Noise in terms of the Commission’s investigation into purported breaches of State aid provisions by Member States has been increasing steadily over the past two and a half years. To recap, the Commission has opened a number of investigations (and produced some decisions) concerning rulings agreed between multinationals and various Member States, namely, Luxembourg, the Netherlands and Ireland. At the same time, it has produced a Decision in relation to a Belgian regime for excess profits. The latter is different in nature from the former investigations. The Belgian regime does not break new ground in terms of the jurisprudence on State aid law. The State is considered by the Commission to have effectively introduced a tax relief, in economic terms, for large companies. Such a selective advantage, if proven, undoubtedly comes within the scope of the State aid provisions.
It is the investigations into the rulings given by Ireland to Apple, by the Netherlands to Starbucks and by Luxembourg to Fiat, Amazon and McDonalds which do break new ground. What is unique about them is that the Commission is questioning the application of the transfer pricing rules by the revenue authorities in those individual countries. In brief, it is posited that the revenue authorities gave unduly favourable rulings to the multinationals in departure from the transfer pricing rules and that this constituted State aid.
Much has been written (including by myself here, here, here and here), and will continue to be written, about these investigations. In the case of Apple, the Irish government and Apple have published their grounds of appeal. Amongst other things, including a throwaway argument on the basis of the Charter of Fundamental Rights, the government and Apple argue that the Commission has misconceived how the law in relation to transfer pricing ought to be applied. Indeed, the Netherlands, Luxembourg and Fiat have appealed against the Commission’s decisions (the McDonalds and Amazon investigations are still ongoing) and they too will question the Commission’s understanding of the underlying law.
I think there is a prior question which needs to be asked however. Let’s assume then that the Commission is correct and that in each case, the revenue authorities did grant rulings which departed from the underlying rules. Should this constitute State aid and so engage the authority of the Commission? What I am driving at, and what I have flagged up elsewhere is the functional sustainability of such a conclusion. Is the logical conclusion not that all multinationals, when seeking rulings from revenue authorities, will thereafter seek to have the Commission approve the ruling? Is that sustainable across all 27 Member States?
Taking a step back, why do we even have revenue authorities?
The idea is that it is necessary to equip a body with the powers and resources to collect tax that is due. Now it is not possible for the revenue authority to ensure that it collects every single penny of tax that is due. Two important reasons for this can be highlighted. First, it is impossible to ensure that there is no evasion, fraud or avoidance. Instead, the relevant authority must make judgement calls as to how best to ensure compliance-whether that is targeted campaigns, engaging with representatives or unions of taxpayers, strategic litigation, threatening letters, whatever. Second, it is incorrect to say that there is a definite amount of tax that is due. For instance, many taxes are not imposed upon revenue streams themselves. They are for instance levied upon income or profits. In that sense, I do not pay tax on the money that I receive from selling a jumper. I pay tax on the profit (which is the sales price minus the costs). The calculation of cost, even in the simple case of an article of clothing like a jumper, is not particularly straightforward. What if I have wool left over for the subsequent tax year. How should I price that wool? What if I make that jumper in my own home-can I offset some of the cost of rent? Thus, whilst revenue streams may have definite values, incomes do not and thus neither do the taxes that are imposed. This is particularly the case when it comes to transfer pricing as we are already looking at something pretty unique and artificial. The arm’s length principle looks for a comparator. A comparator is not identical, it is comparable. The relevant revenue authority and the company must hammer out a price that is agreed to be appropriate in the circumstances. Thus again there is a requirement that the relevant revenue authority make a judgement call as to the appropriate amount.
In essence then, we equip a revenue authority to collect and manage taxes and in doing so endow the body with authority to make judgement calls as to how to go about the collection of tax. That is the nature of administrative discretion-the public body is given authority to decide between a range of choices and the courts respect the public body’s resulting decision. It is not right that the courts should micromanage the manner in which the public body carries out its task, as the public body is infinitely better placed to perform the particular function and nor would it even be practically possible for the courts to do so. In the case of tax, that means that there is some allowance given to decisions which are technically incorrect in law. Thus, in the famous Fleet Street Causals case, the House of Lords endorsed a decision of the Inland Revenue to collect less than the full amount of tax that was due. The tolerance is not unlimited of course. The standard that the courts apply in determining how far they should intrude into decisions made within the public body’s discretion is known as reasonableness or rationality. In short, it is a high standard-it requires the decision to be so incorrect that no reasonable official would arrive at it.
Bringing this back to the issue of rulings and assuming that the rulings themselves are indeed a departure from the underlying law, does that in itself mean that a supervising authority should condemn the rulings? Is there a case to be made that EU State aid law should make some allowance for incorrect applications of the law by the revenue authorities? The argument could be made that there should be some degree of deference afforded to revenue authorities to misapply the law, as is afforded by courts. This would be in line generally with the principles underpinning the EU concept of subsidiarity, namely, that decisions should be taken by those best placed to take them. This would also ensure that the Commission is not called upon to investigate all errors of law committed by revenue authorities. That is an important consequence that should also be considered. If the Commission’s case is upheld, this would mean that any time there is a misapplication of law by the revenue authorities in favour of multinationals, then that could potentially amount to State aid. So this would not just be transfer pricing arrangements, but also all settlements of tax disputes and any tax arrangement, binding or otherwise, between multinationals and revenue authorities.
Of course, that is not to say a carte blanche should arise and that there is no place for the Commission in this area. It is simply that Competition rules should not overreach unnecessarily upon the machinery of tax collection. Indeed, it has been found in other areas that Competition law must have tolerance for arrangements which may have a distortive effect on competition (such as in allowing for trade unions). The fundamental issue appears to be that there is a lack of trust that the revenue authorities of certain countries in the EU properly carry out their function of dispassionately collecting tax. There is merit to that concern. There is merit too to the idea that favourable treatment can distort competition. I have flagged already elsewhere that a route could lie through the application of Article 102 TFEU which proscribes abuse of a dominant position. Another hypothetical solution could lie in institutional reform. The EU has long been in the business of reforming national institutions and could set out guidelines for Member States for the institutional set-up of their revenue authorities so that observers can be assured that the revenue authorities are properly and dispassionately carrying out their functions. Of course, it would first be necessary to distil what is best practice in this area and that would be a mammoth research task (although I can think of one early career scholar who would fancy the task…)