Case Nos: CA-2024-000713 - [2025] EWCA Civ 1271

The cross-appeal, like the appeal, involves a challenge to a finding of fact based on an evaluative judgment. Therefore it faces the same difficulties, with one additional hurdle, namely, that FCE bear the burden of establishing that they would not have been advised that an ACT Challenge was worthwhile/they would not have had the confidence to embark on the preliminaries to litigation, had they sought advice from an Appropriate Adviser on 6 June 2000. The Judge recognised that it was for FCE to prove that, despite Verkooijen, and its accepted implications for a DV Challenge, an Appropriate Adviser would still have advised that an ACT Challenge would not be worthwhile (in the sense explained by the Supreme Court). He decided they could not discharge that burden. Indeed he found as a fact that a well-advised multi-national would have had sufficient confidence to embark upon the preliminaries in respect of both a DV Challenge and an ACT Challenge in the wake of the judgment in Verkooijen.

FCE’s case was that an Appropriate Adviser would not have appreciated that an ACT Challenge was worthwhile until, at the earliest, the CJEU’s judgment in Hoechst (and thus a claimant seeking advice from such an Adviser could not have discovered the mistake before then). The argument is based on the premise that the ACT Challenge is so fundamentally different to the DV Challenge that the publication of the CJEU’s judgment in Verkooijen would not have led to an Appropriate Adviser advising a prospective claimant that an ACT Challenge was worthwhile.

One of the fundamental problems for FCE is that even if they could establish that the Judge erred in failing to appreciate that there was a significant difference between the two types of challenge, which meant that Verkooijen would not have alerted an Appropriate Adviser to the viability of an ACT Challenge, the FCE claims (all of which are ACT Challenges) would be time-barred if the date of discoverability was the date on which the Advocate General’s Opinion in Hoechst was published (12 September 2000). In order to avoid that consequence, FCE would have to establish that the mistake of law was not discoverable by the exercise of due diligence even then.

Hoechst was concerned with the tax treatment of UK subsidiaries of non-resident parents. As the Judge set out at [96]-[98], the issue in Hoechst was whether a UK subsidiary of an EU resident parent company could make a group income election in respect of ACT when it paid dividends to that parent, as it could if it made the payment of dividends to a UK parent company. If permitted, this would enable the subsidiaries of non-resident parents to enjoy the same cashflow advantage as the subsidiaries of UK parents. As Mr Gammie explained in his report, and the Judge accepted, Hoechst raised no issue of double taxation relief. The argument did not require the UK to forego charging ACT, nor to confer any tax credit on the non-resident recipient of the dividend. The challenge in Hoechst was based squarely on the contention that the difference in treatment was contrary to the freedom of establishment.

Importantly, in the light of Avoir Fiscal and subsequent cases, it was not argued by the UK that there was an objective difference between subsidiaries of UK and EU parents which could justify a difference in treatment between them. However, arguments were mounted on both non-comparability of the two situations (because a foreign parent would not be liable to ACT) and the need to preserve cohesion in the UK tax system. It was argued that, although making a group income election relieved the UK resident subsidiary of the obligation to pay ACT when it paid the dividends to its UK parent, that payment was merely deferred, because a resident parent would pay the ACT when it made distributions subject to that tax. A non-resident parent would not pay ACT at all. If resident subsidiaries and non-resident parent companies were able to make a group election, then no ACT would be paid in the UK. A UK resident subsidiary of a resident parent company was therefore not in a comparable situation to a UK resident subsidiary of a non-resident parent.

Secondly, it was argued that in order to prevent double taxation in economic terms, resident shareholders are exempt from paying corporation tax on dividends they receive from resident subsidiaries, since that exemption is offset by the ACT charge on the payment of dividends by subsidiaries to their parents.

The Advocate General rejected both those arguments. At [25] he held that the differences in the respective fiscal situations of the parent companies did not provide a justification for denying the subsidiary of a non-UK parent the same tax advantage as the subsidiary of a UK parent. He then considered the cohesion argument and distinguished Bachmann, relying among other matters on the reasoning in Verkooijen and the absence of the requisite direct link. He concluded at [35] that the objectively different corporation tax positions of resident and non-resident parents cannot justify the imposition of an effectively higher corporation tax burden only on the subsidiaries of the latter.

The CJEU, in its judgment, agreed with the Advocate General’s analysis and his conclusions. It expressly rejected the non-comparability argument at [52] to [60], adopting and approving the Advocate General’s reasoning at [25] of his opinion. It then considered the cohesion argument and rejected it at [66] to [73], distinguishing Bachmann at [69] on the basis that there is no direct link between, on the one hand, the refusal to exempt subsidiaries in the UK of non-resident parent companies from payment of ACT under a group income election and, on the other, the fact that non-resident parent companies are not liable to corporation tax in the UK. It held at [70] that parent companies, whether resident or non-resident, are exempt from paying corporation tax in respect of dividends received from their UK subsidiaries and it is irrelevant for the purposes of granting a tax advantage (such as exemption from ACT under the group income election regime) that the reasons for that are different, namely, in the case of a non-UK parent because it has no liability to pay corporation tax, and in the case of a UK parent because the profits have already been taxed at source and there would otherwise be double taxation.

Mr Bremner submitted that there would be an insufficient level of confidence to embark on the preliminaries to litigation based on the Advocate General’s opinion, but there would be sufficient once there was a decision handed down by the CJEU. I am unable to accept that submission. In my judgment, if an Appropriate Adviser, when asked if there was enough of a legal basis to make a viable ACT Challenge, would have worked out from the CJEU judgment in Hoechst that there was, and would have given advice to the multi-national which would imbue them with that level of confidence, then that Adviser would have done the same on reading the Advocate General’s opinion which adopts the same reasoning. The mistake was discoverable when the Appropriate Adviser would realise that the CJEU might well accept the argument that the situations were comparable and the difference could not be justified on grounds of cogency. That is fatal to FCE’s case even if they were to succeed in undermining the Judge’s finding that the date of discoverability for an ACT Challenge was the date of the CJEU judgment in Verkooijen.

So far as that finding is concerned, I can see no legitimate basis for this court to disturb the Judge’s conclusion that the reasoning in Verkooijen was enough to dismantle the professional consensus, and was therefore enough to indicate to an Appropriate Adviser that there was now a real possibility that the CJEU might reject all the reasons why, in the past, it was believed that there were insurmountable obstacles to a claim based on an argument that matters of direct taxation which involved double tax relief and differential tax treatment between companies with non-UK and UK parents were contrary to one or more of the fundamental Treaty freedoms. The precise detail of the legal argument does not matter.

Whilst Verkooijen contained no argument concerning an obligation on the part of the Netherlands to give relief for foreign tax, or to abandon its taxing rights in relation to overseas income, that was the practical result of obliging the Netherlands to give the exemption in respect of dividends paid by the Belgian company as well as dividends paid by the Dutch company. It would not have been difficult for an Appropriate Adviser to reach the conclusion that if the CJEU was prepared in that case to find that the tax treatment of foreign dividends in the hands of a resident taxpayer was comparable with the tax treatment of dividends paid by resident companies to that taxpayer, there would be a viable argument (that would at least get past a strike-out application) that the refusal to treat EU dividend income as FII in the hands of a UK company was unjustified as a matter of EU law.

Of course, Verkooijen concerned the taxation of individuals, not companies. However commentators at the time (such as David Oliver KC) began to query whether it could be read across to taxation of corporations. It is clear that they did not discount that possibility, though Mr Oliver was both cautious and pessimistic in his conclusions.

Mr Bremner argued that although Verkooijen flagged up that a DV Challenge would get off the ground to a sufficient extent to give confidence to a claimant to embark on the preliminaries to litigation, a great deal more work had to be done before an Appropriate Adviser would reach the same conclusion about an ACT Challenge, even if they suspected it might be arguable. Suspicion would not be good enough to establish the level of confidence indicated by the Supreme Court. Verkooijen indicated that EU law may have something to say about a Member State’s system for dealing with double taxation relief, but that was just a stepping stone. The UK’s ACT system was (on the face of it, at least) even-handed, as there was a single ACT charge which applied both to UK source income and foreign source income. The ACT charge funded the shareholder credit, so there was an internal logic to the system. It was only at a much later stage that it became clear that as a matter of EU law, ACT was treated as a prepayment of corporation tax and could be equated to foreign corporation tax. That only emerged, he submitted, at [88] of the CJEU judgment in Hoechst.

In support of that argument, Mr Bremner referred to articles that were written by commentators in the 1990s, prior to the decision in Hoechst, in which they described ACT as being in concept an income tax, because although it looked like a company tax, so far as the shareholders were concerned the company had pre-paid income tax on their behalf. However, despite the fact that ACT provided an economic justification for the shareholder tax credit, from the perspective of the relevant taxpayer complaining of the discriminatory treatment (i.e. the UK subsidiary) ACT operated as an advance payment of corporation tax on those profits at the time when they were distributed. Mr Bremner accepted that this was true, but said that the Advocate General and the CJEU in Hoechst looked at that function to the exclusion of the others, which was something no-one had foreseen.

All these arguments were aired before the Judge, who regarded them as flawed. I agree with him that the CJEU’s treatment of ACT as an advance payment by the relevant taxpayer on account of corporation tax was hardly surprising, given that this is the way in which it is described in s.14 of ICTA, quoted by the Judge at [138]. As he said, Neuberger J had found that ACT was in the nature of corporation tax in his judgment of 2 October 1998 which settled the terms of the reference to the CJEU in the Hoechst case (a point also made by the Advocate General in Hoechst). As the Judge also said, the test as explained by the Supreme Court did not require the claimant to foresee in advance all the ways in which the court might ultimately determine the challenge in their favour. They would just need to have sufficient confidence to embark on the preliminaries. He found as a fact that they would have done, once it became clear that the factors previously seen as insuperable obstacles were no longer incapable of being overcome. That finding cannot be disturbed, on the application of classic Volpiv Volpi principles.

As I have already mentioned, there was evidence that shortly after the judgment in Verkooijen was published, BAT started examining internally whether an ACT Challenge could be brought. The Judge was therefore entitled to make the finding at [140] that the inconsistency between the judgment in Verkooijen and a substantial proportion of the professional consensus described by Mr Gammie would have given a well-advised multi-national sufficient confidence to embark upon the preliminaries to litigation. Bearing in mind where the burden of proof lay, his conclusion that he saw no other secure basis in the evidence for concluding that the date of discoverability in respect of an ACT Challenge was any later than 6 June 2000 was fully justified.

For those reasons, I would also dismiss the cross-appeal.