A conclusion: the disconnection between freedoms of movement and tax jurisdiction shopping;
and a question: what remains of the Marks & Spencer judgment?
On 25 February this year, the European Court of Justice gave judgment in the X–Holding case (C-337/08). Following the Advocate-General's opinion, it held that in permitting a resident parent company to form a single tax entity with a resident subsidiary (being controlled to the extent of at least 95%), while preventing that same parent company to form such a tax entity with a non-resident subsidiary, Dutch legislation was not only compatible with EU law, but was entirely proportionate.
Thus there is no question of a non-resident subsidiary forming a single taxpaying entity with its resident parent company, or of transferring its losses to that company to take advantage of the more favourable tax system in force in the parent company's state.
What lessons can be drawn from this judgment, and what are its implications for the Marks & Spencer judgment (C-446/03, 14 December 2005)?
The X–Holding judgment is clear: while Dutch law is applicable in all circumstances, that is, whether or not there is any risk of tax evasion, fraud or double benefit from a tax advantage, and while its sole objective is to protect the revenues of the Dutch treasury, the Court considers that the differential treatment it establishes, to the detriment of non-resident subsidiaries, is justified solely by the need to preserve a "balanced allocation of the power to tax between Member States". What is the meaning of this new justification?
In reading the passages of the Marks & Spencer judgment where this ground first appeared in 2005, one might understand it to be applicable where companies have "the option to have their losses taken into account in the Member State in which they are established or in another Member State …" (para. 46).
Hence it is where the taxpaying company can choose the national tax system in which it will be taxed that the balanced allocation of the power to tax might be compromised. The necessity of this element of choice of national tax system was confirmed in the Lidl Belgium judgment (C-414/06,  ECR I-3601) which stipulates in para. 52 that "...to give the principal company the right to elect to have the losses of that permanent establishment taken into account in the Member State in which it has its seat or in another Member State would seriously undermine a balanced allocation of the power to impose taxes between the Member States concerned".
In the X–Holding judgment, the decisive element is again the choice between two national tax systems: "Since the parent company is at liberty to decide to form a tax entity with its subsidiary and, with equal liberty, to dissolve such an entity from one year to the next, the possibility of including a non-resident subsidiary in the single tax entity would be tantamount to granting the parent company the freedom to choose the tax scheme applicable to the losses of that subsidiary and the place where those losses are taken into account" (para. 31).
The Court's examination of the proportionality of the Dutch measure is startling, in that it gave no consideration to the fact that the Netherlands could have limited the taxpayer's choice by imposing, for example, a minimum duration for unitary taxation of the entity including the foreign subsidiary (not forgetting that while the single tax entity means that the subsidiary's losses must be taken into account in the Netherlands, it also enables its profits to be taxed there). In the same way, it gave no consideration to the fact that the Netherlands could have made the legislation which applies to non-resident permanent establishments (which have the benefit of a system of provisional transfers of losses, combined with a system for recuperation in subsequent years of profit) equally applicable to non-resident subsidiaries.
It is surprising to see the Court revisiting the question of whether the positions of non-resident subsidiaries and non-resident permanent establishments are comparable, for the purposes of justifying a departure from the fundamental freedoms, and equally surprising to see it coming to a negative conclusion. In fact, in both cases, the subsidiary or permanent establishment is taxed in the same state. Further, it is difficult to make out the specific tax implications of the Court's argument that the permanent establishment "remains in principle and in part subject to the fiscal jurisdiction of the Member State of origin" (para. 38).
This is especially so in light of CLT-UFA (C-253/03  ECR I-1831), where the Court held that German legislation under which a Luxembourg company had profits made in Germany more heavily taxed where they were made by a permanent establishment than where they were made by a subsidiary was inconsistent with the Treaty. In para. 40 of the X–Holding judgment, the Court simply stated, without very much by way of explanation, that the subsidiary and the permanent establishment were not "in a comparable situation with regard to the allocation of the power of taxation".
It can be added that the objective of the provisions is the same whether they relate to non-resident subsidiaries or to non-resident permanent establishments. This seemed obvious before 2003, when Dutch legislation did not exclude non-resident subsidiaries.
The Marks & Spencer case involved a challenge to UK legislation which permitted a resident subsidiary to transfer its losses to the UK parent company, while prohibiting non-resident subsidiaries from doing so. The judgment was novel in that it held that such differential treatment could be justified by three important considerations of general interest "taken together". One of these, the need to combat tax evasion and fraud, had already been recognised, although it was defined in wider terms than in previous judgments. The others had not been seen before: preventing double use of losses, and protecting the balanced allocation of the power to tax between Member States.
In the Marks & Spencer case itself, on the issue of proportionality, the Court nevertheless held that UK law went too far in applying even where the taxpayer had no choice as to the national tax system in which the losses of its non-resident subsidiaries were to be declared. In the case of Marks & Spencer, these had ceased their activities in their respective countries of establishment and, the losses having become permanently unusable in those countries, they could only have been used in the UK. Such losses are generally described as "terminal".
There are three factors leading to concerns that even such "terminal" losses made by a non-resident subsidiary may no longer be transferred to the parent company, even though this would be possible for a resident subsidiary:
The fact remains that this situation was not addressed in the X–Holding case, and that it is still reasonable to think that in the case of terminal losses (and on the conditions set out in para. 55 of the Marks & Spencer judgment) the taxpayer does not have, or certainly no longer has, the choice of different national tax systems in which it could seek to use the relevant losses. This is so even though there is plainly no risk of the tax advantage being used twice.
This was clearly the case in Marks & Spencer, where the losses were permanently unusable in the Member States where its subsidiaries were established. It was also the case in Krankenheim Ruhesitz (C-157/07  I-8061), where losses made by an Austrian permanent establishment of a German company were unusable in Austria because Austrian law did not provide, either for resident companies or for non-residents, any mechanism for transferring such losses.
The Marks & Spencer judgment may therefore survive insofar as it concerns losses which are permanently unusable in the state where the non-resident subsidiary is established, or in the state of the permanent establishment, but on reading the recent judgments of the Court one might be tempted to wonder… for how long?