 Now we would like to give the floor to Paul John Lohenthal, who is turned into legal services of the European Commission. Thank you very much. My name is Paul John Lohenthal. I'm a member of the legal service of the European Commission. At the legal service, I work in the state aid team, about 12 of us, and I am responsible for the past four years on fiscal state aid matters. Which basically means that the cases that have been discussed till now, I have been working on them for the past four years. I know them inside and out. I will not comment too much in detail, since this is all ongoing litigation in which I am representing the European Commission. But I've been asked more to look into the notion of state aid and how fiscal state aid fits into the basic notion of state aid. In relation to recent case law and decisions that we have recently taken. Now I was very happy to hear before my presentation that fiscal state aid is legitimate. So we are legitimate in looking at fiscal state aid because otherwise I would be out of job. At that point I just wanted to highlight the judgment that I always liked. It was a judgment from 1973 by the Court of Justice dealing with cases, Italy vs. Commission, where Italy had given tax breaks to its manufacturing sector. And the Commission deemed that to be state aid. And Italy argued, no, we have fiscal sovereignty. We are able to determine our tax policies and how we want to benefit different sectors of the economy. And the Court said, no, this is the same as just granting a subsidy. So fiscal state aid is certainly legitimate. Now, taking that to how our recent cases and other cases fit into the notion of state aid. The question I heard very often today is does it stretch the state aid rules? And like Professor Smith, I will jump to my conclusion, no it does not. Here's the reason why. So Article 1071, we know it. I've broken down here into four conditions for finding of state aid. As regards to fiscal state aid cases, I know there were some mention of distortion of competition. It being, let's say, too easy to show a potential distortion of competition. On that, let's remember that fiscal state aid is basically, I would say in almost all circumstances, operating aid. Simply operating aid. And the Court has said operating aid is not compatible. It is obviously distorted. There's basically a presumption of distortion of competition and the effect on inter-eval trade when we have operating aid. I will focus mostly on the condition of selectivity and advantage, which I refer to here as selective advantage, and the Court does as well. As a matter of fact, we look at the case law of the courts up until, I think it's actually just the world duty-free judgment. I think it was one judgment before that, when the Court set out the four conditions for finding state aid. And in that list, we did not see the word selective. It just said it must be measured as imputable. It must be a distortion of competition, effect on trade, and an advantage. And then, of course, selectivity is a condition for finding state aid. All of a sudden appears and a selective advantage. So what is a selective advantage? Now, the mole judgment says rightly that a distinction should be made between selectivity and advantage. Now, the U.S. White Paper, which was referred to before, took this to mean that when you assess selectivity and advantage, you must assess them in a decision completely separately from one another. Now, let's first see what these two notions actually mean. Now, advantage, as mentioned here, is that your net financial decision has been improved by measure. Selectivity means it has been improved as compared to comparable undertakings. Undertakings in a comparable factual and legal situation. And the reason why we make this distinction is just to stress that not all advantageous measures lead to state aid, which I think is obvious. If you have a general exemption, general tax exemption, applying across the board to all companies in a comparable situation, you don't have state aid. An advantage, you don't have state aid. Now, the last point I made comes back to the first point. When we think about the three-step selectivity analysis, which I will get to in a second, where what's asked of us is to look, what is the reference framework? First, what is the benchmark against what we determine that there is selectivity? And the second step is to ask, has there been a derogation from that benchmark? Now, let's think about that very, very precisely. Let's say we have a provision, a provision that basically is saying to us, this is the manner in which a company should be taxed in that situation. And then we already have a derogation saying, you are not taxed. You get a tax exemption under these rules. Isn't that tax exemption in the end? The granting of the net financial position, so improving the net financial position, it's the tax exemption. And is it also the derogation from that rule? And what I mean to say about this is ultimately, isn't that test the same? Isn't it actually artificial to break this apart in fiscal state aid cases? What I mean by that, we go later on to the, let's just take an argument of what a lot of the member states are saying in the state aid cases. They're saying the benchmark you need to apply is our arms length principle rule in our national legislation. And then you need to determine whether we derogate it from that rule. And what does that rule say? That rule says you must price transactions at arm length. If you don't price transactions at arm length, you grant an advantage through reduction of the tax base. So the derogation and the advantage are the same. Now, I ran a little bit ahead of myself, but the first main question here, of course, is in these cases, how do we establish selectivity? And here, a big distinction that the case law has made and emphasized over and over again in many recent cases. I think of the mole judgment here, the derogation one I mentioned. The Belgian BSE test case, I think World Unifreeze also stresses it at the outset is distinction is made between individual aid measures and general aid schemes. We have an individual aid measure. If we can show an advantage, then selectivity is presumed. The presumption is not an irrebutable presumption, but the idea is if you've granted only to one entity, one taxpayer an advantage, then obviously you have selectively favored them compared to other taxpayers. And when it comes to general aid schemes, we have what was devised. I think possibly for the first time with the Pink Grausels judgment way back in 2008, the three-step test. And that basically requires you to follow these three steps with the first identification of the reference system, let's say the benchmark, the normal taxation system, the normal tax regime, then showing that there's a derogation from that reference system. There's an exception has been made. If you can show the first two steps that have been fulfilled, you have what's known as prima facie selective. And then seeing whether or not that derogation can be justified again by the nature and logic of the references. Now I imagine here, this is a point that was made recently in the airport Lubeck judgment, that the three-step test also applies to non-fiscal state aid measures. That wasn't clear before the judgments. I think that's wrong. I think it's wrong for two reasons, which I mentioned here. First of all, Lubeck, a bit of the background of Lubeck, was about the airport charges that airport Lubeck was setting, the landing rights and landing fees. And the commission had deemed that to be a selective advantage granted to airlines that landed at Lubeck airports. The court disagreed with the commission, basically pointing very strongly to the fact that under German legislation, the airports had their own autonomy, that they were separate from other airports, and therefore what we had to do was to see whether specific airlines, specific operators, undertakings at airport Lubeck were being discriminated vis-à-vis other ones. And the fact that airport Lubeck had better charges than, let's say, the airport of Berlin is not a selective advantage. To me, that's really much more a question of regional selective. And the question you have to ask if you read that judgment is, this works in Germany, a very large member state. But what if it was an airport in Luxembourg? What if Luxembourg was the one determining, the member state itself was determining the landing rights for different airports? What if, as in Spain, there's one entity determining the landing fees for all airports? There's the same reasoning applied. And now we have a very recent judgements going back on the streets that test here, and how I find it a bit problematic is the Retabelle judgements, which deals with the digital, professional television saga that happened first before the general court, and then at the court of justice. The court of justice upheld the commission's decision saying that there had been a selective advantage granted to providers of broadband televisions and digital television, vis-à-vis satellite television, the court just upheld that. But in one of the judgements, in the Retabelle judgements, the court annulled the decision for saying that the commission had not properly shown that these companies were in a comparable factual and legal situation. The most important thing is that you cannot say, and not to escalate the case, you cannot say that the selectivity commission is automatically satisfied if the measure applies exclusively to a specific sector or specific region. And that is contrary, I think, to a lot of the cases that came before, because just at all in the beginning, Italy adopts the measure that is selected only for its manufacturing industry against which benchmark is that supposed to be compared to? All sectors, all industries, do we need to set up a reference system saying, well, the manufacturing industries aren't in a comparable situation to all industries? And therefore here, where you get subsidy, where you get exemption, there's all the sudden some kind of selective treatment. I find it a bit confusing. So I want to move on back to fiscal state aid. And for fiscal state aid, I think a very important case that came on recently is the world of free judgements, which basically dealt with the Spanish measure that allowed undertakings that had made transactions in foreign undertakings to make deductions to the goodwill that they had acquired there. That was found by the commission to be a selective measure because domestic undertakings, sorry, domestic transactions, undertakings that had made the same kind of acquisitions in domestic undertakings could not benefit from that same deduction. Now the general courts annulled the commission decision, finding that there was no selective treatment there because the commission had not set out a specific category of undertakings that benefited from that measure. The commission appealed and the court of justice upheld, or annulled the judgment of the general court and basically upheld the commission decision for now, to send out the general court. What's very useful from that judgment is the first point I put here, and that is where you see the courts perhaps shifting a bit in its logic, its internal market logic, of what matters for selectivity is discrimination, discrimination of comparable situations. That's what it's about. I mean, we often talk when we have the references and the interrogation, but what they're saying is it really comes down the first two steps about discrimination. And then to put it back into what all the case law says before, a derogation from the reference system constitutes discrimination. Is that correct? I don't know, but it's really all the effort that's going back to discrimination. And here is basically the points that the court has kind of said, look, any undertaking can engage in this transaction. Anyone can make a foreign acquisition. So how can this be selected? And they say the fact that any undertaking can engage in that does not mean that it cannot be selected for that reason alone. I set out there's a three-step test, but honestly when you think about it, this is really a three-step test or two-step test. It's a very good opinion by a general go-back in the Belgian BSEing test case, it's a C-275, where he says actually it doesn't matter, it's the same thing. And perhaps academically it's more sound to really speak of the two-step test and the three-step test. The first step being, is there discrimination between comparable undertakings and how do we determine they're comparable, which we'll get to, but how do we get to that comparability? First of, is there discrimination between comparable undertakings determined by the objectives of the legal system, the reference framework under which that measure is granted? If there is, it's cream of fashion selective, and the second step is that discrimination justifies. Now the burden of proof, I point out here, the commission must show the first step, the first two steps from the last three-step test, whereas the member state has to show that it's justified. The commission can show it's discrimination. Now, the identification of the reference system, I think the statement was already made, in the first step, that the case law is very sparse on this. And I know because for the past four years I've been looking at it in detail, and it really is very, very well said about. I take somewhat offense to Abhijit General Walz's comment that when asked at the hearing how we had identified the reference system, we had no answer, because Richard Lyle, who if anyone should have an answer, spoke 15 minutes on that. So we were very surprised to read that in his opinion. But basically there is very, very little case law on this. And ironically enough, I think the best case that we have for this is the Gibraltar Judgment. The Gibraltar Judgment, if finally enough, is not applied recently. That's what gives the impression that when you have a tax system that is designed in a discriminatory way, there at the general, let's say the Gibraltar and corporate income tax was designed in a way that offshore companies would not be taxed at all because they just didn't have these taxable events and tax cases they need to be taxed, whereas onshore companies would. And I think what we can see in that judgment is that determining, basically when we come to the first step, maybe it's not so good to find, let's say, the normative rule, but to find out who is in a comparative and factual legal situation when we look at the objective of the measure that we're looking at, and the framework of which it forms a larger part. So what was said in the Gibraltar Judgment again is that determining which other things are in a comparable factual legal situation is key to defining the reference system. We need to know who is comparable to say, okay, what is the overarching principle that should apply to them all. And here the court says, you know, we're going to actually ignore what Gibraltar did. Gibraltar created an obviously discriminatory tax system to ensure that offshore companies are not taxed and onshore are taxed. And we're going to look at the objective is of a system like that, what their objective was to introduce a general system of taxation for all companies in Gibraltar, all of them, not just the onshore ones, the offshore ones too. And then they point out, of course, it's not a random consequence that the all shores are not taxed, that was deliberately done in that situation. Now again, when it comes to determining the reference system, the case law also is extremely contradictory on this point because when they speak about comparability, we have a number of judgments that say, I would say the vast majority that say that what matters to determine whether undertakes are comparable is the objectives of the reference system. We'll explain also why it's a good circular, but it is. But still, that is what we look at, the objectives of the reference system determine whether they're comparable. We have other cases, and I mentioned here the Adriavine one, which I think was mentioned earlier this morning on the energy charges and the Innoix case here in the Netherlands, where the course represents the objective to measure. Now the problem of looking at the objective of the measure is if a member state adopts a tax exemption for multinationals, and as well, the objective of the measure is to not tax multinationals. Therefore, only multinationals are comparable with the actual legal situation, no solitivity, no sitting, very easy. So it can't be that. And that's why I said, could it be both of them? Because it can also not just be the reference system, that's exactly what we're trying to figure out. So how can we determine the objectives of the reference system are if we don't know what the reference system is? So I would say in cases like this, by giving an example now of the tax rulings, what is the objective of a tax rule? The objective is to determine the taxable profit of a company that belongs to a multinational group. The taxable profit for which it focuses to tax it under the general corporate income tax system. That is the larger framework of which that ruling forms a part. Who is in a comparable factual legal situation in regards to that larger framework? All undertakings to generating profit in the member state. Any question? I refer here to the Russian doll. Now I thought it was very admirable that you explain the Germans' neurons' causal case because it's so incredibly confusing, but I would just refer to the Russian doll is from this opinion I mentioned from the general blowback, saying when you're looking for that reference system, you can just be, you know, basically opening a Russian doll. And what I think we get from the serious causal cases, you shouldn't take the largest, most broad references you can find. You need to go to the lowest level of generality. You have to ask yourself, what is the objective of this measure that we are investigating, where, for instance, this lost heart forward rules, the exception to that, and then the exception to the exception. The exception to the exception is what we're looking at. Then we look at the purpose of that within the framework of the exception. We don't have to go all the way up to the German corporate income tax code system. So on what cases like world to be free and say, sonorous causal and all these more recent cases come down to, has a shift to discrimination, actually heralded the novel approach. I would say the problem that we have and that is some very unique about fiscal state agencies, we don't really have a lot of cases, especially not about this. And the reason why we don't have a lot about this is that most of the cases that I've mentioned before, pink graphos, then OICS, the Adriavine, these are primitive references. And when it comes to determining a reference system in a state-aid investigation commission, commission must establish the first two steps, reference systems, derogation. But in a national case, it's a national judge, who generally has to do that. And that's not what the Court of Justice is doing when these questions are referred to. The Court is explaining to how they have to do not what it is. So we don't really see many direct practical applications with the exception of Gibraltar and World to be Free. So my conclusion is that volatility again comes down to discrimination of comparable situations and whether there's a justification for that. And I mentioned here at the end that there's an interesting question because we go then, let's say we're doing our analysis whether there's state aid, we have to determine is there discrimination? There's only state aid if there's discrimination because then you have selectivity. And the question is, is that state aid compatible? State aid cannot be compatible if it violates the fundamental principle of non-scrimination. Where are we going? What kind of discrimination? What does it mean? And I think the crux of the message that I want to send and that's why something we know is playing with the fiscal sovereignty of the member states. And I think the most important thing is fiscal sovereignty ends where discrimination begins. None of the decisions we have adopted attack Ireland's 12.5% tax rate. The double Irish Dutch Sandwich, these things are not looked at. This is fiscal sovereignty matters that Hungary now has a 9% corporate income tax rate. It may be deplorable, it may be bad for the European Union as a whole, but that's not a matter of state aid. What is a matter of state aid? A matter of state aid is that a company like Amazon which happens to belong, not even Amazon I would say, Amazon's operating company in Luxembourg, which happens to belong to a larger multinational corporate group can agree with another Amazon company that it will transfer 90%, 2% of its profits to that company that has no employees, no board, no functions, does practically nothing but holds entirely passively an IP that it gives to that operating company. Now for bookstore, when Amazon still needs to sell books, bookstore here in Netherlands can't do that. So bookstore here in Netherlands pays 25% of its profits. Amazon's operating company, Luxembourg, pays the fact of 5%. Is there discrimination? I don't think I have much time to continue so I will point out some of the interesting cases we have recently. One is here the Hungarian and Polish turnover tax cases which are a bit more in the line of the Gibraltar case law because what we have here is an inherently discriminatory system that is set up in order to basically punish larger undertakings, very often foreign-owned undertakings and has poled itself very very strongly and said what adopted the law in question is the Polish business. Small Polish business. We looked into it and we say well we have here discrimination that is caused by the way these progressive rates are applied to turnover which means that larger undertakings are taxed at a higher level than smaller undertakings. Can I quickly do the tax rules? Or do you want me to stop? Quickly. Tax rules, selectivity. For us there are individual aid measures granted to a specific company dealing of those companies in the specific situations individual measure we need to show an advantage. However, there is a desperate advantage according to the Court of Justice in the Form 187 judgment which deals with the exact same situation as all our tax ruling cases. Form 187 is about determining taxable profits of companies belonging to a multinational group. There it was done through a scheme through a formula here it was done through a tax rule. There an OECD method was used, and it was distorted in a way that allowed certain multinationals in Belgium that had a certain number of employees, a certain number of turnover, to pay less tax as we said in the decision than other multinationals but also than other standalone companies that pay arms and profits. The test they set is you need to compare the method for determining taxable profit with the ordinary tax system with the way normal standalone companies not forming part of a group are taxed under conditions of free competition that having been said all of our decisions contain some sort of very lines of reasoning but first we first say we do the three step testing we say there's a discrimination as compared to standalone companies under the ordinary tax system and then we look at the case of just companies belonging to a multinational corporate group and the specific provision in national law saying that they need to be taxed at arms and profits and we say there's discrimination there as well and the admiral decision also contains the third finding selectively aspirating what criteria do you use to determine the taxable profit of non-resident companies offering through a branch and five years later we still don't know again the benchmark is the treatment of independent standalone companies under the ordinary tax system which ordinary tax system, the ordinary tax system of the Netherlands, of Luxembourg of Belgium not an EU arms like principle I hear this over and over again here's all conferences it's not EU's arms like principle but then again there's no arms like principle there's no laughing on the arms like principle there's the arms like principle, it's a principle treatment of standalone companies the treatment of their transactions how their profit and their transactions are taxed under the ordinary tax system what's the right tax system separate energy approach plus the arms like principle and the reason why again the reference system has to compose with standalone companies and companies belonging to multinational groups that's the purpose of the arms like principle that's what the arms and principles are all about if we were to say the reference system just companies belong to a multinational company you could have a measure in a member state that says we tax companies belonging to a multinational corporation at zero percent no city we're all treated exactly the same but our standalone companies pay in Belgium 39.99 so is this all total political crap I think not these are our cases thank you very much and sorry for going