The BEPS Gambit: Will the OECD Know When to Resign?

There are two ways to lose a chess match.

Under the preferred approach, the player resigns with dignity upon realizing that defeat is inevitable. Checkmate is lurking a few moves away, and because it cannot be stopped there’s little purpose in delaying the outcome longer than necessary. You learn your lessons and move on to the next challenge.

Under the alternate approach, the player tries to squirm out of trouble through a series of desperate maneuvers, but each futile move only prolongs the death spiral. Knowing the correct moment to resign is more art than science. The world is full of well-meaning people with clever plans who don’t know when to quit.

The etiquette of defeat is a prominent theme in the popular Netflix series, The Queen’s Gambit, based on a novel by Walter Tevis. The plot centers on a young orphan, Beth Harmon, who turns out to be a chess genius with self-destructive tendencies. She is her own antagonist. One of the most memorable scenes from the series occurs when a mentor tries to teach Beth how to lose in the proper way. The lesson does not go well but remains a formative experience that sticks with the character into adulthood.

Why this talk of defeat with dignity?

It’s a suitable metaphor for where the tax community stands at the beginning of 2021. Consider where we’ve been during the last half-decade. The final reports produced by the OECD’s base erosion and profit-shifting initiative are now more than five years old. While there have been a few notable accomplishments, such as country-by-country reporting, the reality is that we are no closer to resolving the thematic material contained in BEPS action 1, the challenges raised by the digital economy. We know perfectly well that the international consensus is no longer suitable for 21st-century business models, but there is no durable agreement on how to rectify the situation.

Given how the permanent establishment doctrine operates, source countries are correct to have misgivings about tax treaties based on the OECD or U.S. models — which are lumped together for these purposes. This is especially true for developing economies. Did those countries realize what they were signing up for when they concluded tax treaties with the United States or other capital-exporting heavyweights? Yes and no. They certainly knew there was a significant trade-off between attracting foreign direct investment and relinquishing key portions of their taxing rights. The tension between economic development and revenue capacity was always built into the process. But it seems doubtful the affected countries realized the extent to which today’s digital giants could reap massive profits from local sales or user activity without being subject to the local corporate income tax.

I’ve come to view the OECD’s inclusive framework as a collective dialogue on buyer’s remorse. A lot of source countries signed tax treaties over the decades, and many now wish they hadn’t. Consider the demise of withholding at source, which could serve as a practical fix for these countries’ inability to reach nonresident digital service providers through the conventional corporate tax. The countries could apply suitably tailored gross-basis withholding tax as a backstop. In the event of excess withholding, the nonresident taxpayers could voluntarily submit themselves to the local taxing jurisdiction to obtain whatever refund they are owed.

We don’t see much of this kind of withholding because tax treaties prohibit it. The cynical perspective is that the whole purpose of a tax treaty is to saddle source countries with the PE doctrine and flaky transfer pricing rules, while denying them the ability to obtain fair outcomes via withholding. Seen through this lens, treaties are a straitjacket. The OECD model might not be the root of all evil, but it is the seed of discontent. Tax treaties explain why BEPS had to happen.

The U.N. model treaty offers a different approach. A pending revision would add article 12B, permitting source-country withholding on automated digital services irrespective of physical presence. Taxpayers could avoid gross-basis withholding by electing net-basis taxation on qualified profits, defined to render transfer pricing inapplicable.¹

Article 12B is such a good idea for source countries that residence countries might never agree to it. This is unfortunate. It’s doubtful the U.S. Treasury would allow a comparable provision to migrate into the OECD model. How I would love to be proven wrong on that point. Widespread acceptance of article 12B would make BEPS 2.0 unnecessary. In terms of policy, the U.N. Tax Committee has delivered the treaty-based solution that the international tax community has been seeking. It’s being ignored. Instead, we have the OECD’s pillar 1 and pillar 2, which are baffling in their design. And even they require too much forfeiture to satisfy U.S. tastes.

When then-Treasury Secretary Steven Mnuchin announced that pillar 1 should be regarded as a “safe harbor” (meaning it should be something that U.S. companies could opt into at their discretion) he was signaling a harsh reality to all stakeholders: You’re crazy if you think Congress will go for any of this. The handwriting was on the wall under the Trump administration. It’s still there under the Biden administration.

But Aren’t Things Different Now?

No, actually. Things aren’t as different as you think.

Some people are shocked when they realize that incoming U.S. officials are embracing the hard-line positions of the Trump administration, especially when foreign relations or multilateral collaboration is involved. Consider the recent exchange on Capitol Hill between Sen. Lindsey Graham, R-S.C., and Antony Blinken, President Biden’s then-nominee for secretary of state (confirmed January 26). This was during Blinken’s Senate confirmation hearing on January 19, one day before Biden’s inauguration.

Earlier that morning, outgoing Secretary of State Mike Pompeo directly accused China of engaging in genocide against the Uighur population of the Xinjiang province. Pompeo could have delivered those comments anytime during the last four years, but he waited until January 19, in part, to place his successor in what was assumed to be an awkward position.

Graham looked stupefied when Blinken said he completely agreed with Pompeo’s statement. On the substance of the Uighur issue, there was no gap between the two administrations. Graham expected the witness to dance around the sensitive topic without providing a definitive answer, as is so common with politicians. Nope; what Pompeo said was fine with Blinken, meaning Biden’s position is the same as Trump’s.

This wasn’t a one-off. Blinken went on to agree with Trump-era positions on a series of sensitive matters, including the location of the U.S. embassy in Jerusalem, recognition of opposition figure Juan Guaidó as Venezuela’s rightful head of state, and the need for conditions-based negotiations with the Taliban concerning U.S. troop drawdowns in Afghanistan. The picture that emerged was one in which Biden’s people would be more tactful than their predecessors, but the endgame would remain mostly unchanged.

The same pattern held for Biden’s then-nominee for Treasury secretary, former Fed Chair Janet Yellen (confirmed January 26), who faced her own confirmation hearing the same day as Blinken and was directly questioned about foreign digital services taxes and the OECD inclusive framework. Her responses were perfect diplomacy. She explained to senators that she looks forward to “actively working with other countries through the OECD negotiations on taxes on multinational corporations.”²

Just to be clear, that’s not a flip. Nor is it a teaser of some future softening in the U.S. opposition to BEPS. Treasury still doesn’t like pillar 1, and it still expects a pass on pillar 2 because the Tax Cuts and Jobs Act ushered in the global intangible low-taxed income regime. Yellen’s language is exactly what it purports to be: an affirmation of multilateral engagement — and not a whiff more.

Someone who didn’t get the message was French Finance Minister Bruno Le Maire, who told a television interviewer the following day that Yellen’s comments mark a change in U.S. policy. Dream on, monsieur.

Le Maire previously stated that global consensus on the OECD pillars is being blocked only by Washington, implying that the moment Biden and Yellen announce their personal blessing a shiny new global reform package would be a done deal. Does he know the United States is not a parliamentary system?

Wishful thinking has got the better of Le Maire, or perhaps something was lost in translation. Whatever the case, he and others should know there is no persuasive reason for the United States to forfeit corporate taxing rights while receiving nothing of perceived value in return. That’s a roundabout way of saying Washington isn’t intimidated by the threat of DSTs.

True, the U.S. government doesn’t like the DSTs that are popping up. But the tax continues to be viewed as an inconvenience that can be satisfactorily dealt with in the trade arena. Any cross-border spat that can be squeezed into a trade box immediately becomes less of a problem than it was previously. The United States has leverage in trade tussles. The task of responding to foreign DSTs will end up on the desk of the U.S. trade representative, not the Treasury secretary. That tells you all you need to know about how this will play out.

Unilateral countermeasures are no reason for Washington to throw in the towel on an international tax regime that generally works fine for most U.S.-based multinationals. In that sense DSTs are no big deal. In a slightly different context, professor Reuven S. Avi-Yonah recently expressed the same idea.³

At last count there were already more than 30 DSTs (or equivalent regimes) in place around the world. That number will certainly grow over the course of the year. Bring them on.

Excluding the United States, there is no reason for any country to refrain from enacting a DST, and to do so as expeditiously as possible. One could argue that the optimal number of DSTs can be determined by the formula “N - 1,” in which N represents the number of countries in the world and 1 represents the country that hosts almost all the major digital service providers.

The scope of DSTs will expand with the passage of time. Their reach will grow to include a broader range of activities and service providers. A good example is India’s equalization levy, which is functionally a DST and was broadened from its original scope. This type of swelling is bound to occur when local lawmakers regard the tax in question as falling on nonresidents. The perfect person to tax is someone who’s not your constituent. That’s how countries try to export their tax burden, a game as old as the hills.⁴

The point here is that the U.S. appetite for BEPS 2.0 depends almost entirely on the view that DSTs represent a calamity. That’s just not the case, and the election didn’t change things.

Dissection by Simplification

I admire what the OECD has been doing with the inclusive framework over the last few years. They get an A for effort. But the realist in me still thinks of the pillars as an exercise in pounding square pegs into round holes.

Note how pillar 1 has changed over the recent months in response to criticism that its design elements were overly complex and unworkable. In October 2020 the “amount C” concept was dropped, leaving amount A and amount B as the remaining means for reassigning national taxing rights. The change did little to silence the criticism. That one of the three tools for sharing taxing rights with source countries could be so easily dismissed reveals a fuzziness about the purposes behind the remaining amounts, and why they’re structured as they are. People are questioning what guiding principles are being served by all the complexity contained in the OECD’s blueprints. As one observer put it, the specific problem that amount A is trying to solve isn’t clear.⁵ If so, then why do we need it as a component that is distinct from amount B?

In hindsight, the OECD might have opened a can of worms when it dropped amount C, as if to dismantle a preserved trinity. One concession implies that others are possible. Now everybody is emboldened to propose an alternate plan, with each variation less convoluted than the OECD’s own blueprints.

Also, we’re hearing simplification ideas from unlikely sources. Lafayette G. “Chip” Harter III, the former Treasury deputy assistant secretary for international tax affairs, has spoken favorably of a German proposal that would address scoping issues under pillar 1 through a quantitative approach that draws data from available financial statements. That would be in lieu of the more subjective, activity-based approach in the blueprint.

A problem with scoping is that multinationals typically have diverse business segments, some of which would fall in scope while others fall out of scope. The German proposal would base scoping on discernible metrics, such as profits per employee and return on depreciable assets. The idea makes sense when you think about it. The more objective the criteria, the better. Why set ourselves up for endless bickering over which business segments constitute an automated digital service or a consumer-facing business?

On that point, Netflix is seriously arguing that its business model, based on streaming movies and television shows such as The Queen’s Gambit, as referenced above, does not constitute an automated digital service. Really? Is there any point in building a new international consensus if the tax base is going to be just as porous as the last one?

The use of financial statements is also preferable to relying on angel lists. Angel lists are inverted blacklists, and they seldom work out well because the selection process is too easily corrupted. Rest assured that digital service providers that can afford the best lobbyists will find a way to be included on angel lists and thereby fall outside the scope of the tax. The process by which such lists are formulated is rarely transparent. These lists are negotiated in smoke-filled rooms behind closed doors.

Another concept for simplifying pillar 1 was recently proposed by professor Michael Graetz. His recommendation focuses on the troubling way that pillar 1 requires differentiation between residual and routine profits, with only residual profits being eligible for inclusion in amount A. Graetz breaks down the policy arguments as to why routine profits should be off limits and finds them unconvincing. He would reallocate a portion of all profits (both routine and residual) of in-scope companies.⁶

The recommendation is generally consistent with the comments submitted to the OECD by the G-24 working group of developing countries. It might make sense to listen to the G-24 because they reflect the underrepresented constituency that has been poorly served by the status quo.

Do we really think tax administrators and taxpayers are going to agree to which bits of a firm’s profits are deemed routine or residual? I’m not suggesting that’s a distinction without a difference, but splitting hairs this way isn’t worth the hassle. A tax regime that’s not easily administrable cuts against the interest of resource-deprived economies.

It’s a bit like when residence countries suggest we can “fix” transfer pricing by requiring multinationals to submit boatloads of documentation, knowing full well that developing economies aren’t equipped to handle it. Such fixes do more harm than good, to the extent they take more practical reforms off the table.

I never liked the idea that pillar 1 should distinguish between routine and residual profits (for the same reason that I don’t like the qualified business asset investment concept contained in the GILTI regime). Requiring these distinctions is asking for trouble. The pillars are supposed to get a dedicated dispute resolution facility. They had better, considering how many controversies these rules would generate if enacted.

This Is Where You Resign

I hate to be the one to say it, but pillar 1 is doomed. Hopes of U.S. support are speculative fantasy. The OECD blueprint earns a distinguished spot on my bookshelf, right next to the Bowles-Simpson report on deficit reduction and other aspirational volumes that contributed much in the way of intellectual stimulation but never went anywhere.

Perhaps the fate of pillar 1 will become clear during the next gathering of G-20 finance ministers scheduled for late February. That would be an appropriate time for Yellen to clear the air about where things stand and what’s not changing with the new administration. The timing would give the OECD a few months to plot its next move before the self-imposed deadline of mid-2021. It might want to focus on whether pillar 2 is viable as a solo act. I have my doubts about that as well.

There will always be other fiscal challenges for the OECD to address. Given how prevalent DSTs are about to become, it might consider drafting principled guidelines for countries seeking unilateral countermeasures. I’m only half joking. Figuring out how to fold DSTs into existing VAT regimes would be a genuine public service.