Tax Notes logo

Treasury’s Pillar 1 Reset: In Praise of Comprehensive Scoping

Posted on Apr. 26, 2021
Robert Goulder
Robert Goulder

Things are happening. We now have movement from the U.S. Treasury Department on what it wants to see from the OECD inclusive framework regarding modifications to the pillar 1 proposal. This is good news. It’s high time we hashed out what’s wrong with the OECD’s October 2020 blueprint.

The preliminary signs from Washington are encouraging, although it’s challenging to write about the evolving U.S. position when there are few known details from which to work. The little we know about Treasury’s preferences can be gleaned from a 21-page slide deck that accompanied the U.S. presentation to the inclusive framework steering committee on April 8.1 What follows is an attempt to decipher the project’s possible course correction based on those nuggets.

The Blueprint Made Me Do It

The blueprint was both a necessity and an obstacle. The nearly 500 pages of detail on the architecture of pillars 1 and 2 was stupefying — not as much in its length as in its sheer complexity. Much of the complexity carries a superfluous flavor. The clutter surrounding pillar 1 convinced some observers that the whole endeavor would eventually bog down under its own weight. That could still happen.

The blueprint leaves the reader wondering if a more straightforward alternative would be preferable. As this article is going to press, another body of international policymakers — distinct from the inclusive framework — is contemplating whether to amend the U.N. model tax treaty by adding article 12B.2 Could the U.N.’s withholding tax solution be any worse than what’s in the blueprint? The business community hates withholding, but it doesn’t like the blueprint either. Among two unpopular ideas, why not opt for the one that’s easier to administer and less prone to gaming?

The blueprint’s main usefulness is that it shows everyone what an unstable global tax environment would look like. This has the beneficial effect of making remedial measures easier to swallow. The Treasury slide deck says all of this in more diplomatic language. It praises the blueprint for providing “significant technical groundwork.” Well, it did accomplish that — if by groundwork you mean laying out design elements that are certain to produce endless litigation.

That these big ideas envision the need for a dedicated dispute resolution mechanism should be taken as an ominous sign. Corporate taxation is litigious enough, why make it worse? The indicia of a successful tax reform proposal is that it would reduce the volume of taxpayer disputes, not increase them exponentially.

The Treasury slides contain a related bullet point:

At the same time, the complexities inherent in the multilateral international tax architecture have made it difficult to reach consensus, especially on scope and related administrative issues.3

Allow me to translate that into curt language: Dear OECD, as things stand, pillar 1 is a non-starter because your concept of in-scope taxpayers is nuts.

Simplify or Go Home

Do you know any teenagers? They have a rule of thumb for sending text messages and posting comments on social media platforms. Don’t put anything in bold font unless your intention is to shout at someone. The Treasury slide deck is shouting at us.

The context is the discussion of consumer-facing businesses (CFB) and automated digital services (ADS). These are the two populations of taxpayers singled out for exposure to pillar 1 under the blueprint. The Treasury slides offer four bullet points under the section on ADS and CFB:

  • Compliance and administrative burdens disproportionate to expected tax benefits: simplification is highly desirable. [Emphasis in original.]

  • Lack of clearly defined policy objectives and principles to distinguish ADS and CFB from rest of the economy (and from each other).

  • Complexity and subjectivity of proposed rules specific to ADS and CFB raise obstacles to consensus.

  • Difficult determinations required under qualitative activity test could lead to many scoping disputes in practice.4

The words in bold font speak volumes. Again, allow me to translate. Simplification isn’t just desirable. It’s necessary if the OECD expects buy-in from the U.S. government.5

For starters, there’s no objective litmus test for determining in-scope status, just a cluster of vague and overlapping standards. That’s baffling when you think about it.

It’s significant that Treasury has publicly outed the problems with scoping and proposed something better. I suspect the previous Treasury secretary figured out that providing constructive feedback to the OECD wasn’t worth the political baggage. When former Secretary Steven Mnuchin made his infamous comment about the United States viewing pillar 1 as a safe harbor, he was effectively saying the hot mess wasn’t worth saving. With Treasury under new management, it’s been made clear that step 1 to achieving a new international consensus is fixing the scope.

Is Apple an ADS even though it earns most of its profit from selling tangible consumer products like phones, tablets, and laptops? That’s all hardware. Apple also has a hand in cloud computing and operates an online music shop in the form of iTunes. Does the music streaming platform cause the whole enterprise to be an ADS? As discussed below, segmentation is no solution.

So we’re still asking what pulls Apple into the gravitational orbit of pillar 1. Is it that its products have Bluetooth connectivity to the internet? These days you can buy thermostats and refrigerators that do so. Cars connect to the internet. Are Volkswagen and Toyota examples of ADS companies?

Perhaps Apple is subject to pillar 1 because it is a CFB rather than an ADS. It has a lot of customers spread across a variety of countries — but so does Starbucks. Is Apple subject to pillar 1 because it is more like Starbucks or more like Netflix; or do the details not matter because Apple is large and profitable? Whatever the answer, it must be something other than a gut feeling that taxing Apple differently is the correct thing to do.

The contours of the corporate tax base, however partitioned, should never be delineated by qualitative factors or subjective considerations. The subjectivity isn’t just an obstacle to consensus; it’s a deal killer. The concept of ADS and CFB need to go. That’s not because of U.S. intransigence or protectionism. Nothing about a corporate tax base should be qualitative.

It should not require a Ouija board to deduce whether a firm is subject to pillar 1.

Scoping Done Properly?

Treasury’s idea for replacing the CFB and ADS standards is eminently practical. It proposes quantitative criteria for identifying the “largest and most profitable” multinational groups and would do so “regardless of industry classification or business model.”6 This is consistent with the idea that the digital economy cannot be ring-fenced, which should be obvious by now. The chosen description for this approach is “comprehensive scoping.” It’s an upgrade over what came before.

In practice, comprehensive scoping would mean the new nexus and profit allocation schemes of pillar 1 would apply only to the 100 biggest multinational enterprises, as measured by objective criteria that could include both a revenue threshold and a profit margin threshold.7

Is 100 MNEs an arbitrary cap? Of course it is. Many numerical thresholds in the tax code are arbitrary. People need to get over that if the project is going to be viable. A cap based on size and profitability are the least discriminatory criteria you’re going to find.

The point of including a metric for profit margins is that it would snare multinational groups with a lot of residual profits, whose success is presumptively driven by intangibles. In essence, comprehensive scoping is more concerned with intangible-driven profits than digital-driven profits, which are not the same thing. Consider how pharmaceutical firms rely heavily on intangibles, yet you’d never describe them as being in the digital economy.

Going forward, digital status should be beside the point. Perhaps the term “digital economy” can be retired once and for all. It was always a distraction.

What about carveouts? The blueprint gave a dubious pass to whole sectors of the global economy, such as financial services and extractive industries. The Treasury slides address that in a bullet point in slide 12:

Sector-based scope limitations, if any, should be principled and based only on fundamental policy mismatches or irresolvable administrability constraints.

The fewer carveouts, the better. The position out of the gate should be that nobody gets a carveout.

The cap of 100 multinational groups is not the byproduct of an elaborate calculus; it’s the starting point. It stems from an awareness that pillar 1 need not apply more broadly to provide sufficient taxing rights to market and user jurisdictions.

The OECD’s impact assessment estimates that roughly 2,300 companies (applying ADS and CFB scoping) satisfy the annual global revenue threshold of €750 million.8 That’s too many players on the field. That number falls to 780 companies when you filter out firms without sufficient residual profits. That’s still too many players.

Sexy Numbers and Fancy Profits

I know what you’re thinking. Treasury came up with the idea of limiting pillar 1 to just 100 MNEs as a backhanded way to shrink the amount of profit reassigned to market and user jurisdictions.

Not so. The slides emphasize that comprehensive scoping is not intended to shrink the size of the pie. It expressly calls for “maintaining [the] blueprint’s level of re-allocable profits to facilitate consensus.”9 The quantum of corporate profit being shared doesn’t change. Treasury acknowledges that this requires flexibility as to the formulas being applied.

How much profit are we talking about? According to the OECD’s impact assessment, the estimated total residual profit subject to amount A is around $493 billion per year. Taking 20 percent of that produces a number in the ballpark of $100 billion per year.

We’ve been assuming the share will be 20 percent of residuals, mostly because it marks the midpoint of the suggested range — a floor of 10 percent and a ceiling of 30 percent. A consequence of the 100 MNE cap is that we should no longer cling to that assumption. The fixed percentage of residual profit that makes up amount A will need to be reworked if the quantum of profit (approximately $100 billion) is to remain the same.

The implication is that far fewer taxpayers would be tangled up with pillar 1, but a greater slice of each firm’s profits would be up for grabs. Mathematically, that’s the only way to maintain the quantum while reducing the pool of affected taxpayers.

Treasury is making the point that it’s more important to get the project’s scoping correct, and then work backward to arrive at the suitable allocation details. The fixed percentage of shared residual profits will always be seen as a sexy number. Here, it takes a back seat to the scoping principle.

While we’re at it, Treasury should demand that the OECD rethink the idea of differentiating between residual and routine profits for purposes of determining amount A. That leaves open the possibility that residual profits could play role in scoping, as discussed above.

It’s never been clear why the taxing rights of market or user jurisdictions should be conceptually linked to residual profits any more than routine profits. Pointing to amount B does not answer the question. A better idea is to ditch the distinction between residual and routine profits once the scoping has been settled, so that amount A could be based on a standardized measure of full profits.

Let’s be honest. Asking multinationals to bifurcate their profits into these pools is asking for trouble. They’re going to game the system — and why shouldn’t they? These are amorphous numbers that don’t exist in any real context.

Label the next dollar of profit as routine, and it’s outside amount A. Label it residual, and it’s included in amount A. Congratulations, you’ve created another cottage industry for profit shifters. You’ve created another transfer pricing. Why would any sentient being think that’s a good outcome?

Distinguishing between routine and residual profits is a task that corporations aren’t doing for purposes of tax accounting, financial accounting, or country-by-country reporting. That its sole purpose is to support amount A means that taxpayers will have no reason to treat it as anything other than a tax-induced construct.

The popular fixation with residual profits makes sense to the relatively small number of people with advanced degrees in economics, and that’s not to be ignored. But the approach is not the optimal solution if your first priority is to come up with model for pillar 1 that isn’t being continually litigated.

A further simplification determines amount A according to the corporate group’s worldwide book income, full stop.10 In terms of the arithmetic, it’s just as easy to allocate the same quantum of profits ($100 billion per year) using worldwide book income as it is using residual profit. It’s easier for everyone and the market jurisdictions would be no worse off.

The last time I checked, $1 of tax revenue culled from routine profit pays for public services just as effectively as $1 of revenue taken from residual profits. Oh, but the economic distortions you’ll cause. Enough with the distortions, please. Enough with intellectual subservience to fancy-pants profits. If pillar 1 is going to survive, it needs to prioritize simplicity, certainty, and administrability. With that stroke of the keyboard, I will no longer receive holiday cards from my economist friends.

Segmentation Is Not Your Ally

The blueprint talks about segmentation. What a clever waste of everyone’s time. Somewhere in the pantheon of bad ideas there’s a spot reserved for segmentation.

Segmentation is the OECD’s idea for dealing with MNEs with multiple business lines, which is basically all of them. It’s anticipated that several of a company’s business lines would be considered in-scope for purposes of pillar 1 (by virtue of ADS or CFB status), while other business lines would not. Each business line would have its own annual revenue tally, unique cost structure, and profit margin.

As a result of segmentation, Disney might be an in-scope taxpayer for its video streaming venture, but not for its theme park operations. Amazon might be in-scope for its web services, but not for its retail stores.

Segmentation is an unnecessary chore that’s reflective of a poor design. The Treasury slide deck seems to agree:

  • Business line segmentation is the most complicated and difficult building block in the Blueprint. [Emphasis added.]

  • The need for business line segmentation is highly reduced under a comprehensive scope.

  • We support an approach that eliminates or minimizes business line segmentation and thus vastly simplifies pillar 1.11

To say that segmentation is the most complicated thing in the blueprint is saying something. The more you contemplate segmentation, as described in the blueprint, the more U.N. article 12B starts to look like the path of least frustration.

Now What?

It’s too early to know what other remedial proposals Treasury has in mind. The calculation of amount A remains problematic. Beyond that lies amount B; and we haven’t touched on pillar 2.

We can surmise that scoping along the lines of what Treasury has suggested still won’t solve one major obstacle — most of the affected taxpayers will continue to be U.S.-based multinationals. When this eventually gets before Congress (if it makes it that far), it could still carry the air of a threat to the U.S. fisc.

I have no idea how Treasury officials are going to convince skeptics in Congress that pillar 1 doesn’t represent some level of forfeiture. For now, the game plan should be to simplify and rationalize it as much as possible and to worry about Congress later. One battle at a time.

In those terms, it’s good to have Treasury officials leading the charge.


1 Stephanie Soong Johnston, “U.S. Offers Key to Unlock Scope Issue in Global Tax Reform Talks,” Tax Notes Int’l, Apr. 12, 2021, p. 148.

2 Sarah Paez and Annagabriella Colón, “U.N. Tax Committee Moves Toward Adopting Treaty Article on DST,” Tax Notes Int’l, Apr. 12, 2021, p. 246.

3 U.S. Treasury Department, Presentation by the United States to the steering committee of the inclusive framework meeting, slide 9 (“Pillar 1 Blueprint provides many solid foundations”) (Apr. 8, 2021) (hereinafter, “Treasury slide deck”).

4 See id., at slide 10 (“Challenges of ADS + CFB as scoped in Blueprint”).

5 Other commentators have already noted that the blueprint desperately needs simplification. See Michael J. Graetz, “A Major Simplification of the OECD’s Pillar 1 Proposal,” Tax Notes Int’l, Jan. 11, 2021, p. 199.

6 See Treasury slide deck, supra note 3, at slide 12 (“Proposal for comprehensive scoping”).

7 See id., at slide 15 (“Proposed quantitative screening criteria”).

8 See id., at slide 11 (“OECD assessment”).

9 See id., at slide 12 (“Proposal for comprehensive scoping”).

10 See Graetz, supra note 5.

11 See Treasury slide deck, supra note 3, at slide 18 (“Building Block: Segmentation”).


Copy RID