found not to be a Covered Group may not be considered a Covered Group in a subsequent Period.

AM Note: As several articles have pointed out, Pillar 1 applies to a small number of very large companies, and further applies only to a portion of their profits. As per the OECD Economic Impact Assessment, 106 MNEs are impacted by Amount A, and these are largely US based firms.

AM: What will happen to Digital Service Taxes?

ChatGPT: This new taxing right is intended to replace unilateral measures such as digital services taxes. However, the adoption and implementation of the Convention by individual countries will depend on their domestic legislative processes and political considerations. Therefore, it is possible that some countries may choose to retain their digital services taxes or similar measures, while others may choose to replace them with the new framework established by the Convention.

AM Note: The United States has pushed back against the imposition of DSTs by other countries, since these taxes are largely aimed at US multinationals and particular business lines. Further they apply to company tax revenues rather than profits, pulling even loss-making digital companies into the fray. Currently, about twenty countries have adopted some form of digital services taxes. When France imposed a DST of three per cent on US companies, the US considered retaliatory tariffs on French products like cheese, handbags and wine.

AM: How complicated is it to apply these rules?

ChatGPT: As an AI language model, I cannot provide a subjective opinion on how complicated it is to apply the rules. However, it is worth noting that the Multilateral Convention to Implement Amount A of Pillar One is a complex international tax agreement that involves a significant amount of technical detail and legal language. The Convention is designed to address the challenges of taxing the digital economy and to ensure that multinational enterprises pay their fair share of taxes in the countries where they operate. As such, it is likely that applying the rules will require a high level of expertise in international tax law and accounting.

AM: How will profits be allocated?

ChatGPT: Profits and taxes will be allocated under the Multilateral Convention to Implement Amount A of Pillar One using a two-step process. First, the profits of a multinational enterprise (MNE) will be divided into two categories: Amount A and Amount B. Amount A profits will be allocated to market jurisdictions where the MNE generates revenue, regardless of whether it has a physical presence in those jurisdictions. Amount B profits will be allocated to the MNE's countries of residence, based on factors such as where the MNE is headquartered and where it conducts research and development.

Second, the Amount A profits allocated to each market jurisdiction will be further divided among the countries in which the MNE has a taxable presence, based on a formula that takes into account factors such as sales, employees, and assets. The formula is designed to ensure that each country receives a fair share of the MNE's profits, based on its economic activity in that country. Once the Amount A profits have been allocated to each country, the Convention provides rules for determining the amount of tax that can be imposed on those profits in each country, in order to avoid double taxation and ensure that the MNE pays its fair share of taxes in each jurisdiction where it operates.

AM Note: This is essentially a destination-based tax that is likely to be complicated and lead to a significant compliance burden. Economist James Hines points out in a recent National Tax Journal webinar that this formulary allocation system may create incentives for companies to manipulate financial income and location of sales with strategic acquisitions and divestments. In other words, the behavioral responses to this profit allocation system may lead to further inefficiencies in the system and may further reduce revenue estimates. For instance, there is a sharp cutoff value for which companies are in scope for Amount A. Companies would have an incentive to report incomes just below this threshold. In addition, there may be an incentive to shift sales to lower tax jurisdictions through mergers with low profitability firms that have large sales in these countries. These incentives are magnified for the very small number of companies impacted by Amount A.

AM: How much revenue will Pillar 1 raise?

ChatGPT: Regarding your question, I'm sorry to inform you that the source does not contain information on how much revenue Pillar 1 will raise. The Multilateral Convention to Implement Amount A of Pillar One is designed to ensure that multinational enterprises pay their fair share of taxes in the countries where they operate, but the exact amount of revenue that will be raised will depend on a variety of factors, including the size and structure of the companies, the countries in which they operate, and the nature of their digital services.

AM Note: As per the OECD Economic Impact Assessment, the gains from Amount A range from $9.8 to $22.6 billion between 2017 and 2021, and $17.4 to $31.7 billion in 2021.

AM: Who is the final customer for digital companies?

ChatGPT: For digital companies that sell directly to consumers, the final customer would be the end-user or consumer who purchases the product or service directly from the company. However, for digital companies that sell to other businesses (business-to-business or B2B) or for which revenue is related to the location of the user of a digital service, the final destination of their sales may be more difficult to determine. In these cases, the MLC envisions the potential use by MNEs of various allocation keys based on macroeconomic indicators. However, the specific macroeconomic indicators are not mentioned in the text.

AM Note: The OECD Economic Impact Assessment shows that the allocation of profits based on final consumer definitions is complicated. In their analysis, allocation mechanisms differ based on the nature of the companies. 43 per cent of profit allocation comes from using data on sales by country for MNEs that sell directly to consumers. For companies relying largely on advertising revenue, 24 per cent of profits are allocated “based on a macroeconomics key in proportion to internet usage and final consumption expenditure”. For companies that are primarily dealing in components and B2B transactions, again profits are allocated based on a “macroeconomic key in proportion to GDP”. These allocation mechanisms suggest that aggregate metrics relating to the macroeconomy, rather than company specific metrics relating to actual sales, are going to play a big role in profit allocations. These metrics may have little correlation with actual profits earned by companies across different countries.

AM: Will developing countries gain from Pillar 1?

ChatGPT: Regarding your question, developing countries are likely to see more revenue gains under Pillar 1 of the Convention. This is because Pillar 1 aims to allocate a portion of the profits of multinational enterprises to the countries where they have market activity, even if they do not have a physical presence in those countries. Developing countries often have a large and growing digital market, but they may not have the resources to enforce their tax laws or to negotiate bilateral tax treaties with other countries. The Convention provides a multilateral solution that allows developing countries to participate in the allocation of profits and taxes, which could result in significant revenue gains for those countries. However, the exact amount of revenue gains will depend on a variety of factors, including the size and structure of the multinational enterprises, the countries in which they operate, and the nature of their digital services.

AM Note: The Economic Impact Assessment shows that there are several special provisions for developing countries.

• For jurisdictions with GDP under USD 40 billion, which includes all low-income jurisdictions in the Inclusive Framework, the nexus threshold to receive Amount A is lowered from EUR 1 million to EUR 250,000. This change increases the share of low-income (and middle income) jurisdictions that are able to receive allocated profit under Amount A due to the nexus threshold.
• (Other provisions in the MLC set aside up to 5 per cent of Amount A that MNEs may have difficulty sourcing to the final consumer, and distribute this share of Amount A to low-income and lower-middle-income jurisdictions exclusively on a pro-rata basis by GDP.
• The de minimis provisions largely exempt small and lower income jurisdictions from the obligation to provide double tax relief. Taken together, these provisions play a significant role in increasing low-income Inclusive Framework member revenue estimates from Amount A.

Based on the assumption of full Amount A implementation, the 2023 EIA Update estimates that low-income Inclusive Framework jurisdictions would gain 2.5 per cent to 3.0 per cent of CIT in 2021 and 1.5 per cent to 2.0 per cent of CIT on average over 2017 - 2021. Middle income jurisdictions would gain 1.2 per cent to 1.6 per cent of CIT in 2021 and gain 0.7 per cent to 1.1 per cent of CIT on average over 2017-2021. High income jurisdictions would gain 0.9 per cent to 1.4 per cent of CIT in 2021 and 0.5 per cent to 1.0 per cent of CIT on average over 2017 - 2021.

AM: How will it avoid double taxation?

ChatGPT: Regarding your question, the Convention includes a number of mechanisms for avoiding double taxation, which is the situation where the same income is subject to tax in more than one country. One such mechanism is the "tax sparing credit" provision, which allows a country to give credit for taxes that would have been paid in another country if not for a tax incentive granted by that country. Another mechanism is the "mutual agreement procedure," which allows countries to resolve disputes over the interpretation or application of the Convention through consultation and negotiation. The Convention also includes provisions for binding arbitration, which can be used to resolve disputes that cannot be resolved through the mutual agreement procedure. These mechanisms are designed to ensure that multinational enterprises are not subject to double taxation and that disputes are resolved in a timely and efficient manner.

Final Thoughts (And Questions)

While the new rules provide more clarity on how Pillar 1 may work, it is becoming increasingly clear that many countries are still not on board with the framework. The US Treasury has opened a period of consultation which means that it may receive additional input that may be at odds with what the OECD is proposing. Other countries, like Brazil, Colombia and India, are objecting to provisions relating to withholding taxes. In August 2023, Canada refused to agree to a moratorium on imposing new digital services taxes as part of the OECD’s Pillar 1 MLC and is moving ahead with its digital services tax. And even if all countries agree, we still don’t completely understand the economic incentives these provisions will create.

Ultimately, it comes down to what the motivation is behind the new rules. To create a fairer and simpler international tax system? Or siphon revenues from one jurisdiction to give to another? If the latter, it is hard to see how incentives across countries may ever align on the path forward.

1   The author tried Claude AI as well, but Claude is currently in Beta stage and could not read these large documents.

Aparna Mathur
Aparna Mathur is a former Senior Fellow at Harvard Kennedy School’s Mossavar-Rahmani Center for Business and Government where she is researching the US social safety net. She is a Visiting Fellow at FREOPP and a Senior Research Manager in Economics at Amazon. Aparna spent a year as a Senior Economist at the Council of Economic Advisers during the pandemic. She joined the Council as part of the COVID-19 response task force at the peak of the crisis in April 2020 and worked with epidemiologists on the health aspects of the crisis, while also tracking the economic downturn that came with the lockdowns. Prior to joining CEA, she was a resident scholar in economic policy studies at the American Enterprise Institute. At AEI, she directed the AEI-Brookings Project on Paid Family and Medical Leave, building bipartisan momentum on paid leave, for which she was recognized in the Politico 50 list for 2017. Her academic research has focused on income inequality and mobility, tax policy, labor markets and small businesses. She has published in several top scholarly journals including the Journal of Public Economics, the National Tax Journal and the Journal of Health Economics, testified several times before Congress and published numerous articles in the popular press on issues of policy relevance. Her work has been cited in leading news magazines such as the Economist, the New York Times, the Wall Street Journal and the Washington Post. She has regularly provided commentary on prominent radio and television shows such as NPR’s Marketplace and the Diane Rehm Show, as well as CNBC and C-SPAN. She has been an adjunct professor at Georgetown University’s McCourt School of Public Policy. She received her Ph.D. in economics from the University of Maryland, College Park in 2005, and is currently serving on the University of Maryland Economics Leadership Council. She is also on the Board of the National Academy of Social Insurance, Simply Green and the National Economists Club.