We commend the OECD’s 15-year effort since its 2005 publication of E-commerce: Transfer Pricing and Business Profits Taxation to address the challenges arising from the digitalization of multinational enterprises’ business models and the evolution of cross-border ecommerce. Our comments and recommendations are submitted in an academic capacity and do not represent an official statement or position of our respective employers, including Texas A&M University.
We additionally commend the members of the Inclusive Framework for the detailed technical framework for both Pillar One and Two and the focus on the need for tax certainty and avoiding double taxation. Beyond our comments and recommendations specific to the questions raised for public comment within the Reports on Pillars One and Pillar Two Blueprints, we stress the need for the OECD to expand its investigation of a unified approach for Pillar 1 to include a cost/benefit regulatory analysis of a withholding based system rooted in established substantive and procedural norms with the relief of double taxation provided via the established norms for foreign tax credits. Moreover, a withholding based system may better allow an amalgamated approach with source-based taxing rights under Pillar Two.
A significant challenge for Pillar One is the formation and implementation of a new international tax regime mired in a lack of institutional knowledge and capacity of resources for audit and MAP. A withholding based system may offer simplicity in contrast to the implementation of Pillar One, including (a) better procedural certainty for taxpayer and tax authority based upon current withholding regimes for services, (b) better revenue estimation for tax authorities, (c) less complex and expensive audits by tax authorities of taxpayers, (d) better tax risk management for taxpayers, (e) an established procedural system for relief of double taxation, and finally, (f) less cause for requiring MAP. We recommend that the OECD, via its Model Tax Treaty, shape the contours of the definition of the income and source issues, and the range of rates, to which digital services withholding tax may apply. This approach will allow the OECD to maintain the bedrock of the arm’s length approach for Article 7 and Article 9 of the OECD’s Model Tax Treaty and its Transfer Pricing Guidelines.
We attach our comments responsive to the questions for public comment of the Reports on Pillars One and Pillar Two Blueprints and welcome the opportunity for a representative of our signatories below to provide input at the public consultation to be held in January 2021.
On behalf of the International Tax Risk Management Curriculum of Texas A&M University School of Law
PILLAR ONE BLUEPRINT
Recommendation for a Study of a Withholding Tax Approach
The primary objective of the OECD Inclusive Framework to address tax challenges of digitalization of the economy is to establish tax equilibrium and capital neutrality for all stakeholders while respecting existing arm’s length principles. The Pillar One blueprint ‘Amount A’ proposes to establish a new international tax regime based on the doctrine of economic allegiance whereby a portion of the value of a digital economy value chain is allocated to the source countries. The new international tax regime requires (a) scope of income and expenses, (b) a new unitary allocation formulae for income and expenses within scope, (c) selection and definition of new nexus criteria, (d) new formulary apportionment for countries with a nexus claim, and (e) new multiparty MAP procedures. This current Pillar One approach neither offers simplicity nor addresses capacity. Moreover, several emerging economy stakeholders via the United Nations Committee of Tax Experts have questioned the responsive of the Pillar One approach for distributive fairness and sustainability as expressed by the UN’s Sustainable Development Goals of 2030.
Our recommendation for a study of a withholding tax approach aligns with the recognition that a score of countries have already enacted a withholding-based digital service tax with the likelihood of continued adoption among both OECD members and emerging economies. Broadly, the digital service withholding tax measures limit applicability to enterprises that exceed a global revenue threshold on a group reporting basis and that exceed a local country revenue threshold. Enterprises that exceed the thresholds and that generate income from digital services sourced within a country are then subject to a withholding tax. Digital services income in general includes advertising on a digital interface, digital marketplaces, and selling data. The withholding tax amount ranges in general from two percent to seven percent. These facets of the current digital service taxes either enacted or proposed represent a starting point of commonality for the OECD, in association with its Global Forum, and the UN International Tax Committee, to find agreement for uniformity that may then be reflected within the respective MTCs. The income potentially included for Pillar One’s Amount A may expand beyond digitally generated income which would not align with the objective of BEPS Action 1.
The Inclusive Framework’s BEPS Action 1 objective is to address the tax challenges from lack of physical presence arising from the digitalization of business. Pillar One has been evolving as a means to establish tax equilibrium and capital neutrality among the Inclusive Framework’s stakeholders while preserving the arm’s length principle codified in the OECD and UN Model tax treaties and respective transfer pricing guidelines. The carrot and stick of the Pillar One consensus-based solution is that business in-scope of the activities of Automated Digital Services (ADS) and Consumer Facing Businesses (CFB) will not also be subject to withholding-based digital service tax regimes among OECD members and Inclusive Framework members alike.
Yet, there appears to be an emerging consensus of another approach, at least among developing countries, that warrants a comparative economic impact study to inform a discussion among the OECD members: the recent inclusion within the United Nations’ Model Tax Treaty of Draft Article 12B “Income From Automated Digital Services” (hereafter “UN Art. 12B”).[1] UN Art. 12B attaches to automated digital services derived from a source state either a withholding tax on gross income, or at the election of the taxpayer, the standard domestic tax rate to the business’ qualified profits. Qualified profits are defined as a deemed amount of 30 percent of the business’ profitability ratio or the profitability ratio of its ADS segment, if available, applied to the gross annual revenue from ADS from the source state. Where the taxpayer is a member of a multinational group, the profitability ratio to be applied shall be that of the group or, if available, of the business segment of the group relating to ADS, in order to address concerns of transfer pricing related profit shifting.
A withholding based system offers an immediately implementable regime built on legacy systems and procedural simplicity. A withholding based system offers: (a) better procedural certainty for taxpayer and tax authority based upon current withholding regimes for services, (b) better revenue estimation for tax authorities, (c) less complex and expensive audits by tax authorities of taxpayers, (d) better tax risk management for taxpayers, (e) an established procedural system for relief of double taxation, and finally, (f) less cause for requiring MAP. Thus, in the context of a withholding system ameliorated by the OECD Model Tax Treaty, we recommend that the OECD investigate commonality for definitional income issues, source of digital income, and the range of applicable rates.
The OECD may support the formation of global consensus by expanding its discussion of Pillar One to include, with the OECD’s regulatory impact analysis approach, a withholding tax based option such as UN Art. 12B. The UN Art. 12B option is rooted in established substantive and procedural withholding norms with corresponding double taxation relief provided by foreign tax credits. A withholding based option may offer lower compliance costs in relation to its implementation, as well as (a) better procedural certainty for a taxpayer and a tax authority based upon current withholding regimes for service income, (b) better forecasting on a country basis for taxpayers and better revenue estimation for national tax authorities, (c) less complex and expensive audits by a tax authority of a taxpayer, (d) an established foreign tax credit system for relief of double taxation, (e) less cause for requiring the already stretched national resources necessary for Mutual Agreement Procedures (“MAP”) between nations, and finally, (f) less opportunity for corruption within the discretion of apportionment determination.
According to the OECD’s 2008 publication Building An Institutional Framework For Regulatory Impact Analysis, such analysis generally requires a comparative benefits/costs and economic impact analysis of alternatives be undertaken.[2] The OECD has presented an economic impact assessment of Pillar 1.[3] Thus, in alignment with the OECD’s goal of securing global consensus through transparent informed and discussion, we recommend that the OECD undertake and publish a regulatory impact analysis that includes a comparative benefits/costs and economic impact analysis of UN Art. 12B with Pillar One, any overlap with Pillar Two, and an OECD withholding tax based approach.
The activity test to define the scope of Amount A. Comments are invited on the design and implementation of the proposed activity test relating to Automated Digital Services and Consumer-Facing Businesses, including any challenges and suggestions on how to address them?
Pillar One evolved to address source country challenges for levying tax on digital service business models that are not adequately captured by the definition of permanent establishment. The inclusion within scope of taxation Automated Digital Services (“ADS”) business, and the more concise definition of ADS business with the positive and negative lists, achieves this goal. However, the inclusion of “consumer-facing businesses” (“CFB”) within the Pillar One approach is unnecessary. CFBs supply goods or services that are not digital in nature because ADS are pursuant to the 2020 Pillar One approach excluded from the scope of CFB. The OECD Model Tax Treaty and the Transfer Pricing Guidelines have for decades well-managed the taxation of the sale and leasing of cross-border traditional goods and services in conjunction with value-added tax and customs regimes. The forthcoming adoption of an economic presence definition test to augment the permanent establishment definition will bring income from the sale and leasing of cross-border (non-digital) goods and services[4] within the scope of Article 7 “Business profits” and the arm’s length approach that has evolved with the publication of the 2017 Transfer Pricing Guidelines. It will be superfluous to pull consumer purchased goods and services into Amount A.[5]
Moreover, the Pillar One Blueprint recognizes the ambiguous differentiation of which goods and services will fall in-scope activities that in turn will lead to additional compliance burden and controversy. The Pillar One suggested solution is that “the early tax certainty process will be available to provide clarification and ensure consistent treatment across the Inclusive Framework.” But such an early tax certainty process is not yet developed and even if agreed, will require years to develop capacity. It is recommended that CFBs for the time being be left out of the Pillar One Blueprint (and thus Amount A) and instead that they be addressed pursuant to an economic presence standard and the arm’s length standard for determining business profits.
If CFB remains in-scope for Amount A, then the definitional regime differentiating CFB from ADS should be clear and unequivocal as to allow proper understanding of which activities are in-scope. Clearer definitions are required as regards CFB since considerable uncertainty remains what types of activities are in-scope and this will trigger disputes. The justification for CFB inclusion is even arguable based on its reasoning for in scope is that they are able to participate in an active and sustained manner in the economic life of market jurisdictions beyond local presence, the blueprint fails to address how current existing ALP rules do not allow adequate taxation of CFB or why minor adjustments to existing rules could address the taxation of CFB. As it is, this Pillar One Blueprint appears to stretch beyond the concerns of digitalized businesses with its discussions of pharmaceutical, extractive industry products, energy products as examples.
If CFB remains in-scope for Amount A, then we recommend that, in the drive to reduce ambiguity, the OECD strengthens its approach of a “positive list” and a “negative list” that supplement its general definition. The general definition will lead to variations of inclusion as a natural outcome of scores of revenue authorities drafting regulations to embed the general definition into local law.
The design of a specific Amount A revenue threshold (in addition to a global revenue threshold) to exclude large MNEs that have a de minimis amount of foreign source in-scope revenue. More specifically, comments are invited on what would be the best approach to define and identify the domestic or home market of an MNE group (e.g., the residence of the ultimate parent entity).
At first glance, a revenue threshold will be helpful to avoid burden on small and medium enterprises (SMEs) that allegedly do not contribute to significant value creation in the market jurisdiction. Such a threshold will exclude the pool of 90 percent of SME taxpayers below it while securing 90 percent of the corporate tax base for Amount A redistribution among the tax administrations exercising a nexus claim. Approximately 9,000 taxpayers at the revenue threshold of 750 million Euro / 850 million U.S. dollars have implemented the CBCR reporting tool.[6]
However, we recommend a different approach than envisioned under paragraph 185, that the Amount A revenue threshold should be a phased-out approach over a five-year period based on the Pillar One 2020 Economic Impact Assessment.[7] By example, the OECD estimates that year 1 global residual profit in scope will be US$74 Billion regarding ADS (US$415 billion if including CFB) if the global revenue threshold is initially set at EUR5 billion at a 10 percent profitability threshold. In year 2, the threshold may be reduced to that which applies for Country-by-Country Reporting (CbCR), i.e. 750 million Euro, raising the scope for Amount A to US$81 billion for ADS and an additional US$413 if CFB income is included. Year 3 would drop the threshold to $500 million, then $100 million for year 4 after which it is phased out altogether. Year 5 would add at a 10 percent profitability threshold US$83 billion of ADS to Amount A (and $433 billion for CFB if included).
The phase out will reduce over the five years the tax regime ring fencing among business based on threshold until it is eliminated. Such threshold discrimination during the four-year period will, in general, advantage the smaller business in order to pay for the adoption of the compliance framework necessary to implement Pillar One for their ADS (and potentially CFB) profits.
For the purpose of determining a de minimis foreign source income exception for falling in-scope of Amount A, the identification of a domestic market based solely on the ultimate parent may produce distortive results jeopardizing MNEs with ADS and CFB located in small economies. Similarly, the domestic market based solely on where sales are primarily obtained may produce distortive results. Thus, we recommend that an election could be afforded the taxpayer. Alternatively, we recommend that a combination of factors could be used for determining home market (e.g. location of the ultimate parent combined with a percentage of in-scope revenue generated). Moreover, it is recommended that a Amount A revenue threshold be agreed to exclude business that has a de minimis amount of foreign source in-scope revenue.
Regarding an approach between the state of residence of the ultimate parent entity or the state of greatest economic presence for the MNE Group, consider if an MNE originates in a smaller state but expands its digital business to a larger market in a large state. The digital business operations and income in the larger state may be significantly larger than those in the original state where it began its operations. In this case, although the residence of the Ultimate Parent Entity would be in the smaller state, a much larger share of the business would derive from another state. It would be reasonable to identify the state where the largest share of this MNE group’s economic activity is located as the home market for the MNE’s allocation to its Amount A.
The “plus factors” suggested for CFB will be examined as potential indicators which denote an engagement with the market beyond the mere conclusion of sales. In terms of compliance costs and administrability, do you have any comments on these plus factors?
Under paragraph 204, the plus factors serve as indicators that the MNE is having a connection to activities connected to in-scope sales. The suggested plus factors have been deemed to be either physical presence or a targeted advertising and promotion activity.
We recommend the implementation of the “plus factors” with a physical presence requirement supplementing a sales threshold for CFB to reduce compliance costs and enhance administrability for both MNEs and tax authorities, aligning with the nexus premise of ‘sustained and significant manner in the economic life of market jurisdictions’. These two indicators should be sufficient to establish taxing rights rather than requiring that MNEs track and allocate costs of advertising and promotion activity that could be deemed as “connected” to the sales. Moreover, this recommended approach can be more easily evidenced when challenged by tax authorities because the MNE is already tracking its physical presence, and the sales and expenses attaching thereto, via its internal and, when required, external country-by-country report.
Do you consider the suggested plus factors (and hence a taxable nexus under Amount A) could be deemed to exist once a certain level of sales is exceeded? If so, what should be the criteria for establishing such level?
If the “plus factor” approach is included in the final report, then we recommend that the OECD consider a “plus factor” based on the number of transactions in a specified jurisdiction which would be administratively easier for an MNE to source to the point of sale for tracking and reporting purposes. For example, if the company is participating in a large amount of targeted marketing and that marketing is having a substantial impact on the local jurisdiction then the sales revenue should reach above the threshold specified, otherwise the deemed “connected” activity does not appear to have an net-positive impact on the market jurisdiction.
We recommend that the criteria for establishing a level of sales should be simplified to reduce the administrative burden and complexity that additional factors may add, for example, context of the GDP of a country. A five-year phased-in local market threshold from high to low should be sufficient to allow MNEs and tax authorities to develop systems for ADS and if applicable CFB compliance and verification thereof. But if criteria for establishing a level of sales are included in the final report, then we recommend measurement of the country’s economy via its gross national income (GNI) as published by the World Bank.[8]
In respect of the SDG 2030 goals of the United Nations, developing countries may be eligible for a reduced threshold. To simplify the criteria, the thresholds could be set into several tiers such low, medium, and high based on the country’s GNI. For example, if the country’s GNI is considered low, then its threshold could be X% of the average GNI of the low-income category; if the country is considered medium then the threshold would be set at X% of the average GNI of the medium category countries. The OECD list of DAC countries which are eligible to receive official development assistance.[9] These consist of all low- and middle-income countries based on gross national income (GNI) per capita and includes the least developed countries (LDCs). We recommend that the OECD leverage its current list method to determine the countries that should be eligible for a reduced threshold for nexus.
Should the market revenue threshold contain a temporal requirement of more than one year? If so, what should it be?
To simplify reporting and enable consistency for typical reporting periods, we recommend that the revenue threshold be calculated on an annual basis. However, we recommend the exclusion from nexus for isolated or one-off transactions to prevent nexus being permanently triggered. For example, if the MNE had documented that a specific, nonrecurring transaction caused nexus and it was reasonable to believe that the MNE would not continue to incur nexus, then an waiver should be applied, assessed on a case by case basis by the tax authority for compliance verification. While this approach causes additional compliance on the MNE, it better aligns to the purpose of Amount A.
Do you have any comments with respect to the proposed sourcing rule and proposed hierarchy of indicators as the basis for the sourcing of revenue for Amount A?
The Blueprint proposed revenue sourcing scheme designates the most reliable indicator as the fallback indicator. We recommend that it would be more administrable and more effective if the most reliable indicator is the default indicator. If an indicator is already the most reliable, then any other indicator is suboptimal reliable. Moreover, tracking a variety of indicators is burdensome on compliance.
Having a principles-based hierarchy based on the most reliable indicator allows necessary flexibility that is not possible in a rules-based hierarchy. The current blueprint version contains a rules-based hierarchy listing specific indicators. This rules-based hierarchy is susceptible to rigidity and obsoletion. Technological changes may render specific indicators obsolete and others less reliable which means that the hierarchy may require periodic, perhaps often, modification. Yet such regulatory modifications are in general not as swift as the evolution of the digital economy, evidenced by this very BEPS Action 1 project.
A principles-based standard based on the most reliable indicator is flexible enough to adapt to evolving technology and business. As technology evolves, the most reliable indicator would change. If the standard is based on the most reliable indicator, then the MNE will be afforded the agility to modify its approach to implement the new most reliable indicator so that its recordkeeping more accurately reflects the source revenue. A principles-based hierarchy may have as examples of a most reliable indicator:
- a user’s GPS data or IP address;
- any available indicator that appears reasonably reliable, such as a user’s billing address or telephone country code; or
- the MNE’s educated conjecture of a user’s actual location, based on the available information.
We note that the Blueprint does not indicate how to treat ADS distributed to consumers in jurisdictions where the particular product is non-existent. Consider the following example. HEB stores are physically present in the United States and Mexico, but not Canada. HEB purchases from an MNE a digital ad package that covers North America. The digital ads will reach consumers in Canada even though HEB has no physical presence there. We question whether Canada’s tax authority should exercise a claim for source taxation due to the digital ads for the sole reason that the GPS and IP addresses in Canada are being logged by the digital provider.
We note that the European Union’s VAT regime includes rules for determining the location of the customer. In order to leverage compliance systems already in place for MNEs, the EU VAT rules may provide the most reliable indicator for MNEs with EU country nexus.
What factors should be taken into account in determining “reasonable steps” required to obtain information that is unavailable (such as changing contracts with third party distributors)?
We note that requiring that an MNE change contracts with third party distributors so that third party distributors are obligated to provide anonymized customer information seems to be a burdensome, even unrealistic, step. We recommend that the factors for considering what steps are reasonable for obtaining information should include the available technology, customers’ consent to have their information tracked, and ability to obtain information with minimal intrusion into personal privacy. The MNE or distributor is generally required by domestic law to obtain user consent in order to track the user’s data, and to notify the user of the possible uses of their data. Moreover, many national legal systems either have rules about, or evaluating rules about, personal privacy protection in the context of data and collectors of that data via digital business models. Thus, we recommend that any accessed data be limited to a necessary minimum for the MNE to accurately track sourcing information. Also, before the data is accessed by any business beyond the distributor, the data should be anonymized. Again, we point to the ability to leverage VAT compliance systems already in place for many MNEs that take into account such data issues.
What simplification measures, if any, should be considered in the revenue sourcing rules, such as safe harbors or de minimis rules?
We note that the Blueprint’s revenue sourcing scheme envisions a complex multi-indicator test for ADS. The test varies depending on which category the ADS fall under for which there are nine different categories of ADS businesses. Also, some categories distinguish between seller-based and purchaser-based sourcing. While we appreciate the attempt to delineate among ADS business models, we think that such complexity will pose an unnecessary compliance burden versus its benefits of specificity. We recommend consolidating the number of categories to a manageable level that will still promote the objective. For example, ADS businesses can be consolidated into five categories:
- digital content services,
- online advertising services,
- sale or other use of user data,
- online intermediation services, and
- hybrid digital services.
The Blueprint’s revenue sourcing scheme for CFB is not as complicated as the scheme for ADS. Thus, we do not consider the need for simplification in the treatment of CFB is not as urgent.
As a safe harbor, certain location data that have stood the test of time should be granted the presumption of reliability. GPS data or IP addresses that are consistent for 183 days in one year should be presumed reliable. Also, billing addresses that have allowed the MNE to successfully receive payment for two consecutive years should be presumed reliable. Because of the time period involved, such location data has a high probability of indicating a person’s habitual location.
Do you consider that VPNs and/or any other emerging technology may have an impact on the accuracy and/or reliability of proposed revenue sourcing rules? If yes, what options or design changes should be considered to eliminate or minimize such an impact.
We note that the use of virtual private networks (VPN) poses a dilemma, albeit the degree of such use by consumers is unknown to us. Consumers may use VPN’s for legitimate reasons, such as protecting online privacy, preventing identity theft, and work-at-home situations. However, VPN’s may compromise the reliability of revenue sourcing by altering a person’s location from one country to another. While governments may have an interest in regulating the use of VPNs for CFB, the effectiveness of such regulation is doubtful. Thus, we recommend further study regarding the necessity of practicality of potential safeguards such as VPN detection and targeted VPN blocking. Incidentally, governments may consider such safeguards from the perspective of ameliorating Internet-facilitated crimes, including digital piracy, hacking and stalking.
Do you consider that hallmarks drawing on IAS 14 constitute an appropriate basis for developing a test to determine whether an MNE group is required to segment? If not, what other options should be considered to identify relevant segments for Amount A purposes?
We recommend that segmenting should be optional. We think that the Amount A tax base should be computed on a group basis and an exception be provided that allows MNE Groups to segment if the segmentation hallmarks are consistent with IAS 14.
Do you consider that existing segments (under financial accounting standards) should be used in the majority of cases as a basis for segmenting the Amount A tax base (for example by using a rebuttable presumption)? If not, what other options should be considered?
We recommend that using existing segments under financial accounting standards is the best measure commonly available that is consistent with the objectives of Pillar 1 while avoiding elevating the cost of compliance or administrability for the MNE. We note that audited financial information is expensive and that it does not offer significant benefit over the financial accounts other than verification of the numbers by an audit firm which is not the tax authority. Yet, if audited information is available, then we recommend a taxpayer election to leverage it if applicable for the taxpayer’s determination of Amount A.
Do you consider that groups should be permitted to calculate Amount A on a geographically segmented basis? If so, what should be the criteria for determining when geographical segmentation is permitted and what those geographic segments should be?
We recommend an election to provide an option to allow an MNE that is geographically managed and/or has the requisite financial accounts available to calculate Amount A on a geographically segmented basis. We suggest that the OECD perform an economic impact assessment to identify which industries would be eligible for geographic segmentation.
Alternatively, do you consider that MNE groups should be required or permitted in some cases to segment their profits before tax between in-scope activities (i.e. ADS and/or CFB) and out-of-scope activities? If yes, what criteria could be used to determine when this approach to segmentation should be applied as opposed to calculating the Amount A tax base on a consolidated basis?
We recommend that an MNE be permitted to elect to segment its profits between in-scope activities and those out of scope to best align with the goals of Pillar One and Amount A. We note that such process may be cost and administratively prohibitive for MNE’s when the financial information is not already readily available and the out of scope activities are insignificant.
Do you consider that the proposed mechanism to eliminate double taxation from Amount A will have an impact on the scope and relevance of possible double counting issues?
We note the risk of double taxation in specific circumstances of decentralized business models or relatively autonomous businesses that have multiple local entrepreneurial activities. These entities already recognize and incur tax on a significant amount of the profit in the market jurisdictions that should be excluded from the Amount A to avoid the incumbent double taxation.
Do you consider that there is an interaction between withholding taxes in market jurisdictions and the taxes under Amount A? If so, how could such interactions, including double counting issues, be addressed?
We note the incidence for double taxation when a market jurisdiction levies withholding tax on an ADS payment when the gross ADS payment is allocated as Amount A. We recommend that Amount A should exclude income that has attracted withholding tax to also avoid potential double counting in case of any divergence in approaches for countries adopting the UN approach.
Should a domestic-to-domestic business exemption be considered to exclude part of a group’s business that is primarily carried on in a single jurisdiction from the calculation of the Amount A tax base? If so, do you have views on how this exemption could be designed?
We recommend that a standalone domestic business should be excluded from the scope of Amount A to avoid the incumbent double taxation of including its domestically taxed profits.
The development of a loss carry-forward regime that would ensure that Amount A is based on an appropriate measure of net profit. Do you consider that Amount A tax base rules should apply consistently at the level of the MNE group (or segment where relevant) irrespective of whether the outcome is a profit or loss (symmetry)? What is your view of the proposal to extend the carry-forward regime to ‘profit shortfalls’?
We note that start-ups and companies with high capital and intangible investment will generate losses in their first years after establishing a business. We recommend that the Amount A tax base rules, irrespective of outcome, be applied consistently to avoid unintended distortions and controversy. Moreover, we recommend that any deficit between the Amount A deemed residual profit and the Amount A actual profits be, for sake of consistency and in alignment with the Pillar One goals, be deemed a loss for purpose of Amount A determination. The purpose of Amount A is to subject a business’ deemed residual profit to tax. In turn, a deficit of such residual profit should be factored into the business’ ability to generate extraordinary profit, by example, due to substantial investment or due to a highly competitive market. By example, a business that underperforms with only an eight percent margin the deemed threshold of 10 percent would incur for Amount A deemed loss equivalent of the two percent deficit.
We recommend that Amount A allow pre-Pillar One implementation regime losses (e.g. because of the global recession caused by COVID-19) to ensure that Amount A actually reflects the ability to pay principle by taxing net profits. Excluding pre-Pillar One loss will hamper growth and does not allow a proper reflection of actual profits.
We recommend that losses should be allowed as an offset against future years Amount A residual. Losses should be allowed to be carried over indefinitely against the group Amount A, or alternatively, at least 20 years, to offset losses resulting from the pandemic, research and development, and market entry costs. The recognition of a group-level loss carryforward account and an earn-out mechanism is both reasonable and relatively administrable.
What do you consider will be the key challenges in the early tax certainty process described in the Blueprint and how do you think would they best be addressed?
We note that the key challenge will be the complexity and burden imposed on the MNEs to accurately delineate business lines, calculate segmented profits, perform the Blueprint’s 10-step process, and the identification of paying entities. As an alternative, the UN Committee of Tax Experts has proposed a 3-step process. We recommend that the OECD undertake a regulatory cost-benefit analysis of proposals and of a withholding regime approach presented in the introduction of this Pillar One section that takes into account leveraging regimes and compliance systems already in place such as VAT and digital service tax withholding.
PILLAR TWO BLUEPRINT
Do you foresee any other technical implications of GILTI co-existence – in addition to those already identified in the Blueprint that should be taken into account?
A key element to GILTI resulted from the final regulations under Internal Revenue Code (“IRC”) §951A, specifically the high-tax exclusion (“HTE”) rule under Treasury Regulations §1.951A-2. The HTE rule exempts foreign tested income from taxation in cases where such income is subjected to tax at a rate equal or greater than 90 percent of the current U.S. rate (i.e., currently 21 percent * .9 = 18.9 percent). Notably, the rule is implemented to foreign income on a “tested entity by tested entity” basis. While there is an exception for income blending between entities within the same jurisdiction, in implementation the rule essentially equates to a CFC by CFC basis. Importantly, the HTE hurdle determines whether the foreign CFC’s tested income should be subjected to GILTI but does not define the related tax. This determination subsequently takes place when the tested income of those entities not meeting the HTE are then combined (along with GILTI loss entities) and the actual GILTI tax is calculated at the consolidated level. Overall, the complexities of the calculation particularly in light of the Pillar II proposals raise several issues.
As mentioned in the blueprint, a primary consideration is an inability to carryforward or carryback of GILTI losses and taxes. While GILTI losses cannot be carried to other years, they can be used to offset the GILTI income of other entities in the consolidated calculation. Any differences between the two regimes may require complex calculations including the need to track the treatment differences. The same would apply to any differences in the carryforward of taxes. Also, note that the taxes related to the tested income of those entities that meet the HTE are then disallowed for GILTI deduction or credit.
Another issue is the interaction of GILTI with the complexities of the foreign tax credit (“FTC”) rules. The FTC, of course, prevents taxpayers from being subjected to double taxation. The FTC affiliation rules are one key element in determining a taxpayer’s ability to use an FTC. Overall, these rules involve the segregation of foreign income into one of several categories including GILTI, passive, and foreign branch classifications. In order to determine the income in each separate category, these rules allocate certain key deductions (e.g., interest, research, and development) amongst the numerous baskets. To the extent that the calculation results in a domestic or foreign loss, these losses are allocated against other income thus triggering complex recapture rules the effects of which impact calculations in future years. Moreover, while certain U.S. affiliates of a foreign entity may not be required to file as a consolidated group, these FTC affiliation rules are applied at the affiliate group level based on IRC §1504, adding a layer of complexity. Any differences between the GILTI and Pillar 2 regimes will inevitably create differences in the tax base from which tax and FTC relief is determined, increasing the likelihood of taxpayer’s being subjected to additional tax while not being able to leverage a domestic mechanism for double taxation relief.
What are the interactions between GILTI and the GloBE rules that would need to be coordinated and how should they be coordinated?
We recommend the grandfathering of the GILTI regime to simplify administration. First, adopting a high-tax mechanism would ensure that foreign income is subjected to a sufficient amount of tax. Second, for those entities not meeting the high tax threshold, mirroring the “use it or lose it” (no carryover) rules related to both GILTI taxes and losses would simplify compliance for both the taxpayer and tax authorities as well as avoid the need for separate calculations based on two sets of equally complex rules. Moreover, double taxation as a result of the tax and the FTC is based on two separate regimes would be mitigated.
Considering that the GloBE rules only protect the tax neutrality of investment funds that are at the top of an MNE Group’s ownership chain, are there specific situations in which the GloBE rules do not adequately protect the tax neutrality of investment funds?
As mentioned in the Pillar Two Blueprint, a primary objective of the treatment of investment funds under the GloBE rules is to protect the tax neutrality of investment funds. However, in the United States, the income taxation of foreign investment funds/companies is both highly complex and is often an administrative burden. For example, U.S. taxpayers that own an interest in a foreign investment fund/company can be subject to the US GILTI rules and/or the U.S. PFIC (Passive Foreign Investment Company) rules that generally result in the current taxation of the income generated by the foreign investment fund/company. In addition, the use of transparent entities can create subtle compliance differences to the U.S. GILTI or U.S. PFIC rules, but generally, also result in the current taxation of the income generated by the underlying foreign investment fund/company that passes through to a transparent entity for U.S. tax purposes in the case of U.S. taxpayers. This may result in a different result than what is desired by the Pillar Two Blueprint.
In the case of an investment fund under the control of an MNE Group, what additional rules would be needed to ensure the tax neutrality of the fund and ensure that: i. The MNE Group’s share of the fund’s income is not excluded from the GloBE tax base? and ii. Related party payments to and from the fund cannot be used to circumvent the UTPR?
In the United States, foreign investment funds can be subject to U.S. income tax under the GILTI rules or PFIC rules. These two regimes typically result in immediate taxation for the U.S. investor of the income generated by the investment company generated by its underlying investments. If this income is subject to U.S. income tax and also subject to a GloBE tax, then we recommend relief from double taxation be afforded either through an exemption from GloBE or through a foreign income tax credit. In addition, a backup withholding tax may be required under local law associated with earnings generated by the investment fund/investment company which may be allowable as a foreign tax credit for the benefit of the MNE. In these situations, we recommend that if the local withholding tax rate is at or above the required GloBE backup withholding rate, no further backup withholding under GloBE is required. If the local withholding rate is less than the required GloBE backup withholding rate, then the difference would be the only additional backup withholding required under GloBE.
Calculating the ETR under the GloBE Rules?
As mentioned in the Pillar Two Blueprint, IFRS and many national GAAPs reflect the significant similarity in their general principles and conceptual frameworks. Yet there are some differences in IFRS and for example, the U.S. GAAP that affect the timing of recognition of an item of income or expense. Given the burden that would be associated with identifying and tracking any required adjustments and the fact that consistency within a taxpayer group is more important than consistency across taxpayers, we recommend that there should be a clear guideline on the requirements for adjustments to the accounting standard used in preparing the consolidated global financial statements in an attempt to reconcile any differences between the financial accounting standards.
Moreover, we recommend that the OECD limit required adjustments to financial statement income to reflect tax policy considerations to align financial statement income with taxable income concepts. To the extent that some limited adjustments are required because of material differences between accounting and tax policy, these should be clearly identified, and the required adjustments should be based on the tax accounting standard and tax policies applicable in the jurisdiction of the group’s parent company. Adjustment for Pillar One outcomes applies before the determination of income subject to Pillar Two. Depending upon the final design of Pillar One, an adjustment may be required to the GloBE tax base to properly reflect Pillar One outcomes. The new GloBE system should include guidance identifying specific areas where adjustments may be made.
We recommend a simplification measure to reduce compliance costs while aligning to the purpose of the GloBE rules. If a jurisdictional ETR based on an MNE’s CbC report is above a de minimis threshold, which may be set above the agreed minimum GloBE rate, then no further GloBE determination would be required for that jurisdiction. The filing of the CbC report would meet the requirements for that jurisdiction for purposes of its GloBE regime. The benefit of a relative (national) threshold approach is that it establishes a ceiling on the number of jurisdictional ETR calculations for an MNE. Still. MNEs would be required to establish the necessary processes and systems in every jurisdiction in order to compute the base year ETR. In order to apply the simplification measure, MNEs would still be required to compute the pre-tax profit for every jurisdiction; though, MNEs would only need to compute covered taxes for jurisdictions with profits above the de minimis threshold.
We recommend that an MNE may elect to compute the base year ETR based on covered taxes and income data from multiple consecutive years rather than a single year to safeguard against inappropriate exemptions. Yet this approach requires that the period would need to extend for several years in order to provide material simplification. On the other hand, the longer the grace period the more likely it is to hide or overlook inaccuracies and distortions. We suggest that a multi-year approach be studied.
For “low-risk jurisdictions”, tax administrations that were following a risk-based approach would eventually stop auditing an MNE’s ETR for operations in low-tax jurisdictions. Yet, we recommend a low-risk re-determination for tax law revisions or reforms that materially change the jurisdiction’s tax base and/or tax rate. The determination of “low-risk” could apply to all MNEs operating within a certain jurisdiction, or it could be restricted to MNEs within or outside certain sectors. It may be the case that in a certain jurisdiction virtually every MNE is likely to be above the agreed minimum rate, except MNEs in a certain sector because of sector-specific tax incentives. In such a case, the low-risk determination could apply to every MNE, except for MNEs in the particular sector. In the case of a sectoral approach, consideration would be required for the treatment of firms that operate across multiple sectors.
Do you have any views on the appropriate ownership threshold and the methodology of how to determine that threshold, both for the exclusion of portfolio dividends and the exclusion for gains and losses on the disposition of stock from the GloBE tax base?
We recommend a consistent threshold of 10 percent to mirror portfolio dividend exemption rules to avoid cumbersome expense allocation.
The treatment of re-organizations?
Our recommendation is to ensure consistency with local nonrecognition domestic rules to avoid risk of over-taxation under GloBE.
Cover Taxes?
The Pillar Two Blueprint excludes digital taxes from GloBE covered tax as these are generally designed to apply to the gross revenues from the provision of digital services and as such would not be considered an income tax. Furthermore, franchise tax which are based on gross receipts are not considered an income tax under the laws of certain jurisdictions and would also not be considered a covered tax. MNE’s with this type of tax might be at a disadvantage when computing the GloBE ETR compared to MNE’s based in jurisdictions that solely focus on net income tax. We note that the covered tax section is not very clear on various taxes MNEs paid, and as to why these are not covered, even if they affect the MNE’s bottom line.
We note with interest the Blueprint proposal for an imputation credit regime for GloBE which is designed as a protection for double taxation to shareholders under the same jurisdiction. However, the imputation credit regime allows for a refund of taxes be paid in respect to the distributions made to non-residents. Under GloBE this credit arises from taxes paid with respect to distributions of a constituent’s net income. The Blueprint’s proposed change from the receiving entity claiming the tax credit to the paying/ distributing entity claiming the credit requires more guidance. For an MNE with many non-resident shareholders by example, how would the Blueprint proposal affect the shareholders if the tax credit is not available to offset taxes generated by the income received? There should be clarification as to whether the distributions are paid gross rather than net of tax because the paying entity will use the refund of taxes for payments to non-residents to reduce covered taxes on the year of the distribution.
Tax Base Determination?
We note that the Pillar Two Blueprint proposal replacing accounting depreciation with tax depreciation rules in each jurisdiction would introduce administrative complexity for MNEs. The guidance is unclear how this approach would be applied. By example, would each entity be required to reduce the value of its assets to match the tax value? Is the requirement effective from a specific date or will there be a grace period to comply with the transition to tax, and if so, over how many years? Allowing accelerated tax depreciation is an incentive by a country to promote foreign direct investment. The unintended consequence of the Blueprint proposal’s change may be the reduction of FDI which is already at its lowest since the 2008 recession because the foreign risk-reward benefit will be reduced. The additional determination to adjust for permanent and temporary differences will create an administratively burdensome task for an MNE’s tax and its accounting staff that must continue to adhere to the tax depreciation rules for domestic purposes. We recommend that such proposal be framed as a taxpayer election, or alternatively, that a sufficient timeline is granted to comply with the new requirements.
Do you consider that the categories of covered payments and the exclusion for low-return payments ensures that the STTR focuses on the transactions that present significant BEPS risks?
We note significant reservations about the implementation of the Subject to Tax Rule (“STTR”). Nevertheless, we recommend that if an STTR is ultimately promulgated, there should be an exclusion for low-return payments. In terms of categories that ensure that the STTR focuses on the transactions that present significant BEPS risks, the following could achieve the goals:
- Interest and royalties,
- Franchise fees,
- Insurance or reinsurance premiums,
- Guarantee, brokerage or financing fees,
- Rent or any other payment for the use of or the right to use intangible property.
We consider that the last category of payment proposed by the Blueprint, for an amount paid to or retained by the payee that is a consideration for the supply of marketing, procurement, agency or other intermediary services, is too broad and will lend itself to overreach by tax authorities. Thus, we recommend excluding it.
We recommend that a STTR should have a narrow scope with defined criteria that avoids ambiguity, potential controversy, and prospective litigation.
Do you have any views on the design and practical application of this rule component as well as potential simplifications?
We note two general concerns in terms of the design aspect. First, we note the withholding strategy on covered payments. While the commentary mentions the “significant tax burden” and the (contrariness) to the policy behind the Subject to Tax Rule (i.e., defined transactions that give rise to significant BEPS concerns), it does not outright dismiss a mandatory withholding. Unsurprisingly, there is already a discussion on what level of mandatory withholding percentage should be required. A primary concern is an impact this would put on certain MNE entities. Although members of an MNE, many entities are in less economically advanced jurisdictions with more limited economic resources. This is especially true in times of monetary restrictions and when payments are due in different (especially more valuable) currencies. One current example is when an Argentine entity, the currency of which is Argentine Peso (a hyperinflationary currency) has a U.S. MNE parent to which all payments would be required to made in U.S. dollars. Notably, with its hyperinflationary currency, Argentina’s Central Bank has enacted a series of actions to control capital within the country to include restrictions on foreign currency and foreign currency transactions.
The second concern with covered payments is the basis of the STTR, especially in relation to a mandatory element. If a Subject to Tax Rule is included in the final proposal, then we recommend limiting both the trigger rate and the amount of top-up tax under the STTR to a rate that is lower than the minimum effective tax rate set under the income inclusion rule and undertaxed payments rule in order to limit the risk of over-taxation.
What are your views on including a materiality threshold?
The inclusion of a materiality threshold is extremely important in the implementation and administration of the STTR. The most effective and efficient method that is consistent with addressing BEPS risks and the goals of Pillar 2 is the threshold based on the size of the MNE Group. Because the Income Inclusion Rule and the Undertaxed Payments Rules apply the EUR 750 million threshold adopted by the Inclusive Framework under BEPS Action 13 (Country by Country Reporting), we recommend to main this threshold for consistency.
Would such a threshold simplify the administration of the rule and limit compliance costs in a material way?
We note that large MNEs already comply with CbCR rules and have consolidated financial statements readily available that will make determination of whether the MNE is within the scope of the rules more cost effective and less burdensome than having to provide different financial reports and perform more calculations. Also, as stated within the Pillar Two Blueprint, the MNE Groups that are within the scope of the CbCR rules earn 90 percent of the global corporate revenues, preserving the effectiveness of the rules.
Do you have any views on the different approaches suggested for the materiality threshold as well as on their application in isolation or combination?
We agree with the OECD’s approach as it relates to having the consolidated revenue threshold ultimately match the methodology currently used under BEPS Action 13 for Country by Country Reporting purposes. However, we recommend that an undertaking of the magnitude of the implementation of Pillar II compliance on a worldwide basis warrants a phase-in period for MNE groups of different revenue levels. Thus. we recommend that the scope of Pillar II be modified to provide a phase-in period of 4 years using the following scale:
Year 1 – Subject to Tax Rules apply to MNE Groups with gross worldwide revenues in excess of €1 billion or equivalent.
Year 2 – Subject to Tax Rules apply to MNE Groups with gross worldwide revenues in excess of €900 million or equivalent.
Year 3 – Subject to Tax Rules apply to MNE Groups with gross worldwide revenues in excess of €800 million or equivalent.
Year 4 – Subject to Tax Rules apply to MNE Groups with gross worldwide revenues in excess of €750 million or equivalent.
Further technical work will be undertaken in the Inclusive Framework on administrative approaches that could deliver these aims. This will include work on (i) applying the top-up tax as an ex-post annualized charge, (ii) a certification system providing for reduced rates of withholding tax, and (iii) the application of contingent withholding taxes set at a level that would generally result in an annual ex-post balancing payment by the taxpayer (rather than a repayment). Which administrative approach do you consider to be the most suitable?
We recommend that the low interim withholding with balancing payment to be the most suitable. This is the most balanced method proposed for taxpayers and tax administrations in terms of cash flows. It also has the greatest degree of predictability of liabilities due to the ascertainable nominal rates of taxation. Finally, because it is based on a lower rate of withholding it minimizes the impact of an over withholding on payments situation that can arise when a payment is initially thought to be within the scope of the STTR but after an ex-post examination of the facts is found not to be within the scope.
[1] The final language of UN Model Tax Treaty Article 12B and its commentary that will include minority perspectives will be agreed at the 22nd session of the Committee of Experts on International Cooperation in Tax Matters scheduled for April 2021. The currently adopted draft language of Article 12B of the 21st session is available at: https://www.un.org/development/desa/financing/sites/www.un.org.development.desa.financing/files/2020-10/CITCM%2021%20CRP.41_Digitalization%2010102020%20Final%20A.pdf (accessed Nov. 15, 2020).
[2] OECD (2008), Building An Institutional Framework For Regulatory Impact Analysis, OECD Publishing, Paris, https://www.oecd.org/regreform/regulatory-policy/40984990.pdf (last visited November 30, 2020).
[3] OECD (2020), Tax Challenges Arising from Digitalisation – Economic Impact Assessment: Inclusive Framework on BEPS, OECD/G20 Base Erosion and Profit Shifting Project, OECD Publishing, Paris, https://doi.org/10.1787/0e3cc2d4-en (last visited November 30, 2020).
[4] The exemplary goods and services that are proposed to be included in-scope when purchased by a consumer are pharmaceuticals, franchising /use of resellers, licensing, dual use goods / services, and dual use intermediate products and components.
[5] By example, a justification for including the income from prescription pharmaceuticals sales to consumers in Amount A, however infrequent, is the previous tax haven planning of pharmaceutical companies regarding income associated with their patents. Yet, such income going forward will be taxed at a minimum rate under the Pillar Two approach and is already included in the income of U.S. shareholders of controlled foreign corporations within the scope of the U.S. GILTI regime.
[6] See OECD, BEPS, Action 13: Country-by-Country Reporting available at https://www.oecd.org/tax/beps/beps-actions/action13/ (last visited November 30, 2020).
[7] OECD (2020), Tax Challenges Arising from Digitalisation – Economic Impact Assessment: Inclusive Framework on BEPS, OECD/G20 Base Erosion and Profit Shifting Project, OECD Publishing, Paris, https://doi.org/10.1787/0e3cc2d4-en (last visited December 5, 2020).
[8] See World Bank GNI by country at https://data.worldbank.org/indicator/NY.GNP.MKTP.CN (last visited December 10, 2020).
[9] See http://www.oecd.org/dac/financing-sustainable-development/development-finance-standards/DAC-List-of-ODA-Recipients-for-reporting-2020-flows.pdf (last visited December 10, 2020).
________________________
To make sure you do not miss out on regular updates from the Kluwer International Tax Blog, please subscribe here.
