INTERVIEW: Ruth Mason and Stephen Daly

This is a bonus post for paying subscribers. Sign up here for an Emperor Subscription to get full access, including past interviews and extra content.

Ruth Mason is a professor of law and taxation at the University of Virginia School of Law, and Stephen Daly is reader and scholar at King's College London. Both have written about the European Commission state aid investigations into the tax practices of multinational entities. I spoke with them on Oct. 7 about the Court of Justice of the European Court's decision in the state aid case involving Apple Inc. and Ireland.

I'm not going to pretend like I read the full decision, I think it was 400 paragraphs. It seemed like people were really surprised by it. Was there a particular sort of logic that ended up winning out, or a particular argument that seemed to sway the court?

MASON: The reason why it was such a surprise was that at the lower court, the Commission won on the law, and Ireland and Apple won on the facts. And then there was an intervening high court decision called [Fiat Chrysler Finance Europe v. Commission], which held that analogous reasoning to the reasoning that the Commission used in Apple to win on the law was an infringement of the Treaty on the Functioning of the European Union. So what you had was Apple and Ireland going into the appeal, they had won on the facts, and then there was this very strong statement in Fiat that they should have won on the law. So, observers said, ‘Well, there's no way Apple and Ireland can lose the appeal.’ So that's why it was such a surprise.

DALY: Just to add on to what Ruth said, the Fiat case tells us that to determine if there's state aid, you ground your analysis in domestic law. It didn't look like the Commission had grounded its analysis in domestic law. But then the court also went even further in the case of [Luxembourg and Engie SA v. Commission], if there is a dispute about what domestic law means, you need to defer to what the member state has said. So not only do you have Fiat on your side saying that you need to ground the analysis in domestic law, the court also said if there's a dispute, you should be deferring to what Ireland has said its law required. And Ireland said it never made a mistake. So it seemed inconceivable on Monday the ninth of September that Apple could lose this case. And yet, I woke up on the morning of the 10th of September to a message from a journalist just saying, ‘Wow.’ And I looked through my emails and saw the result.

MASON: So you might wonder, how did it happen? Well, the Commission had to appeal on an issue of law. You can't appeal the facts to the Court of Justice, because those are decided by the General Court. So the Commission appealed a question of law, arguing that the General Court had not applied Irish law correctly. Which is a great irony, because what the Commission had reasoned in its final decision against Ireland was that, not Irish law, but rather EU law and [Organization for Economic Cooperation and Development] guidance ought to apply to determine whether Ireland granted state aid. So then we come to the Court of Justice, and the Commission suddenly is arguing that the General Court misapplied Irish law, and the Court of Justice agrees with it somehow. And it does so by saying, the General Court held that Irish law requires this [Authorized OECD Approach], but the General Court had come to that conclusion by reasoning that was knocked out by Fiat and Engie. Fiat and Engie said that you can't, over the Member State’s objection and evidence as to what its domestic law is, apply external law. You can't apply the OECD standards. But the Court of Justice says, well, the General Court said you could. The General Court said Irish law requires this AOA. But of course, the General Court held that before the high court decisions came out in Fiat and Engie, saying you can't do that. So the Court of Justice had to ignore that very important legal development. The Court of Justice referred to that particular issue—the General Court’s holding that Irish law required application of the OECD guidance--as res judicata. And then the Court of Justice said: ‘Well, what Irish law requires is was decided by the General court, and since Ireland didn’t appeal it, it stands as a matter of res judicata. So, yes, in accordance with the decision below, Irish law requires the AOA. But the General Court applied it wrong.’ And that's how you get to this decision at the high court that unravels the General Court's decision.

It’s a $14 billion decision resting resting on a procedural technicality—a failure to appeal a particular legal issue—where the legal issue that should have been appealed was established after all the appeals were in.

DALY: The res judicata point is just so slippery, because Ireland and Apple are barred from making an argument that they never had the opportunity to make, because the case of Engie, the case of Fiat, they all postdated the appeal. So the European Court justices slammed them for not specifically identifying an argument that they couldn't possibly have identified at the time that they made their appeal. The second thing is the European Court of Justice cites this case of [European Commission v. Fútbol Club Barcelona] for this point that if you fail to appeal a point from the General Court, then you cannot raise it later. But that case is very different, because in Fútbol Barcelona, what happened was the parties appealed to the General Court, they lost the point, and then they failed to appeal again. It’s very different here, because Ireland and Apple, they won the case at the General Court. And in fact, the General Court's reasoning is kind of contradictory, so you can pick and choose depending on which side you want to win. The General Court both agreed with Ireland's analysis of what Irish law required and agreed with the Commission's analysis of what Irish law required. So the two cases are very different.

There is a third point, which is that the principle of res judicata is not absolute. If the Court needed an excuse to reopen this issue, as Professor Mason has already said, the reference framework was raised as a point of appeal, so the Court of Justice could well have used that as a basis for reopening this issue, and then properly applying Fiat or Engie. And also, the European Court of Justice, in its own judgment, doesn't faithfully, rigidly apply res judicata, because it has reopened an issue later on the judgment. About, specifically, how the General Court had conflated selectivity and advantage. A point which neither party had appealed, and yet the European Court of Justice found reason to still reopen that issue.

MASON: Moreover, if you take seriously the decision of the Grand Chamber of the Court of Justice, that’s 15 judges, in Fiat, they describe the lower court’s application of an external standard as an infringement of Luxembourg's entitlements under the treaty. That is, the Court of Justice held it was a violation of the treaty for the General Court to apply this external law. What part of the treaty did it violate? It violated the part of the treaty that reserves to the member states the entitlement to define their own tax law. It violated the sovereignty of Luxembourg for the commission to use this external OECD standard. Why then would the Court of Justice, when faced with exactly the analogous reasoning, not want to avoid infringing the Treaty itself, or avoid having the General Court infringe the treaty? Instead, the Court of Justice upholds as a matter of res judicata a thing that it had said two years previously constituted an infringement of the treaty.

So if I’m following you right, this is an interesting issue. One of the things that I wanted to follow up on was, does this case signal a fundamental shift in the balance of power, both at the EU and also sort of between the OECD and the European Commission? Does this case establish that the European Commission is a de facto tax authority for Europe? And it sounds like the court’s own decisions are sort of mixed on this, that it’s not totally decided yet.

DALY: One thing that I've been warning for years about this litigation was the premise of the Commission's arguments. Because the basic premise is, member states got their law wrong. And I said that that’s quite tricky territory that you're going to be getting into, because if you can establish that simply showing a misapplication means that there is state aid, then misapplications can happen not just when rulings are provided, but also any time there's an interaction between the tax authority and taxpayers. So whether it's guidance, settlements, any kind of advice that might be relied upon–but conversely, you can also apply to it omissions. So any time that a tax authority passively accepts a tax return which has been submitted on an erroneous basis, all of those are examples of misapplication. I suggested instead that you should need to demonstrate both a misapplication and something more before you engage state aid analysis.

But that's the problem. Once you have a misapplication amount to state aid, that means that the Commission can investigate any time there's a potential misapplication. You're right, the Commission then becomes a de facto tax authority, and then the Courts of Justice become de facto tax courts. It wasn't just me that warned about this. It was Advocate General [Julianne] Kokott, the best tax advocate general, she also warned about this possibility. So I think that's problematic for various reasons, but a simple one is this: the Commission is not a tax authority. It is not specialized in understanding the tax rules and collecting taxes. And the General Court and the European Court of Justice are not specialized tax courts. So that is a problem, and we see that being a problem in this case itself, where the Commission is getting tax law wrong and the European Court of Justice is also getting tax law wrong.

MASON: I agree with all that. Moreover, what you have here is inexpert institutions applying tax law in a way that's highly politicized. In a way that seems quite unfamiliar to people who are accustomed to national tax administrations where–at least in the United States, my impression of Treasury is that they're always trying to get to the right answer on the law. They try to make sure that taxpayers don't engage in avoidance, yes, but it's always within the bounds of the law. I don't get the sense that Treasury is trying to twist the law to arrive at a desired outcome regardless of the law. But in this case, there's lots of evidence that the Commission and the EU courts were twisting the law to get to a result, and that is a cause for concern for other member states that are now subject, potentially, to similar actions by the Commission. But it's also a real warning to third countries that in any reform efforts they're making with European countries, they should try hard to make sure that the Commission and the European courts are kept out of tax.

So is there an argument that, looking at this from a big-picture point of view–there is the slippery slope issue, there are a lot of these legal technicalities, but in a common sense way, the in the Apple case the evidence was clear, and the details of it were striking in a way that isn't true for most cases, and it either shows that the arm’s-length standard was being flouted or that it's ineffective?

DALY: It was quite an aggressive tax structure. So what Apple did was it got the IP out of the United States, the non-U.S. IP out of the United States, and gave it to these Irish-incorporated, but non-resident companies. And because the beating heart of Apple's business model is its intellectual property, that meant that the profits were ultimately attributable back to those companies. And so all that would be taxable, would be whatever is done in these tiny Irish branches. So it's quite an aggressive tax structure.

MASON: I'm of two minds on this, I think Steve and I agree that there was state aid here. It's obvious that there was tax avoidance here. That it was obvious should have made it easier for the Commission to prove it using the law. Steve and I have described several methods in academic work that would have followed the law but still found that Ireland conferred state aid. What happened here was that people in the Commission and people in European institutions became convinced that Apple had gotten away with something, and that Apple should be punished. It’s possible that Apple got away with something, but that EU law does not provide a way to punish that. And if it doesn't, you can't twist the law to punish Apple just because Apple deserves it.

It is dismaying to see how little concern the EU institutions show for their own law. I have just re-read all the materials in the case from the beginning, and the impression you get from the Commission is that it was aiming at a particular outcome and treating Irish law as an obstacle to be surmounted in that goal. And then the Court of Justice accepts the Commission’s analysis, even though it was wrong on the law, and blatantly so. Yes, Apple was aggressive, but the outcome here, although it punishes Apple, and in some sense that may be just, what it also does is put a substantial portion of Apple's global income into Ireland. And that doesn't seem right. That does not align with economic substance. Even if Apple was engaged in tax avoidance, Apple was not principally engaged in Irish tax avoidance. But the outcome of this case is saying, ‘Apple, you avoided Irish tax,’ and that's nonsense. It did use Irish law to help it avoid taxes of other countries. If you want to stop corporate tax avoidance, you have to go after it in a more direct way. You can't just pretend that Irish law stopped this Apple tax plan.

DALY: There's a really fundamental point in all this, which is that the state aid rules are not designed to combat tax avoidance. Because if this is just a pure tax avoidance case, the tax is ultimately being avoided in the United States. To my mind, it was the United States tax rules, or their application in the U.S., which enabled it. The first was, you had the deferral system–so as soon as you got the money outside of the U.S., it wasn't taxed, so long as you didn't bring it back in. The second was that you could even shift the intellectual property to these letterbox companies, companies that had very little substance. And apparently they were supposed to be involved in this very complex cost-sharing arrangement, which is nonsense. And the third thing is the idea that there was management of the intellectual property in the United States, but that somehow, that didn't mean there were permanent establishments in the U.S. of these Irish-incorporated, non-resident companies. So if their true concern is tax avoidance, then it's one of those three mechanisms, or possibly all three mechanisms, which led to the avoidance of tax in the first place. And then you can think of a thought experiment–well, if the U.S. had, in fact, taxed the profits, would there still be state aid? Surely, none of the facts differ as to how Apple uses Ireland's tax code. But would it still be state aid? I don't know.

MASON: Another way to look at this is that the EU institutions have, after-the-fact, ratified Apple's tax plan, at least vis-a-vis shifting out of the United States, by saying the income is Irish, instead of American.

You mentioned the deferral of US income. That gets to something else which I wanted to ask about, the fact that this tax structure no longer exists. It wouldn't be really possible, or at least in a practical sense, it wouldn't be something that you would do today, in the post-TCJA world, and with the changes that have happened in Irish law. So what do you think the impact of this is going forward, given that a lot of the tax structures that are similar to this are no longer in place?

DALY: In terms of looking forward, you can still look back. So the European Court of Justice has said this structure did not work as a matter of Irish law to reduce taxes. So it's an open question then as to how many taxpayers had adopted this structure in Ireland. We don't know. We were told that there were 13 multinationals that had applied for rulings between 2004 and 2014, who had used the same structure in Ireland. But that doesn't mean that they were the only multinationals. They were just the only multinationals that had asked for clearance in advance. There could be countless other multinationals that had simply submitted their tax returns relying upon this structure. That's one point to note. The second thing is, yes, it is a historic structure in the sense that you could not adopt this structure of having stateless companies with intellectual property after 2015. Ireland legislated it out in the 2014 Finance Act, this ability to create stateless companies, with effect from 2015. But there was an allowance for those companies that had used this structure, pre-2015, to restructure themselves up to 2020. So the years that are relevant here are 2014 to 2020. The Irish Revenue Commissioners can only look back four years as a matter of Irish procedural tax law. That means that any multinationals that had adopted the structure in Ireland are probably clear of the Irish Revenue Commissioners as a matter of Irish procedural tax law. Unless, of course, there were open investigations in relation to any of those multinationals. That's the first thing in relation to limitation rules.

The European Commission's limitation rules are different. They can look back 10 years. So they can look back to 2014 to see if there were any multinationals that had adopted the structure. Is the Commission barred? I don't think so. So what those 13 lucky taxpayers that had the rulings could try is, they could try to suggest that they had the legitimate expectation that the commission would not investigate them, or that the principle of legal certainty would bar the commission from doing so. And I don't think either of those legal doctrines would be of great assistance unless they had a specific assurance from the Commission that they wouldn't investigate.

So you’ve got to look backwards to look forwards.

MASON: It's bizarre, because what you have here is a General Court decision saying what Irish law is, but the General Court used a method to draw that conclusion that is, after Fiat, impermissible. And then you have a Court of Justice decision affirming that and saying, ‘It's true that Irish law requires the application of this comparative approach, and because Ireland didn't apply the comparative approach in the Apple case, Ireland must recover from Apple.’ But the decisions of constitutional courts, the Court of Justice included, don’t just say what the law is today. When the Court of Justice gives an interpretation of EU law, that interpretation is not only prospective. It tells us what EU law has always meant. So we have Apple and Fiat, which point in different directions, but which ostensibly both represent valid interpretations of EU law.

If you're the Irish Revenue Commissioners, do you follow Apple and reopen all of these rulings that you didn't analyze in a comparative fashion? And what do you do with Fiat and Engie, which say that what matters is domestic law, and Ireland is the best interpreter of Irish law? Put differently: should Irish Revenue and the Irish courts follow Apple or Fiat/Engie? Apart from what the Commission is going to do, must the Irish tax authority avoid state aid by complying with this wrong EU pronouncement about what Irish law is?

And this is why it's so important in cases like Fiat and Engie that the Court of Justice held that you have to use the member state law, because otherwise it just becomes really weird. Because there can be significant differences between how EU courts and national courts interpret national law, and EU courts really aren’t competent or empowered under the Treaty to interpret domestic law. EU courts’ misinterpretation of Irish law produces a rule that applies only to Apple. Irish law applies to every other taxpayer in Ireland, but Apple gets special-made law by the Commission and the General Court of the EU. And so what are the implications now? I think the Court of Justice has been careful to try and limit the effect of the judgment in Apple. First it comes out with Fiat and says you have to use domestic law. Then you have this little detour in Apple. But then right after that, you get the U.K. [Controlled Foreign Corporation] case [United Kingdom v. Commission], which says, no, no, you can only use domestic law. So you basically have this special decision that just applies to Apple. Again, people should be worried about that. People should look at it askance.

DALY: In the Fiat case, in the Engie case, in the [Luxembourg and Amazon v Commission] case, in the U.K. CFC case, and also in a series of cases concerning turnover taxes–those were cases against Hungary and Poland–the Commission got slapped back by the European Court of Justice because the Commission was invading member state sovereignty, over how you determine the fundamental rules of your tax system. And then in Engie, how you interpret your own tax rules. So the European Court of Justice is championing member state sovereignty, except when it isn't, except when it comes to Ireland. So not only has Apple been singled out for special treatment, Ireland has also.

It’s like the European Commission is the new tax authority for Europe in the case of Apple, but not most of the time.

DALY: Yes, exactly. Special Apple tax collector.

So one thing that people have said about this a lot is that it’s like you have to arm’s-length standards, you have the OECD’s, and you have the European Commission’s, and they might apply differently. Is that going to be an issue for companies going forward?

MASON: I don't think so. I think that's a special problem that Apple has, and no other taxpayers should expect to have it, because the Court of Justice has really gone out of its way to say ‘No, apply domestic law.’ And even in the Apple case, the Court of Justice tries to distance itself from what actually happened in that case, which was the rank application of external law by the General Court, and its reasoning.

So when the Court of Justice talks about the res judicata issue–the part that was decided by the General Court, the part that the Court of Justice is not going to look at again–issue that is res judicata is the General Court holding that Irish law required OECD standards. The Court of Justice is saying, effectively, ‘We didn't decide that Irish law requires the external standard. The General Court decided that, and nobody complained.’ But then you have all these other cases saying you're not allowed to apply external standards, including the CFC case, which comes after Apple. I've been surprised before, but I would be surprised if the Commission started bringing this argument again, that it can just apply whatever standard it wants, irrespective of domestic law.

Contact the author at amparkerdc@gmail.com.