It is possible to be hopeful and constructively critical of the OECD’s Two-Pillars strategy that flows out of the BEPS project, most recently advanced in the OECD’s July 1 announcement now a feature component of the July 2021 OECD Secretary-General Tax Report to G20 Finance Ministers and Central Bank Governors for their meeting in Venice and the OECD July 2021 publication under the auspices of the OECD/G20 Base Erosion and Profit Shifting Project entitled Addressing the Tax Challenges Arising from the Digitalisation of the Economy.
But, on both counts, it is also helpful to the cause to be realistic, and in so being to make a useful contribution to the more orderly interaction of countries’ tax regimes without sacrificing their independence. Any accommodations by countries to what amounts to a global “plan” for a “new framework for international tax reform” foreshadowing a new system to recalibrate “key elements of the century-old international tax system”, are likely to come at a price, for now an undisclosed and possibly unknown price, for countries, other than the United States perhaps, which recondition their fiscal and tax policy in the image of the common denominator of a larger group.
The announcement puts me in mind of a common trend in retail real estate sales in continuing “hot” markets. First, the sign on the lawn says “Coming Soon”; no details about the property or the terms – including the price – are advertised. They are left to the imagination of possibly interested buyers and the expectations for the market generated by that imagination. This July 1 announcement is a little bit like that – offering few, if any, new insights into what the Pillars Blueprints announced in October 2020 except, possibly, the inertia heralded by the title of the release that “130 countries and jurisdictions join bold new framework for international tax reform”. The rest, including the “price” to countries which sign on, is “coming soon”.
What “International Tax System”?
We commonly refer to the more or less coordinated interaction of countries’ tax laws as an “international tax system”. But that is both inaccurate as a legal matter and an exaggeration descriptively.
What we have most recently is an outgrowth of the post-WWI work championed by the League of Nations - systematic ways in which the inevitable intersection of countries’ sovereign taxing rights and the private law regimes to which they are accessory are rationalized in the interest of avoiding gratuitous costly trade distortions, without depriving those countries of their entitlement, and to establish in their own images the fiscal choices and policies they choose and then choose to enable with the assistance of tax law. In the broadest of terms this has qualities of a “system” – but it is not and does it require a single system of unified tax and private laws. In fact, it assumes the opposite and then offers ways to accommodate the friction of self-interested actions by countries, primarily through tax treaties and unilateral domestic law provisions like foreign tax credit which are to the same effect.
The difference is not semantic. Yet the Pillars initiative, almost by the fast moving and brochure like presentations of the ongoing work which presume the existence of a system, and implicitly then the unification in important respect of countries laws, would suggest that that it is a fundamental, though largely undiscussed, implication of the Pillars.
These comments look at some of the more revealing aspects of the “Statement on a Two-Pillar Solution to Address the Tax Challenges arising from the Digitalisation of the Economy” that accompanied the July 1 release.
“Agreed a Two-Pillar Solution”
The words “agreed” and “solution” suggest, well, actual terms of agreement, which in this case are in many ways legislative and an actual solution – something the terms of which are well defined, including the necessary legislative commitments by countries necessary to make “it” happen. Yet, the release to which the Statement is attached speaks of a “plan” and the establishment in the Statement of a “framework” for “reform” in respect of a “system” that really is not a system, but is more in the nature of certain self-interests which motivate mutually beneficial accommodations by countries concerning the application of each-other’s tax laws. We are some distance from a “solution” as the Statement itself clearly reflects. A solution will require massive and universally consistent and coherent law changes by countries, not only to provide missing definition to the announced proposal, but to also reconcile and rationalize the changes meant to apply to very large multinational enterprises with countries domestic tax laws as they otherwise apply – including to those very large taxpayers too, unless and to the extent that the enabling law changes say that they do not. And such fundamental changes somehow have to be durable in a longer term than the moment. That seems to foreshadow a great deal of work and self-denial by countries that has not happened so far in the evolution of “the century-old international tax system” that, as it is, encountered and tried to address, in much the same ways interestingly, the challenges to inter-nation tax friction that are attributed now to “digitalisation”.
And “agreed” means agreed. It does not mean maybe, as one or a few of the agreeing parties insist. Reporting about reactions to the July 1 announcement already foresee political objection and serous legislative impediments in the United States, the main driver of this advance of long-running OECD efforts in this regard. Indeed, though perhaps largely unspoken in the present conversation, the operations and planning of US multinationals abetted deliberately by US tax policy and legislation, are at the core of, and a principal catalyst to, the motivation for the BEPS initiative from the outset. The US’s agreement – in a reliable, durable, consistent way – and its execution of that agreement legislatively and in its tax treaties, are essential. Yet, there is reported political opposition in the US - the legislative capacity has been, and still is, fragile at best. Even so, there are continuing suggestions, also reported, that US agreement to the July 1 proposal or anything like it depends on the US being excused from adopting elements of the proposal that the US sees, in its view, already reflected in existing US tax law. This is a not uncommon reaction by the US to OECD-led initiatives which despite its influence the US does not formally embrace.
While it is not for commentators to say, Finance Ministers and governments of other countries might consider exercising considerable caution before moving headlong to adopt changes to their tax laws and attendant underlying fiscal policy that amount to ceding or restricting tax jurisdiction, if the main "cheerleader" for the changes does not play by the same rules or even possibly play at all. A notable example of why caution is desirable, if not indeed necessary, is the abandonment by countries of their “digital services taxes”, if they have or are proposing them, as a quid pro quo for agreement on the Pillar One “solution”. Canada has proposed its own version of a DST, to be effective January 1, 2022. It addresses the core of the “problem”, which is how the notion of “permanent establishment” has so been treated as a prescriptive limitation on tax jurisdiction that outmoded interpretations of that concept in fact restrict countries’ ability to tax multinational businesses clearly carrying on business in those countries because they lack a kind of taxable business presence that, inherently, those kinds of businesses do not need and consequently do not employ.
Nexus and Revenue Sourcing – and “Intangibles”
“Source” is the Key
The key to the entire Pillars initiative lies in the aspiration to develop universal “source” rules for income and prescriptive rules for allowing source countries to tax it.
From the outset, the BEPS project of the trunk of which the Pillars are branches has fundamentally been about “source." There is no international agreement on the source of income; it is not the same thing as the origin of income, and it is not an economic or geographic notion even though these may be legal and factual evidentiary factors in applying countries’ source rules. The “source” of income is a legal notion, that determines tax consequences based on private law constructions of various kinds and how the relationships they frame are performed including where, as a legal matter, they are accountably performed. As I have written elsewhere, with reference to an important paper by Professors Hugh Ault and David Bradford, there is no “natural law of source”; the source of income is what countries define in their law - not just their tax law - that source to be. This appreciation carries with it another important consideration, namely that to reconcile and rationalize source as a tax notion it is necessary also to address countries private laws, those that imbue legal personality and define relationships, the nature of property, and enforceable outcomes. No small matter. Even if some sort of over-arching or supervening umbrella law were enacted to apply to some taxpayers according to new universal standards, the influence in domestic tax laws of private law regimes that implicitly are part of those domestic tax laws because they underpin the tax laws has to be internally reconciled with the laws applicable to others within those regimes in the same country which are not as “large” as this OECD / Inclusive Framework “solution” conceives.
In fact, from the outset, the object and undeniable focus of the BEPS project, although not in so many words, has been to attain even by indirection a universal “source of income” notion. It is on that that all of the Actions, notably those dealing with substantive tax law and treaty matters, converge. Several of the BEPS Actions would accomplish that essentially by arranging the application of tax rules in such a way as to frame what amounts to a “negative source” convention – establishing by various observable standards where income cannot and should not be considered to arise, notably in countries with preferential tax regimes, and leaving for them a modest allocation of income consistent with a risk-free return. It spares the bother and the controversy associated with “lifting the corporate veil” on multinationals’ legally fragmented business and economic “unities” and/or admitting to the inherent and intractable limitations of transfer pricing’s arm’s length standard, while in fact addressing both covertly. The result is to assign the income to those places where, reasonably, it could have been earned according to cognizable and observable business and economic markers of how it is earned, in ways no less consistent with the goals and even standards found in the League of Nations work that, as the release observes, is “century-old”. (For other related commentary and references, see: Scott Wilkie, The Way We Were? The Way We Must Be? The ‘Arm’s Length Principle’ Sees Itself (for What It Is) in the ‘Digital’ Mirror, 2019, INTERTAX, Volume 47, Issue 12, 1087-1102; Scott Wilkie, The ‘Source’ of the International Tax Conundrum; and Scott Wilkie, New Rules of Engagement? Corporate Personality and the Allocation of “International Income” and Taxing Rights, in Brian J. Arnold, Ed., Tax Treaties After the BEPS Project A Tribute to Jacques Sasseville (Toronto: Canadian Tax Foundation, 2018), 349-371.)
Allocating and Taxing “Residual Value”
The “Nexus”, “Quantum”, “Revenue souring”, and “Tax base determination” elements of the Pillar One “plan” / “framework” / “solution” contemplate : (i) “a new special purpose nexus rule permitting allocation of Amount A to a market jurisdiction”, (ii) to permit that jurisdiction to tax a measure of “residual profit in excess of 10% of revenue [that] will be allocated to market jurisdictions with nexus using a revenue-based allocation key”, (iii) according to “detailed source rules for specific categories of transactions [which] will be developed”, (iv) measuring profit or loss “by reference to financial accounting income”.
How demanding is this? Let us look at the pieces.
- Allocating Amount A (and Amount B):
Broadly, Amount A in the Pillar One formula is a measure of the residual value of a provision of services or sale of goods beyond the amount that compensates for so-called “routine” contributions, the latter being addressed as Amount B associated with “in-country baseline marketing and distribution activities” about which no detail is provided or anticipated beyond a stated intention to “simplify” and “streamline” this notion notably respecting “low capacity countries”. Amount A is closely identified, generally and in the Pillar One Blueprint, with “intangibles” value – the value of the corporate opportunity that arises from operating as a commonly controlled multinational enterprise within the “halo” – a transfer pricing description – of the corporate family, rather than as separate enterprises simply contracting for each other’s supplies. It is also closely identified with contributions to profitability associated with unique characteristics of particular markets and their consumers. Even more broadly, it is through Amount A that the core profits of a multinational enterprise would be split, even if the outcome of the split were to be described with reference to typical transfer pricing methodology purporting to rely on comparable external circumstances.
The value contribution denoted Amount A lies at the core of transfer pricing income allocations from their outset in the modern era of international taxation, reaching back to Mitchell Carroll’s study for the League of Nations of how to allocate international income of enterprises conducting business in more than one country. It was the driver of the OECD’s study of “intangibles” leading to revisions to Chapter VI of the OECD Transfer Pricing Guidelines. It was at the core of the OECD’s study of “business restructuring” ultimately captured in guidance set out in Chapter IX of the Transfer Pricing Guidelines. It accounts for the value of the “uniqueness” of being a multinational enterprise, giving rise to continuing intractable questions, if they are to be answered qualitatively with judgment rather than formulaically, about the relative value of contributions made by legally distinct (as entities or establishments) components of a multinational group. It is the core and essence of the transfer pricing “problem”. And, of course, it pits countries “against” each other in their views about which corporate group member, which equates to the country with which it most closely associated in the relevant way, contributes the most to a common enterprise.
A “safe harbour” is contemplated for “marketing and distribution profits” that would be allocated to a market jurisdiction which “are already taxed in a market jurisdiction”. But no detail is offered, beyond the kind of detail contemplated by the Blueprint. Suffice it to say though, that dividing residual profit, which inherently means dividing profit attributable to some manifestation of “intangibles” has proven difficult at the best of times, and this takes it another step in the direction of tracing components of residual profit to “marketing and distribution” contrasted with other functions and contributions. The notion of “marketing intangibles” has been controversial in Canadian tax law and administration for a long time and it has been equally controversial as an object of the OECD’s work on “intangibles”, “business restructuring”, and BEPS Actions 8 – 10.
In other words, it is not only a hard question to resolve, let alone by the Fall of 2021 with global consensus and suitable legal formulations. It is the hard question.
How ambitious is it to imagine that finally we will have universal agreement on profit splits, in an environment that has resisted, denied, and then only guardedly acknowledged that that is what really is going on in, for example, the transfer pricing arena?
- “Residual Profit” Measured as a Percentage of “Revenue”
First, what does it mean to measure “profit” as “revenue”? And what is a “revenue-based allocation key”? The Blueprint offers extensive methodology on how to extract residual value and measure it. Is this really formulary apportionment (fractional apportionment, using language closer to the original Pillar One proposal) but set in a more diplomatic way, to avoid the inflammatory connotations of formulary apportionment in transfer pricing?
- “Detailed Source Rules”
BEPS and the Pillars are a dense and sometimes confounding weave of threads all converging on “source” not only as a question but “the” international tax question, as it has been from, again, the modern era of international taxation in the work of the League of Nations. Earlier comments have addressed the “source” notion, a legal, not an economic or geographic concept, that is notably infused with private law constructions. Even as it is, not all countries have detailed source rules explicitly in their tax legislation. Canada is a case in point. The only source rule in the Canadian Income Tax Act is found in section 4 of that Act, which requires taxpayers to separately measure by location four qualities of income: from employment, services, business and property; capital gains and losses, i.e., gains and losses from disposition of capital property and therefore exclusively property-oriented, are accounted for in a separate “code within the Act” that filters into the computation of income.
Given that “source” is a legal notion that usually depends heavily on an infusion into the tax rules of private law, even if not readily so visible, how likely is it that through “detailed source rules” the tax and general law regimes of all relevant countries will be reconciled in an enforceable, predictable way – for those enterprises within the Pillar One scope, allowing for the inevitable possibility that for other taxpayers whatever “source” means will continue to be what it means in each particular tax regime?
Also, the Statement contemplates the use by multinational enterprises of “a reliable method” based their own circumstances. Coupled with the inherent difficulties associated with devising a common “source” notion and finding common ground on financial accounting standards, it is not clear how “a reliable method” to source income on a case by case method while still reconciling with putative universal standards and internationally consistent outcomes baked into the legislative schemes of participating countries will be achieved, or more importantly be readily administrable and verifiable by tax authorities even if they are not, to use a term used in the Statement, “low capacity”.
- “Financial Accounting Income”
Canadian taxation has plenty of experience of reconciling financial and tax income. Most recently, an example of how difficult it is to infuse tax rules with accounting notions, even for a single tax regime, is the “difficult” experience with the ill-fated “foreign investment entity” (FIE) rules. One of the early challenges in that unhappy tax policy and legislation experience was associated with how to choose among and, more importantly how to reconcile and rationalize for the targeted tax purposes various generally accepted accounting conventions that taxpayers might use, rather than others that other taxpayers would use. It proved not easy to accommodate the reality that not all taxpayers use the same accounting conventions, and possibly not all accounting conventions were equally fit for purpose. Indeed, as experienced more generally with Canadian recognition of International Financial Reporting Standards (IFRS) has shown, some very difficult tax versus financial income measurement issues can be encountered, as well as the possibility that taxpayers in some countries do not use those Standards. All of this is to say that while it seems it should be easy to adopt financial accounting standards for measuring tax income, in Canada’s experience at least, it has been easier said than done. And that wholly discounts the international dimension, where the overall objective effectively is to allocate a commonly determined pool of international income.
Certain, Non-Duplicative Outcomes
“Double taxation” will be relieved in two ways, through exemption or credit features of tax systems, and generally, but with possible exceptions for, “developing economies”, “mandatory and binding” “dispute prevention and resolution mechanisms” for Amount A. In principle, none of this is new as such; countries are familiar with deferring to each other’s tax and right to tax through exemption and credit, as components of their domestic tax regimes and via agreements to limit the assertion and/or relative pre-eminence of taxing rights through the distributive articles of tax treaties. As well the mutual agreement procedure enhanced by some countries to include arbitration by some countries, and the contemplated broader adoption of arbitration to solve tax disputes in the “Multilateral Instrument” are already well-known. What may be more unique, and possibly unwieldy, is the expectation that pre-emptive dispute resolution, with features of transfer pricing’s “advance pricing agreements/arrangements”, might become a year-by-year institutional features of income measurement, seemingly in addition to the transfer pricing and MAP processes that already apply. Typically, both MAPs and APAs take considerable time to complete, and there is as yet no universal embrace of binding (on countries) dispute resolution via arbitration, a prospect that is particularly contentious for “developing economies” for which “an elective binding dispute resolution mechanism” could conceivably apply.
It is not clear how inserting tax authorities, with different interests and perspectives among themselves, into the real time determination and measurement of income, taking account too that some of those authorities on the same brief may be from “developing” and developed economies, would be accomplished efficiently within the available resources of tax authorities. Equally, among no doubt many other things, it is not clear how here and elsewhere “developing economies” (which may also be “low capacity” though the Statement does not explicitly make that connection), no doubt recognizing their unique interests and challenges, could participate on an elective basis when counterparty nations that are not “developing economies” and their taxpayers would be engaging with each other on mandatory basis
It is to be noted that the focus of dispute resolution is on Amount A, which in its other guises historically has proved to be the “source” of so much difficulty in establishing income’s source by country and contributing taxpayer.
Again, this on its own seems to be an ambitious objective in the international tax world as we know it, let alone to be achieved as a structural feature of how to agree among countries and taxpayers on the allocation particularly of residual income that one way or another is influenced by “intangibles” that in turn are difficult to value (consensually).
The Quid Pro Quo for Pillar One
The expectation, said euphemistically by the phrase “appropriate coordination” is that those countries that have or would have “digital services taxes” will abandon them. This already has been noted in these comments. Again, assuming that the Pillar One measures are desirable and would otherwise proceed in some fashion and recognizing that inherently they involve not unfamiliar compromise of countries’ “tax sovereignty”, this is particularly dramatic and possibly perilous step unless there is durable, reliable, concerted universal adoption of the Pillar One paradigm in legislative and enforceable (including between countries) ways. To this end, given the undeniable significance of the US tax system for spawning many BEPS situations and of US taxpayers whose business by kind and size dominate the scope of Pillar One, it seems that countries would want to know, first, whether the US has adopted Pillar One as such and as proposed before conceding jurisdiction to tax by abandoning DSTs.
To What End?
The direction and effects of Pillar One generally are notable, not only for its objectives but its design features.
Essentially, for in scope taxpayers, Pillar One essentially has two aspects. First, it abandons doctrinaire notions of “permanent establishment”, which with private law constructions accounts for so much of “profit shifting” and for businesses where it is obvious identifies carrying on business with business presence and business presence of suppliers with the facilities and personalities of consumers. In short, consumers in market jurisdictions have dual roles, a premise interestingly of the Canadian DST proposal in the April 2021 Federal Budget, as manifestations of suppliers (resident and non-resident alike) and as self-interested consumers engaging with their suppliers effectively through themselves in their other role. Second, parsing the manifest uncertainty in how income is expected to be assigned to market jurisdictions, formulary or fractional apportionment will be a guiding force at least directionally if not more, so as to avoid the inevitable complications – some of which even many of which may persist – associated with qualitative judgments of how much income should be assigned according to legal determinations of source.
Pillar Two and Minimum Tax
The Pillar Two minimum tax regime, orchestrated through the “GloBE rules” anticipates a “common approach” that is ostensibly elective but in fact, by negative implication mandatory. Those countries that chose to adopt the GloBE rules are expected to follow Inclusive Framework agreed standards. Those that would not adopt the rules must defer in applying their tax rules to determinations under the GloBE rules for those countries that elect to participate. Consequently, to avoid whipsaw by inconsistent outcomes, it is not hard to imagine countries conforming even if that participation is grudging and reluctant.
Both Pillar One and Pillar Two contemplate exclusions or qualified participation by “developing economies” and “low capacity countries’ in certain situations. Notable for Pillar One is the exclusion of extractive and financial industries. The principled basis for their exclusion is not mentioned nor, for that matter is any other basis, though it might be imagined that extractive industries are less mobile in the first place under tax systems as they ordinarily would apply and financial industries’ capital deployment commonly is so heavily regulated that allocations of income even at the best of times is complicated as the unique financial industries segment of the “Authorized OECD Approach” (AOA) for allocating income to permanent establishments reflects.
There is an outright exception from the GloBE rules for various large pools of capital that despite their special designations and prohibitions that may apply to carrying on business, in fact behave and operate in (all) material respects like multinational business enterprises with which in many instances they no doubt compete. Mentioned are: [g]overnment entities, international organisations, non-profit organisations, pension funds or investment funds that are Ultimate Parent Entities (UPE) of an MNE Group or any holding vehicles used by such entities, organisations or funds”. The focus seems to be on the ultimate parent or holding entity, rather than subordinate entities that may still be required to attorn to the GloBE rules. This is, still though, a notable exception.
In some ways, though, a more interesting exclusion in light of the notable US interest in the Pillars standing up, so to speak, concerns Pillar Two. The OECD Statement uniquely mentions the US and its tax regime by name with reference to “GILTI co-existence”. It is hard to imagine that this is other than a “nice way” of saying that since, in the US’s view, its GILTI (Global Intangible Low-Taxed Income) rules are consistent with the objectives of a minimum tax to ensure that income is taxable somewhere by a robust tax regime, the US should not have to recondition its tax rules legislatively or otherwise to do what it insists all other should do. It is not to be missed that the force and effect of the GILTI rues target intangible income relating to services and it is proposed also goods under the Biden Administration’s Made in America Tax Plan and repatriate that income to be taxed in the US. It would not be surprising if a collateral effect, if not an implicit expectation, also is to cause the income earning activities to relocate to the US. In other words, it is clear that the US is angling for a "pass" on Pillar Two in deference to its own country specific, country oriented relevant domestic tax rules. The anticipated exception with respect to the US rules is expressed in terms of “ensur[ing] a level playing field”. Laudable as that objective might be, the question remains - define “level”? Is the playing field to be levelled everywhere else to accommodate the US perception of “fairness”, or is the US to be held to account according to the view of fairness and what is “level” held by the other 129 Inclusive Framework countries?
There also de facto exclusions or attenuations of effective taxation in both the Pillar One and Pillar Two aspects. These show up in scope and transition rules that affect both the enterprises that would be subject to the new “rules” and the amounts subject to them coupled with the effective rates of tax that would apply including in transition, though like all of the Statement, these elements too are not fully or even at all resolved.
As real estate brokers so often say these days about properties being sold into hyper active markets, the listing details and the "open house" for Pillar One and Pillar Two are "coming soon". And, a lot of details and a very "open" house for inspection there will need to be. Sometimes sacrificed in "hot" real estate markets as bandwagon bidding takes hold are "home inspections" for which there may not be time or which fall away as a resource for assessing the consequences of such a large purchase beyond the heat of the moment of purchase because so many potential purchasers clamor to be "winners". Hopefully, critical, objective, and appropriately self-interested inspections of the new global "tax home" will not be sacrificed simply to make the "purchase" or get the "sale".