The issue of whether certain large companies have been able to get out of paying tax that they should legally (or morally) have paid could hardly be more politically sensitive at the moment.
The Commission’s decision – published on DG Comp’s website on 30 September C(2014) 3606 final, case SA.38373 – to initiate a full State aid investigation of Ireland’s tax treatment of Apple represents a significant intervention by the Commission into this highly-charged issue. It also exposes Apple to a risk of having to repay very large amounts of money to the Irish government. (Some press reports referred to a “fine” on Apple but that is in fact not a risk in State aid cases – though the difference between a fine and repayment may not be of much comfort to Apple).
Ireland’s ability to attract large corporations by means of a very favourable corporation tax regime is well-known (and controversial in those countries that feel that they have lost out as a result). But it is well-established that the State aid rules have nothing to say about a decision by a Member State to set a low general tax rate. Put shortly, the State aid rules deal with differences of tax treatment within States, not between them. More precisely, when you look at whether the tax treatment of an undertaking involves State aid, you have to compare that treatment with the treatment of other undertakings in that State that are in a comparable legal and factual situation: you do not compare it with the tax treatment of comparable undertakings in other States.
The case for looking at Apple’s tax treatment in Ireland under the State aid rules therefore turns on the proposition that Apple has received favourable treatment compared with other undertakings subject to the same Irish tax rules.
What has given rise to the Commission’s concern in this case is two tax rulings by the Irish tax authorities relating to transfer pricing arrangements. As is well known, one of the difficulties that arises in the tax treatment of multinational corporations such as Apple is how to deal with intra-group sales. The risk is that the corporation will seek to overstate the intra-group price charged by an entity in a jurisdiction with a low corporation tax to an entity located in a high tax jurisdiction, the effect being to increase the profit of the lower-taxed entity and to decrease the profit of the higher-taxed entity. Tax authorities will therefore seek to ensure that, for tax purposes, the price at which intra-group sales are made is the arm’s length price. As can be imagined, that objective is rather easier to state than to achieve, particularly where there is no obvious open market equivalent to the sale being made..
The rulings in the present case are both advance pricing arrangements (“APAs”). The first was reached in 1991, the second in 2007. As the Commission explains, APAs are arrangements that determine, in advance of intra-group transactions, an appropriate set of criteria (e.g. method, comparables and appropriate adjustments thereto, critical assumptions as to future events) for the determination of the transfer pricing for those transactions over a fixed period of time. It is notable that the Commission is at pains to distinguish between APAs and “private rulings”, which the Commission defines as rulings by tax authorities as to the correct tax treatment of a prospective transaction notified to them. The Commission notes that private tax rulings deal only with legal issues (the facts being assumed – and the ruling typically only holds if the assumed facts are the real facts) and that they will only apply to the transaction notified). The point here appears to be that APAs have a certain vulnerability to favourising individual undertakings in that they involve factual assessments by the tax authorities (so that there is a risk that the authorities will find convenient facts in cases where favourable treatment is being aimed at) and will then apply ex ante to a large number of transactions and over a number of years. The assumption seems to be that private rulings are less risky from a State aid perspective because it is easier to ascertain whether tax authorities are maintaining a consistent approach on purely legal issues (and there will be both internal and external pressure on them to have a consistent legal approach).
For present purposes, it is unnecessary to go into the precise details of the APAs at issue. What is of general interest to those concerned with the interface between tax and State aid is the broad reasons why the Commission considers that State aid may have arisen in this case. These appear to be the following: –
- First, the Commission has obtained documents recording discussions between the tax authorities and Apple that appear to indicate that certain bases of assessment were set so as “not to prohibit the expansion of [Apple’s] Irish operations” (a statement highlighted by the Commission); the documents also record references to Apple’s being the largest employer in the Cork area.
- Second, the documents do not appear to explain how critical figures were determined. So the Commission is concerned that they were an unprincipled compromise rather than a good faith attempt to implement the arm’s length principle: in the Commission’s words, the “methods used to determine profit allocation to [the relevant Apple subsidiaries] result from a negotiation rather than a pricing methodology”.
- Third, there were indications in the documents that the methodology chosen was “reverse engineered” so as to generate a desired level of taxable profits.
- Fourth, there were unexplained and substantial anomalies in the methodology used.
- Finally, the Commission expresses concern that the first APA lasted unchanged for 15 years, during which there were likely to have been significant changes to the economic environment. That period is contrasted with the usual length of such arrangements elsewhere in the EU: and the Commission also noted that the 2007 APA departed significantly from the 1991 APA (which suggests that by the end of the 15-year life of the 1991 APA quite a lot had materially changed).
Although it should be remembered that none of this is more than an opening statement by the Commission (it is simply a decision to open a full-scale investigation), a number of lessons need to be drawn from this by both tax advisers and tax authorities. First, it is important to make sure that there is a full and convincing paper trail justifying tax treatment decisions of this kind. Second, care has to be taken in negotiation to ensure that the tax treatment that is reached does not just look like a “split the difference” result but has a rational justification. Third, it is very unwise to refer to wider economic considerations (jobs, investment decisions) in the context of discussions of appropriate tax treatment. Finally, the arrangement reached has to be kept under review and looked at again as and when facts change on the ground.
The importance for the companies concerned of avoiding State aid risk can hardly be over-estimated. The effect of a finding of State aid in these cases will inevitably be an order, binding on the State concerned and its courts, to obtain repayment of the value of the advantage (subject only to a limitation period of 10 years). In addition, the taxpayer will have to pay substantial and compounded interest on those sums. Given the Commission’s evident interest in these matters, tax advisers of large companies entering into these types of discussions with tax authorities need to have the State aid risk (as well as the risk of a nasty encounter with the House of Commons Public Accounts Committee) right at the front of their minds.
Monckton Chambers, London