Oscar Wilde famously said that the only thing worse in life than being talked about is not being talked about. In yesterday's NYU Tax Policy Colloquium, our first of the year (in our 26th year) and the first ever on Zoom, I had a somewhat different experience. I found that the only thing better than being needed is not being needed.
To explain: I was the lead commentator, so I offered introductory remarks, which my co-convenor Lily Batchelder then supplemented, and to which our author, Steven Dean, then responded. But towards the end of Steven’s comments, the Internet in my house crashed, taking me off the session entirely. Luckily, Lily ably kept the ship afloat by running the queue. I eventually got back into the meeting by downloading Zoom on my phone and rejoining that way, but phone Zoom has much worse functionality (plus I’m unfamiliar with it).
This meant that I missed a lot of the discussion, from what was probably our largest ever colloquium audience (I think we topped out at about 68 people). This was so much the worse for me, but I hope no problem for everyone else. (Back in the past when we had smaller audiences, I used to participate actively throughout the sessions, but in recent years, people in the audience have had enough to say that it’s generally best for me to pipe down after my opening comments.)
I suspect I am not the first ever person teaching a class on Zoom to have the system crash. Fun! Obviously, with a regular lecture or seminar class and no co-teacher, this would be pretty bad.
Anyway, the title of the paper is A Constitutional Moment in Cross-Border Taxation. Here is an overview of what I raised, with additional context added as well.
The paper is part of a book project that Steven plans to pursue when (so much the worse for us at NYU) he has returned to Brooklyn Law School rather than running our tax program. The project’s main aims, as the paper makes clear, are normative. Dean (I’ll switch to the last name from now on, for convenience) wants international tax rules around the world better to promote social justice, both within a given country (progressivity) and between countries (the global South and marginalized, low-income countries, versus the global North and powerful, high-income countries).
This paper, however, is mainly descriptive. Its chief premises, as I read it, are as follows:
1) International tax law around the world is dominated by what one could call (slightly modifying the paper’s terminology) a “classification and assignment regime.”
2) This is substantively a “constitutional” regime, even if not formally so. (The paper combines these two points by discussing a “classification and assignment constitution,” but I find it useful to separate the two pieces.)
3) This constitutional regime is subject to informal but effective amendment at what Bruce Ackerman called “constitutional moments.”
4) The world may currently be at such a moment.
And finally, back to the normative:
5) Multinational companies (MNCs) should pay more tax on their profits, and marginalized / low-income countries should both get more of the revenue, and have more say regarding rule design, than they now do.
I’ll turn now to specifying more fully what these claims mean in the paper, and then commenting briefly on them.
1) The classification and assignment regime –
This term refers to tax systems’ first “classifying” income (e.g., as belonging to a cubbyhole such as interest, dividends, royalties, active business income, etc.), then “assigning” it for tax purposes to a given country or else to the given affiliate of a multinational company (henceforth, MNC) that the MNC most likely exclusively uses in that country, if it has a permanent establishment (PE) there.
In particular, however, it rests on the importance to tax planning of separate entity accounting within an MNC’s affiliated group of companies, accompanied by the use of transfer pricing and includability/deductibility for intra-group cash flows (such as interest and royalties) to determine each affiliate’s share of the overall group income. This regime leads naturally to the easy creation of stateless income, along with the enhancement of tax competition between countries.
What isn’t or wouldn’t be part of the classification and assignment regime? Examples include the following:
a) Unitary taxation of the entire worldwide corporate group, with some method such as sales-based formulary apportionment (FA) being used to determine what group income arises where.
b) Possibly, although this is debatable (as I discuss further below), such prominent recent features of the international tax landscape as the enactment of digital service taxes (DSTs), the EU state aid cases such as that between the European Commission & Ireland with respect to its arrangement with Apple back in the day, and ongoing OECD-BEPS initiatives.
c) Depending on how one ends up deciding to use the term, deemed dividend rules might be viewed as inconsistent with the CAA regime. Once you have deemed dividends, as under controlled foreign corporation (CFC) rules such as the US’s subpart F, or likewise under the US’s GILTI rules, then in effect you have taken a step towards unitary taxation. E.g., if ALL of the CFCs’ profits are treated as deemed dividends to the parent, then the prior steps of applying CAA have been pretty much overridden. As it happens, the existing deemed dividend rules are more limited than that in their scope.
In terms of the semantic answer to the question of whether existing deemed dividend rules are part of CAA, I think the answer is Yes if you are using CAA to refer to the existing regime, but it’s No if you’re asking whether a CAA methodology is being used in full. The deemed dividend rules still have Classification, but they override the formalism of Assignment.
2) It’s a Classification and Assignment Constitution, not just Regime
For the point that is being made here, we can start with Reuven Avi-Yonah’s prominent argument, first made a number of years back, that there is such a thing as the “international tax system.” The cynics had tended to say: “No, there isn’t. All we have is a whole bunch of countries with their own rules.” But Reuven argued, with analogies (for example) to customary international law, that there is a coherent whole with continuities and consistent principles. David Rosenbloom prominently disagreed, and I’ll discuss further below the context in which they had this debate, and why they agreed that it mattered.
I read the Dean paper as saying, yes, there’s a coherent system, and not only that but it’s materially a constitutional system even though there is no formal or written global international tax “constitution” as such, nor are there courts enforcing it.
To give an example of what I think the paper has mind by way of a constitution, consider the possibility that Congress will enact a wealth tax in 2021. (Just a hypothetical; I’m not expecting this to happen.) It would be litigated straight up to the Supreme Court, where it’s a pretty good bet that the 5 committed Republican justices (yes, that’s how I view them) would vote together to find it unconstitutional.
We would then have learned ex post that Congress could not implement a wealth tax (at least, without apportionment between the states), because the relevant actors within the system relied on the Constitution to strike it down.
In positing a constitutional system in international tax, the paper posits that there are similar constraints in place here, albeit not enforced in the same manner as US constitutional limitations on the domestic taxing power.
3) Constitutional moments
Bruce Ackerman famously posited that the US Constitution can effectively be amended, without formal resort to the Article 5 process, at what he called “constitutional moments.” The basic idea is that, at a period when there is great public focus and debate on what the Constitution permits, one of the parties (say) endorses legislation that would be unconstitutional under the current understanding, wins the requisite elections, implements, the other party eventually assents, etc.
In illustration, suppose we know from the history of Supreme Court jurisprudence that minimum wage laws were unconstitutional in 1920 but constitutional in 1950. Without Ackerman’s constitutional moments idea, it would seem that one or the other of these pronouncements must have been wrong. But under the Ackerman theory they could both have been correct when decided, with the reversal reflecting informal but nonetheless legally effective amendment of what the Constitution means in the 1930s.
Key examples he has in mind include the New Deal and the Civil Rights Era. I seem to recall that Ackerman pronounced the “Reagan Revolution” to have been a failed constitutional moment.
In the paper, Dean cites a couple of examples of his own. He views Justice Marshall, in the 1803 Marbury v. Madison decision, as having effectively amended the Constitution, in an analogous way, by proclaiming judicial review (which, at a minimum, had not previously been universally accepted). And in the international realm, he views the US enactment of subpart F in 1962 as having perhaps been effectively a constitutional moment, amending the Classification and Assignment Constitution as it had existed previously, by providing that one could modify or depart from it by taxing deemed dividends. This one took, just like New Deal legislation in the aftermath of the famous “switch in time that saved 9,” in that other countries not only accepted subpart F but enacted their own CFC rules.
Next, here are some of my main thoughts in response to the paper’s analysis:
1) Do we have a classification and assignment regime?
Again, deemed dividend rules for CFCs, such as subpart F and GILTI, appear to be inconsistent with it. Also, if FA is inconsistent with it, note that transfer pricing rules often have a formulary component. And one could certainly argue that DSTs, the EU state aid approach of the European Commission, and OECD-BEPS are all consistent with CAA, and merely apply it differently.
Avi-Yonah, when he argued that there is an “international tax system” (albeit, he didn’t say “constitution”), was making a normative argument for US “anti-hybrid” legislation that he argued was in the spirit of the existing regime. (Taxpayers exploit hybridity by structuring their affairs so that different countries’ rules apply inconsistently as to the same concept, e.g., Affiliate A’s cash payment to Affiliate B is deductible interest in A’s country, but isn’t includable interest in B’s country). So he was saying: There’s a coherent regime, it’s a good one on balance, hybridity violates it, and therefore, as a normative matter, we should voluntarily enact new legislation that helps preserve its intended functioning.
The paper, by contrast, is saying that the regime exists as a descriptive matter, and that, because of its “constitutional” status, countries are disempowered (whether we're glad of this or not) from violating it.
2) Is this regime a “constitutional” one?
This raises the questions of what the enforcement mechanisms are. Not a higher court, so presumably the difficulty of enacting a rule that departs from the norm without running into ineffectiveness and pushback.
Would convergence and imitation, creating similar rules in different countries and sustaining the similarity as rules change first here then there, be enough for this to be a “constitutional’ matter? And what is at stake when we address this question? How does it affect the analysis? These questions will, I’m sure, be further addressed in the book manuscript.
3) Constitutional moments
Ackerman’s descriptive claim is not only obvious, but meant to be. We knew before reading his work that, yes, the minimum wage was unconstitutional according to the Supreme Court in 1920, and constitutional in 1950. And he counts on our knowing this. The novel claim that he is making is normative: he’s saying that this reflected, not correct vs. incorrect constitutional interpretation, but the fact that the Constitution had been amended in between.
One question one might ask here is whether this interpretation risks tautology or non-falsifiability. A country proposes something that’s rather novel. Then it either retreats, showing that it faced a constitutional barrier, or it sticks to its guns & shows that the CAA has been amended.
Are there simply livelier and duller moments in the evolutionary process for the international tax regime, such as it is? And again, does it matter for the analysis whether there are constitutional moments, rather than just periods when the observed regularities are changing fast rather than slowly?
4) Back to the normative side: advancing social justice
Again, the central concern motivating the paper is with social justice, defined both in terms of progressivity within a given country, and the interests of the global South as distinct from those of the global North.
The paper’s analysis arguably suggests that what it considers the injustices rife in the current system are mainly a legal problem, and hence also political in the sense of the political factors that drive legal and constitutional outcomes. (E.g., by analogy, the US Supreme Court would be making a legal claim if it struck down a wealth tax, and its so deciding would reflect the politics of Supreme Court appointment and confirmation.)
As one who is somewhat of an economist fellow traveler, albeit a law professor rather than an economist, I have been inclined to view the distributional characteristics of the existing international tax state of the play as reflecting economic forces, along with countries’ incentives when they have limited jurisdiction and are “selfish” (i.e., concerned mainly with the welfare of inside players).
From that perhaps more cynical perspective, why would we ever expect “marginalized” countries to get much sway? One might think they deserve it, without expecting them to get it.
Also, are many of the aspects that the paper suggests are non- or anti-CAA regime actually more orthogonal to the global north vs. south debate than directly relevant to it? E.g., a lot of those items (DST, EU state aid cases, much of OECD-BEPS) are arguably about the competing interests of “producer” and “consumer” states (with respect to highly profitable MNC activity) within the global North. And while those efforts are often hostile to a third set of countries, tax havens, the tax haven countries are far from being coterminous with or identical to the global South.
POSTSCRIPT: As an aside, I am glad for Steven, but very sad for NYU and myself, that he will be leaving us at the end of the year (he has been the director of our tax program) to return to Brooklyn Law School. I'll miss him, although hopefully still see him frequently, at least via Zoom. One of the reasons I so value him, not just as a friend but as a member of our tax community, is that he brings what in our environs is a unique, albeit badly needed, perspective on international tax policy matters. It's so hard around here to escape from just getting US perspectives on things. Through international travel and connections, at least before the pandemic, I was able to encounter and learn from UK and EU perspectives, and to a lesser degree those from certain leading Asian countries. But Steve adds a global South perspective that is really an important part of having a balanced, or at least less parochial, viewpoint.
