By Leslie Prescott-Haar & Sophie Day at TP EQuilibrium | AustralAsia LP
Following a meeting of the OECD’s Inclusive Framework members, the OECD on 12 October released its Pillar One and Pillar Two ‘blueprints’, which reflect significant proposals to reform the international tax system.
These blueprints have involved a massive amount of technical work by the various contributors and international leaders from the OECD.
This article covers the Pillar One blueprint, a detailed 230-page document, which addresses the tax challenges arising from the digitalisation of the economy.
The blueprint is designed to provide a foundation for future agreements across jurisdictions.
The Pillar One proposals reflect a structural shift in international taxation and transfer pricing through a reallocation of both taxing rights and non-routine profits, together with fixed routine returns for in-market marketing and distribution, in favour of ‘market jurisdictions’ for digital services and consumer-facing businesses based on a series of standardised rules detached from the arm’s length principle that are reasonably complex.
The scope of these Pillar One rules is a key issue for international consensus at this stage.
Arguably, implementation of a parallel system of international taxation based on government-imposed profit allocations for only certain taxpayers seems punitive, inequitable, and lacking the appreciation that most businesses are digital to some extent in 2020, and markets drive financial returns across industries.
Pillar One represents the OECD’s response to the recent proliferation of unilateral digital taxes that have been introduced and/or contemplated by various nations.
Digital taxes arguably expose multinationals to double or multiple taxation and excessive administrative costs through fragmented systems.
In October, the US IRS issued proposed regulations confirming the non-availability of foreign tax credits for digital taxes, increasing pressure for enhanced tax certainty through an extensive multilateral approach.
Notwithstanding these issues, the complexity of the Pillar One proposals may advance the fragmented digital taxes.
Elective Pillar One rules, which encompass both approaches and all multinational taxpayers, may be preferable.
How did we get here?
In 2001, the OECD’s Committee on Fiscal Affairs originally agreed to amendments to its Model Tax Convention Commentary regarding the conditions that, under international tax treaties, would determine a country’s right to tax profits from electronic commerce.
The main elements of the amendments confirmed that: a web site cannot, in itself, constitute a permanent establishment; a website-hosting arrangement typically does not result in a permanent establishment for the enterprise that carries on business through that web site; an internet service provider generally would not constitute a dependent agent of another enterprise so as to constitute a permanent establishment of that enterprise; and for a place housing computer equipment, such as a server, to constitute a permanent establishment, it is also required that the functions performed at that place be significant as well as an essential or core part of the business activity of the enterprise.
The OECD’s press release noted that these amendments “marked an important clarification in the fiscal environment for e-commerce which was likely to assist in its continued growth.”
Oh, how times and technologies have changed!
These tax challenges associated with massively expanded e-commerce were identified as one of the main focus areas of the OECD/G20 base erosion and profit shifting (BEPS) project, leading to the 2015 BEPS Action 1 report.
The Action 1 report found that the whole economy was digitalising and, as a result, it would be difficult, if not impossible, to ring-fence the digital economy.
In March 2018, the Inclusive Framework issued “Tax Challenges Arising from Digitalisation – Interim Report 2018,” which recognised the necessity for a global solution.
Then on 31 January 2020, the OECD released a “Statement by the OECD/G20 Inclusive Framework on BEPS on the Two-Pillar Approach to Address the Tax Challenges Arising from the Digitisation of the Economy”, which outlined the conceptual framework for Pillars One and Two.
Are we there yet?
As indicated, the blueprint details the proposed features of Pillar One, which in essence expand the taxing rights of ‘market jurisdictions’ in respect of residual profits where there is an active and sustained participation of the multinational in that jurisdiction’s economy through activities in, or remotely directed at, that jurisdiction.
Conversely, the taxing rights of jurisdictions where residual profits are presently allocated under existing rules would be reduced. Hence, hubs and havens will likely be most significantly impacted.
The Pillar One blueprint also aims to limit compliance and administrative burdens, by making the Pillar One rules as simple as possible, including through simplification measures such as a fixed return for baseline marketing and distribution.
This simplification measure should be positive, so that multinationals can then invest compliance costs savings in higher wages, R&D, environmentally-friendly processes, etc. However, the compliance costs for multinationals associated with addressing the complexities of Pillar One will exceed any savings gained from such measures.
Pillar One further aims to significantly improve tax certainty by introducing innovative dispute prevention and resolution mechanisms. However, the necessity for government involvement in and approvals for profit allocations in respect of business transactions is neither efficient nor effective administration, given the pace of change in this globalised world.
At present, bilateral advanced pricing agreements (APA) and other mutual agreement procedure (MAP) cases can take years to agree with revenue authorities, and changes and shocks are typically problematic under APAs.
The level of resources necessary within revenue authorities to implement and manage the Pillar One rules, together with all other taxation regimes, will likely involve high administrative costs for governments.
However, global APA or MAP agreements for multinationals would be helpful, where requested by multinationals; the blueprint’s proposed dispute prevention mechanisms take this direction.
The Pillar One blueprint does not reflect consensus views from the Inclusive Framework members and accordingly, the blueprint recognises that there are a number of ‘open issues’ relating to key features that must be resolved through political decisions, as well as further technical work in relation to a myriad of additional Pillar One aspects.
Proposed Scope
The blueprint provides for two categories of activities as within the scope of the new taxing right created by Pillar One: automated digital services and consumer facing businesses, but as noted, an agreement has not yet been reached on these in-scope categories.
Some Inclusive Framework members have advocated for a phased implementation, with automated digital services coming first and consumer facing businesses following later.
The US has proposed that Pillar One be implemented on a ‘safe harbour’ basis as part of an overall agreement that abolishes and prohibits digital services taxes and similar unilateral measures.
Given the OECD’s acknowledgment in its Action 1 Report of the difficulties in ring-fencing the digital economy, it is inconsistent that the proposed Pillar One scope seeks to draw such lines.
Presumably, more industries would fall within the ambit of Pillar One over time.
Automated digital services
Automated digital services are services provided on an automated and standardised basis to a large and global customer and user base and provided remotely to customers in markets with little or no local infrastructure.
The definition of automated digital services is comprised of a positive list and a negative list.
Positive list activities include online search engines, social media platforms, online intermediation platforms, digital content streaming, online gaming, cloud computing services, and online advertising services.
Meanwhile, negative list activities include customised professional services, online teaching services, online sale of goods and services other than automated digital services, revenue from the sale of a physical goods irrespective of network connectivity, and services providing access to the internet.
Consumer facing businesses
Consumer facing businesses are those that generate revenue from the sale of goods and services typically sold to consumers, including those selling indirectly to intermediaries and by way of franchising and licensing.
Consumer facing businesses activities include personal computing products (e.g., software, home appliances, mobile phones); clothes, toiletries, cosmetics, luxury goods; branded foods and refreshments; franchise models, such as licensing arrangements involving the restaurant and hotel sector; and automobiles.
Segmentation complexities arise for multinationals with multiple business lines, both in and out of scope.
For various reasons, a variety of industry sectors have been specifically carved out at this stage.
For example, the blueprint carves out natural resources (including non-renewable extractives such as petroleum and minerals, renewables such as agricultural, fishery and forestry products and renewable energy products and similar energy products such as biofuels, biogas, green hydrogen); most activities in the financial services sector (including banking, insurance and asset management); international airlines and shipping businesses; construction, sale, and leasing of residential property.
With respect to the pharmaceutical industry, to the extent that medical devices are products of a type commonly sold to consumers, these will be in scope.
However, there is still a decision to be made as to whether pharmaceuticals (i.e., drugs) will be in scope.
Proposed thresholds
The blueprint provides that the new taxing right would operate within certain thresholds.
These thresholds are intended to minimise compliance costs to multinationals and facilitate administration of the new rules for tax administrations.
However, is it fair to subject only major multinationals to various international tax rules? In our experience, taxation and transfer pricing problems are more common with other taxpayers, as major multinationals often adopt conservative policies and positions to avoid disputes, reputation issues, etc.
The revenue threshold would be based on annual consolidated group revenue, coupled with a de minimis foreign in-scope revenue carve-out.
The EUR 750 million revenue threshold presently applied for the purposes of country-by-country reporting rules would likely be adopted.
The de minimis foreign in-scope revenue threshold would apply to multinationals that exceed the revenue threshold, but only have a de minimis amount of foreign source in-scope revenue, which would be based on an absolute threshold amount, rather than being relative to the size of a given multinational’s domestic business.
This de minimis foreign in-scope revenue threshold would be applied under two steps: first, determine whether the multinational earned more than the de minimis foreign in-scope revenue threshold amount from automated digital services or consumer facing businesses activities, and then determine whether the multinational earned more than the de minimis foreign in-scope revenue threshold amount from foreign in-scope activities.
To avoid tax administrations being overwhelmed with the initial operation of the new taxing right, these thresholds may be implemented on a phased approach, such as starting with higher thresholds that could be either gradually reduced over time, or reduced after a post-implementation review has been undertaken.
Proposed new nexus rules
Scoping rules covering automated digital services and consumer facing businesses are intended to capture those businesses that make profits from active and sustained interactions with customers and users in a market jurisdiction, through activities extending beyond the mere conclusion of sales.
The new nexus rules would provide taxing rights to market jurisdictions in relation to an allocation of Amount A residual profit only. Without such nexus, nil profits would be allocated to that jurisdiction under Amount A.
Under the Pillar One blueprint, the new nexus rules could apply differently for automated digital services or consumer facing businesses.
For automated digital service, exceeding an in-scope market revenue threshold would be the only test to establish nexus.
Automated digital services are typically provided remotely, and such businesses generally have an active and sustained engagement with the market without physical presence, which is one of the key challenges in taxing the digitalising economy.
The extent to which consumer facing businesses, on the other hand, may participate remotely in a market jurisdiction would typically be limited given the physical distribution necessary for consumer products.
Accordingly, a higher nexus standard for consumer facing businesses should apply given additional possible complexity and compliance costs associated with sourcing revenue derived by consumer facing businesses (e.g., third party distribution) and taking into account that profit margins are typically lower for consumer facing businesses as compared with automated digital service.
In this regard, the blueprint indicates that for consumer facing businesses, a higher in-scope market revenue threshold would apply, coupled with a “plus factor” to indicate an active and sustained engagement with the market such as a subsidiary or a “fixed place of business”, carrying out activities that are connected to in-scope sales.
Under the blueprint, the new nexus rules would operate on a stand-alone basis to limit any unintended spill-over effects on existing tax or non-tax rules.
Pillar One details
The key elements of Pillar One are a new taxing right for market jurisdictions over a share of residual profit allocated as Amount A; a fixed return for certain ‘baseline’ marketing and distribution activities taking place physically in a market jurisdiction allocated as Amount B; and dispute prevention and resolution mechanisms processes, as outlined below.
Proposed revenue sourcing rules would apply for purposes of determining the revenue that would be treated as derived from a particular market jurisdiction.
These detailed revenue sourcing rules for automated digital service and consumer facing businesses would be relevant in relation to the scope rules; the new nexus rules; and the Amount A formula discussed below.
The new taxing right – Amount A
As indicated, Amount A represents a newly-established taxing right over a share of the residual profit of multinationals that fall within its defined scope.
The OECD has indicated that international consensus exists, in principle, for this new taxing right. The new nexus rules would apply to identify market jurisdictions eligible to receive Amount A residual profit allocations, subject to the scope and threshold rules.
Given the residual profit-basis for Amount A allocations, underlying issues relating to standardised measures of financial accounting profits and profits before tax; book-to-tax adjustments; segmentation of financial accounts, on either a business line or geographic basis; and loss carry-forward are significant, and may potentially be insurmountable in practice.
Systems and data limitations within multinationals may further complicate matters.
Under the blueprint, eligible market jurisdictions would receive a portion of residual profit using the following three-step formula:
Step 1: A ‘profitability threshold’ to isolate the residual profit subject to reallocation, and limit double counting between Amount A and the remuneration of routine and other activities. To avoid complexity, this threshold would be based on a profit before tax to revenue ratio; hence, the Pillar One rules would determine allocable residual profits, rather than the arm’s length principle. The formula would apply to the Amount A tax base after the deduction of loss carry-overs.
Step 2: A ‘reallocation percentage’ to identify an appropriate share of residual profit that should be allocated to market jurisdictions under Amount A (the allocable tax base), as distinct from residual profits attributable to other factors. This step ensures that other factors, such as trade intangibles, capital and risk, continue to be remunerated and allocated residual profit. To avoid complexity, this allocable tax base will be determined through a simplifying convention, which would be a fixed percentage. Thus, again, the Pillar One rules would determine residual profits allocable to markets, rather than the arm’s length principle.
Step 3: An ‘allocation key’ to distribute the allocable tax base amongst the eligible market jurisdictions where nexus is established for Amount A. This would be based on locally sourced in-scope revenue determined by applying the rules on scope, nexus, and revenue sourcing.
This three-step formula for determining the Amount A quantum could be computed under a profit-based approach or a profit margin-based approach.
Both approaches would, in theory, apply the three-steps of the allocation formula similarly and hence should deliver the same quantum of Amount A taxable in each market jurisdiction, ceteris paribas.
The administration of each approach may, however, present some variations (e.g., foreign currency exchanges), and these practical differences will influence the adoption of the most appropriate approach to calculate and allocate Amount A.
The revenue-based allocation key for Amount A would differ depending on whether the formula is implemented through a profit-based or a profit-margin approach.
The Pillar One blueprint outlines a marketing and distribution profits safe harbour which would put a ceiling on the allocation of Amount A to market jurisdictions where the multinational already leaves residual profits under existing arm’s length profit allocations.
Further, a domestic business exception would exclude from Amount A profits derived by an automated digital service or consumer facing businesses in a market jurisdiction which can be seen as autonomous from the rest of the group, i.e., sale of goods or services that are developed, manufactured and sold in a single jurisdiction.
It is noted that complex multinationals with multiple business lines, both in and out of scope, will likely have challenges applying these rules, as existing internal systems may not support such computations and such could require material system costs to comply.
The Amount A taxing right would be implemented through changes to domestic laws around the world, and by way of public international law instruments, in particular, a multilateral convention.
This approach could result in the proliferation of different regimes and nuances across jurisdictions, which would massively increase compliance/administration costs for multinationals and revenue authorities.
Implementation of this Amount A proposal will require agreement on the approach to computing and allocating residual profits, as well as further technical work on various aspects thereof, including financial accounting and taxation matters; treatment of certain items such as joint ventures profits, minority interests; interest expense; non-recurring gains and losses; dividend income; etc.
Fixed return for baseline marketing and distribution – Amount B
Amount B would standardise the remuneration provided for baseline marketing and distribution functions that occur in each market jurisdiction.
Application of Amount B would be prospective, and would not supersede MAP settlements nor unilateral, bilateral or multilateral APAs entered into before the implementation of Amount B.
Whilst standardisation of Amount B should reduce compliance, administration, and controversy costs for revenue authorities and multinationals, application of Pillar One to only automated digital service and consumer facing businesses would result in incongruent rules across industries.
As noted, increased compliance costs of Pillar One compliance will likely overshadow any benefit.
The blueprint describes baseline marketing and distribution as activities involving purchasing from related parties and reselling to unrelated parties, and routine distributor functionality.
In-scope activities would be defined by a ‘positive list’ of typical functions performed, assets owned, and risks assumed by uncontrolled routine [limited risk] distributors.
A ‘negative list’ of atypical functions, assets and risks of uncontrolled routine distributors would be applied to qualitatively identify additional activities beyond the scope of Amount B.
The blueprint contains extensive details in respect of these positive and negative lists, which will likely be difficult to apply in practice given the complexities of multinational operations and pace of operational changes and will arguably depreciate the OECD’s simplification efforts and limit the effectiveness of Pillar One with regard to Amount B.
Further, these positive and negative lists may advantage automated digital services and disadvantage consumer facing businesses especially relative to excluded industries.
For multi-functional distributors, Amount B may still be applied to the controlled activities in-scope, depending on the availability of reliable segmented financial information, which may further complicate matters.
The blueprint notes that Amount B would be determined in accordance with the arm’s length principle, based on benchmarking analyses under the transactional net margin method and a return on sales profitability ratio, with the ‘required’ percentage potentially varying by industry (such as fast-moving consumer goods, motor vehicles, ICT, pharmaceuticals and general distribution), as well as region, as may be supported by relevant benchmarking data.
Accordingly, Amount B could have a number of ranges of potentially appropriate fixed financial returns for industries and regions.
This Pillar One Amount B proposal appears broadly similar to the Australian tax office’s risk assessment guidelines for inbound distribution, which guidance is not sufficiently effective in a Covid-19 world where distributors have had widely varying financial experiences across industries (including issues relating to government support), and for purposes of addressing market shocks, losses, start-up operations, business changes, etc.
For this reason, Pillar One should be implemented as an elective alternative to digital taxes and/or as a series of safe harbours.
Amount B may operate as a rebuttable presumption, but not all Inclusive Framework members have agreed to this at this juncture.
In this regard, an entity that acts as a buy-sell distributor and performs in-scope marketing and distribution activities would be covered, but multinationals could potentially rebut the application of Amount B by providing evidence that another transfer pricing method is most appropriate.
The burden of proof for this would rest with the taxpayer. For example, the presumption would be rebuttable if the taxpayer had a sufficiently reliable comparable uncontrolled transaction, which was more appropriate under specific circumstances.
There remain divergent views on the breadth of baseline activities that should be included in t Amount B’s scope.
This blueprint assumes that in-scope distributors are defined based on a narrow scope of baseline activities, which is a view shared by many Inclusive Framework members.
There is interest, however, by some members to explore the feasibility of broadening the scope of Amount B. There is also some interest in implementing Amount B as an initial pilot programme.
Further, the precise definitions and applications of regions, industries, relevant activities, and indicators must be finalised through additional technical work.
Elimination of double tax and dispute prevention / resolution mechanisms
The blueprint provides a mechanism to reconcile the new taxing right (i.e., calculated at a group or segmented level), and the existing profit allocation rules (i.e., calculated at an entity basis) to prevent double taxation.
The proposed mechanism would first identify which entity(ies) within a multinational will bear the Amount A tax liability, which then leads to determining which jurisdiction(s) must provide relief for the double taxation arising from Amount A taxing right.
Thus, this mechanism has two components: the identification of the paying entity(ies) within a multinational (or segment); and the methods to eliminate double taxation. The proposals relating to identification of the paying entity (ies) are reasonably complex and could potentially be simplified
Under existing regimes, jurisdictions typically apply two main methods to eliminate international double taxation: the exemption method and the foreign tax credit method. The blueprint envisages that taxes imposed on Amount A could be relieved by paying entities using either the exemption method or the credit method, taking into account the broader preferences of Inclusive Framework members in this regard.
The blueprint indicates there are various technical issues that require further work, including interactions between Pillar 2 and the application of the exemption method; the centralised and simplified administration system and the tax certainty process for Amount A; and Amount A and certain withholding taxes collected by market jurisdictions.
The blueprint also outlines a variety of approaches to provide tax certainty with respect to Amount A, essentially all of which require government involvement.
The proposed processes have a number of elements such as a standardised Amount A self-assessment return / documentation package and centralised filing, validation and exchange of information; request for tax certainty by multinationals; an optional initial review by the lead tax administration; creation of the review panel, the review panel process and approval by affected tax administrations; creation of a determination panel and the determination panel process; solutions for unresolved cases; etc.
For the resolution of disputes, mandatory binding dispute resolution mechanisms will be developed in respect of Amount A issues and potentially for other matters.
Arguably, in this respect, the OECD and its Inclusive Framework members have materially underestimated the complexities of implementing, administering, and adhering to Pillar One (coupled with Pillar Two) across jurisdictions, industries, accounting standards, and multinational business segments and over time as market, technological and operational changes occur. This will likely lead to massive mid to long term compliance costs for multinationals, and potentially increased international disputes.
Where to next?
As noted, the blueprint identifies matters where a political decision is needed and proposals to bridge remaining divergences.
It also recognises the further technical work that will be required to finalise numerous aspects of Pillar One, in particular to reduce complexity, improve administrability, and address the implementation and operational capacities of developed and developing countries.
In this regard, essentially all key aspects of the blueprint require additional consideration, and significant simplification of the Pillar One proposals should be considered. Overall, the blueprint seems more akin to a sketch at this stage, notwithstanding the massive efforts of the OECD thus far.
The blueprint, along with the report on Pillar Two, has been submitted to the G20 finance ministers for consideration in their meetings in October and November.
Public consultation on Pillar One is now open, with submissions required by 14 December. The OECD and its Inclusive Framework members are targeting mid-2021 for resolution of technical issues and consensus on Pillar One.
It is noted that delays and complexities in implementation may decrease the prospects of international consensus and thereby increase the proliferation of digital taxes. More broadly, internationally consistent digital taxes may be more effective and efficient.
HAPPY BIRTHDAY TO A VERY SMART, BEAUTIFUL, YOUNG LADY ON JANUARY 13TH.