Between 1973 and 1999 a company resident in the United Kingdom which paid a dividend was liable to pay advanced corporation tax ("ACT"). This payment could be set off against its liability for corporation tax on its profits ("mainstream corporation tax" or "MCT"). The recipient of the dividend was entitled to a tax credit for the ACT. If the recipient was a company, the dividend and the tax credit constituted franked investment income and could be distributed to its own shareholders without further payment of ACT.
The economic purpose of this system was to ensure that a company's profits were not taxed twice: first as profits earned by the company and then again as dividends received by the shareholders. The payment of tax by the company was partially imputed by means of a tax credit to the shareholders. But the system also had to ensure that credit was given only for tax which had actually been paid. Hence the requirement that ACT be paid at the time of the distribution and then set off against MCT.
Section 247 of the Income and Corporation Taxes Act 1988, like many other of its provisions, recognised the unity of a group of companies which are in law separate persons but economically a single enterprise. It provided that a parent and subsidiary, both resident in the United Kingdom, could jointly elect that the subsidiary would pay dividends free of ACT and the parent would receive them without the benefit of a tax credit. The dividends would not be franked investment income and the parent would be liable for ACT when it passed them on as dividends to its own shareholders. The advantage to the group was that money could be moved from subsidiary to parent, either with a view to subsequent distribution or for some other purpose, without attracting an immediate liability to ACT.
In Metallgesellschaft Ltd v Inland Revenue Comrs (Joined Cases C-397 and 410/98)  Ch 620 ("the Hoechst decision") the European Court of Justice decided that the denial of the advantage of a similar right of election to a group with a parent company resident in another member state was a restriction on the latter's freedom of establishment, contrary to article 43 of the EC Treaty. Such groups were entitled to compensation from the United Kingdom for having had pay ACT which could have been avoided by an election.
The question in this appeal is whether a similar claim to compensation can be made by groups with parent companies resident in countries outside the European Community. Such groups obviously do not enjoy freedom of establishment under article 43 but they put their claim in two ways.
First, the United Kingdom has entered into double taxation conventions ("DTCs") with many foreign countries. These are based on a model drafted by the Organisation for Economic Cooperation and Development ("OECD") and contain a standard article prohibiting various forms of discrimination by the tax laws of the one contracting state against nationals or residents of the other. The denial of a right of election on the ground that the parent company is not resident in the United Kingdom is alleged to constitute such discrimination.
Secondly, the foreign groups claim that section 247 infringes article 56 of the EC Treaty, which prohibits restrictions on the movement of capital and restrictions on payments, not only between member states but also between member states and third countries.
This appeal arises out of test cases brought by groups of companies with parent companies resident in the United States and Japan. Both countries have entered into a DTC with the United Kingdom. The current Japanese DTC dates from 1970 (see the Schedule to the Double Taxation Relief (Taxes on Income) (Japan) Order 1970 (SI 1970 No 1948), that is to say, from before the introduction of ACT. The US DTC was redrafted after the introduction of ACT with that tax very much in mind: see the Schedule to the Double Taxation Relief (Taxes on Income) (The United States of America) Order 1980 (SI 1980 No 568). But the relevant non-discrimination article is in the same terms in both. In the US DTC, it is article 24(5) and in the Japanese DTC it is article 25(3):
Enterprises of a contracting state, the capital of which is wholly or partly owned or controlled, directly or indirectly, by one or more residents of the other contracting state, shall not be subjected in the first-mentioned contracting state to any taxation or any requirement connected therewith which is other or more burdensome than the taxation and connected requirements to which other similar enterprises of that first-mentioned state are or may be subjected.
Both Park J  STC 489 and the Court of Appeal  STC 606 held that section 247 infringed this provision because it subjected an enterprise in the UK, namely the subsidiary of a US or Japanese parent, to taxation (namely, payment of ACT) which a similar enterprise, namely a subsidiary of a UK parent, would (if they had made an election) not have had to pay.
The next question was whether this infringement of the DTC gave rise to any (and if so, what) remedy in English law. An international treaty does not give rise to any rights in English domestic law unless incorporated by legislation. The EC Treaty is so incorporated, in its entirety, by the European Communities Act 1972. But with DTCs the position is more complicated. Section 788 of the 1988 Act provides that Her Majesty may by Order in Council declare that arrangements made by a DTC shall "have effect". But the result is not to make the whole DTC part of English law. By section 788(3), the arrangements have effect
notwithstanding anything in any enactment:
in relation to income tax and corporation tax in so far as they provide–
Again, both Park J and the Court of Appeal agreed that, except for the implied reference to ACT in the provisions about tax credits in paragraph (d) (which was added when ACT came into force), the section did not give effect in domestic law to any arrangements which a DTC might make about liability to ACT. "Corporation tax in respect of income or chargeable gains" in paragraph (a) meant mainstream corporation tax, not ACT. The infringement of the DTCs therefore gave rise to no cause of action in English law.
As for the claim under article 56 of the EC Treaty, Park J was willing to accept that discrimination against a subsidiary of a company resident in a foreign state could be a restriction on the movement of capital or the making of payments between that state and the United Kingdom. But article 57(1) provides that:
The provisions of article 56 shall be without prejudice to the application to third countries of any restrictions which exist on 31 December 1993 under national or Community law adopted in respect of the movement of capital to or from third countries ....
The claimants argued that while article 57(1) might prevent them from claiming that section 247 fell within the prohibition of restrictions on the movement of capital, it did not exempt it from the prohibition on restrictions on payments. Park J thought it was obvious that article 57(1) disapplied the whole of article 56. The Court of Appeal regarded the point as doubtful but left to your Lordships' House the question of whether a reference to the Court of Justice should be made.
The result was that the claims failed. Both Park J and the Court of Appeal went on to discuss what remedies the claimants might have been entitled to if they had succeeded and whether the claims had been properly pleaded but these questions did not need to be decided.
The claimants appeal against the ruling that the DTCs were not incorporated into English law and seek a reference on a question said to arise under article 56. The Revenue support the decision of the Court of Appeal on incorporation and the view of Park J on article 56. But their chief argument is that both the judge and the Court of Appeal were wrong in holding that the denial of a right of election to groups with foreign parents infringed the non-discrimination articles of the DTCs.
The reasoning of the judge and the Court of Appeal was that article 24(5) of the US DTC (for example) requires one to compare the positions of the UK-resident subsidiary of a US parent and the UK-resident subsidiary of a UK parent. If the latter can elect under section 247 and the former cannot, that is discrimination contrary to article 24(5). It is irrelevant that an election would transfer liability for ACT to the UK parent but not to the US parent. The DTC is concerned with the taxation of enterprises in the UK and not with the tax position of their foreign-resident shareholders. The authoritative commentary on the equivalent article of the OECD Model convention with respect to taxes on income and on capital 1963 says that its object is:
to ensure equal treatment for taxpayers residing in the same state, and not to subject foreign capital, in the hands of the partners or shareholders, to identical treatment to that applied to domestic capital.
I respectfully think that this particular observation does not take the matter much further forward because it is directed to a different point. It draws attention to the limited application of the non-discrimination article, which provides only for treatment of resident tax payers and does not prevent a state from discriminating in its treatment of the income of foreign shareholders; for example, by imposing a withholding tax. It does not say that parentage cannot be a relevant characteristic of a resident tax payer.
The question, as it seems to me, is whether section 247 discriminates against a UK company on the ground that its capital is "wholly or partly owned or controlled, directly or indirectly" by residents of the US, or Japan, or some other foreign state. In relation to article 24(1) of the OECD model convention, which prohibits discrimination between residents on grounds of nationality, the commentary says that the "underlying question" is whether two residents are being treated differently "solely by reason of having a different nationality". It does not repeat this observation in relation to article 24(5), but the principle must be the same. Does section 247 discriminate on the grounds that the capital of the subsidiary is controlled by a non-resident company?
In my opinion it plainly does not. For example, if a US parent were to interpose a UK resident holding company between itself and its UK-resident subsidiary, the control would remain in the US but there would be no objection to an election by the UK subsidiary and its immediate, UK-resident parent. On the other hand, an individual US shareholder and the company he controls in the UK could not elect, but the reason is not because the company is subject to US control. An individual UK shareholder and his company could not elect either, for the same reason that a non-resident company cannot elect. It is because an individual is not liable to corporation tax. An election is a joint decision by two entities paying and receiving dividends that one rather than the other will be liable for ACT. This is not a concept which can meaningfully be applied when one of the entities is not liable for ACT at all.
Unfortunately the judge and the Court of Appeal did not have the benefit of the discussion of the nature of the section 247 election in the speeches in this House in Pirelli Cable Holding NV v Inland Revenue Comrs  1 WLR 400. The point was luminously made by Lord Nicholls of Birkenhead, at para 19, in a speech with which the rest of their Lordships agreed:
A group income election is a group election. A group income election cannot be made by a subsidiary alone. It is an election made jointly by the subsidiary paying the dividend and the parent receiving the dividend. By making such an election both companies seek the fiscal consequences of making the election. One consequence is that by making the election the subsidiary will obtain the advantage of not paying ACT in respect of the relevant dividend. Another consequence is that the subsidiary will obtain this advantage at the cost of depriving the parent of a tax credit in respect of the dividend. These two fiscal consequences are inextricably linked. You cannot have one without the other. That is why the election has to be made jointly. The advantage to the paying subsidiary comes at a price to the recipient parent.
In my respectful opinion, it is not possible to decouple the positions of parent and subsidiary as the judge and the Court of Appeal sought to do. To allow an election by a group with a US-resident parent would not be to give a relief available to a group with a UK-resident parent. It would be something different in kind. It would not be an election as to who would be liable for ACT but as to whether the group should pay it at all.
These arguments were in substance those put forward by the UK government to the Court of Justice in the Hoechst decision. But they were rejected. Why? Because the prohibition on discrimination implied in article 43 of the EC Treaty has an altogether different purpose from the prohibition on discrimination in the DTCs. Freedom of establishment under article 43 is the freedom of the resident of a member state to establish itself in another member state. In the case of parent and subsidiary, it is the freedom of the parent to establish a subsidiary. As Lord Scott of Foscote explained in the Pirelli case  1 WLR 400, para 77:
[T]he right to freedom of establishment conferred by article  is the right of a company (or an individual) with its seat in one member state to establish itself in another member state. Unwarranted restrictions imposed by the latter member state on the branch or agency or subsidiary company by means of which the parent company is seeking to establish itself in that member state is plainly a breach of the .... right to freedom of establishment of that parent company.
Discrimination against the group as a whole is thus a restriction on the parent's freedom of establishment. If a group with a UK parent has a cash flow advantage which a group with a parent in another member state does not enjoy, that is a restriction on the latter's freedom of establishment. Fortunately, by the time of the decision in Hoechst, ACT had been abolished. So no one had to decide how the decision should be given effect for the future. But it obviously would not have been possible simply to give a group with a parent in another member state the right to elect under section 247. That would have enabled it to elect not to pay ACT and put it in a better position than a group with a UK parent. It would have been necessary either to repeal section 247 or abolish ACT.
A DTC, on the other hand, does not give a company or individual resident in one country a right of establishment in the other. As the commentary on the OECD model says, the equality it ensures is only that any enterprise it owns in the other country will not be subject to taxation which discriminates on the ground of its foreign control. In my opinion, the denial of the right of election was not on the ground of the company's foreign control but on the ground that section 247 cannot be applied to a case in which the parent company is not liable to ACT.
It follows that in my view the lower courts were wrong in holding that section 247 imposed discrimination on subsidiaries of US and Japanese parent companies, contrary to the respective DTCs. As for the incorporation issue, I think it would be artificial to decide whether section 788 incorporated a provision of the DTC which did not actually exist and I therefore propose to express no opinion.
The claim under article 56 has been overtaken by decisions of the Court of Justice. In its judgment in Test Claimants in FII Group Litigation v Comrs of Inland Revenue (Case C-446/04) (unreported), 12 December 2006, as the claimants concede, the Court of Justice decided that Park J had been right in construing the exclusionary force of article 57(1) as applying to restrictions both on capital movements and payments. This left the appellants only with an argument that a statutory provision which was contrary to an incorporated or even unincorporated provision of a DTC was not a restriction "under national .... law" within the meaning of article 57(1). However, since in my opinion there was no infringement of the DTC, the premise for this submission disappears.
For good measure, the Court of Justice has decided in Test Claimants in the Thin Cap Group Litigation v Comrs of Inland Revenue (Case C-524/04) (unreported) 13 March 2007 that in the case of legislation which is "targeted" only at relations within a group of companies in which the parent had "a definite influence" on the subsidiary, article 56 had no application. The question was solely whether the legislation affected freedom of establishment under article 43 and a company which was not resident in a member state and therefore had no freedom of establishment could not rely on article 56. The appellants questioned whether this doctrine applied to section 247, which they said could, at any rate in theory, deem a company to be a subsidiary on the basis of a shareholding which did not give the holder a "definite influence". Your Lordships were invited to refer to the Court of Justice the question of whether, to exclude article 56, the legislation in question must necessarily apply only to a company which would (if it was resident in a member state) enjoy a right of freedom of establishment or whether it was enough that such a right would have existed on the facts of the particular case. The Revenue submitted that the latter was clearly the correct answer and relied upon the decision of the Court of Justice in Baars v Inspecteur der Belastingen Particulieren/Ondernemingen Gorinchem (Case C-251/98)  ECR I-2787. I am inclined to think that the Revenue are right, but in view of the exclusion of article 56 by article 57, it is unnecessary to express a definite opinion.
In these circumstances, it is unnecessary to express any view on the remedies to which the appellants would have been entitled if they had succeeded. I would dismiss the appeal.
I have had the advantage of reading the opinion of my noble and learned friend Lord Hoffmann in draft and for the reasons he sets out in that opinion I would dismiss this appeal. This means it is not necessary before dismissing this appeal for their Lordships to deal with all the issues that were considered by the Court of Appeal. One of those issues was a pleading and limitation point involving the Limitation Act 1980.
Normally, in the case of issues involving practice and procedure the House defers to the views of the Court Appeal. This is because the members of that court have had more recent experience of those issues than members of this House. However, in the case of this appeal I consider that it is desirable to draw attention to one aspect of the Court of Appeal's decision. This is as to the pleading and limitation point, even though to do so could be said to constitute a departure from the House's normal approach. It is important to do so because these proceedings were brought under a Group Litigation Order or "GLO" as a test case under Part 19 of the Civil Procedure Rules (the "CPR") and GLOs are a relatively recent significant innovation whose distinct procedure has so far received little detailed attention from appellate courts. However, if the experience in the United States of class actions, the United States equivalent of GLOs, is anything to go by, this is likely to change.
Before the trial judge the appellants sought to amend their statements of case to clarify the basis on which they were seeking a remedy. I am far from satisfied it was necessary for them to do this. But despite this Park J gave them the necessary leave to amend and, if amendments were desirable, then I would have thought Park J had a discretion to give the leave which he did. However, his decision was overruled by the Court of Appeal. The Court of Appeal did so based upon the provisions of the 1980 Act, dealt with in Part 17.4 of the CPR. That Part, having set out the general powers of a court to allow amendments, places a restriction on those powers if a limitation period has expired. Then the powers can only be exercised so as to add a new claim "if the new claim arises out of the same facts or substantially the same facts as a claim in respect of which the party applying for permission has already claimed a remedy in the proceedings".
In view of the outcome of this appeal in accordance with the opinion of Lord Hoffmann, the decision of the Court of Appeal as to the amendment is of no practical significance to the parties. However, my concern is as to its possible effect on future practice in relation to GLOs. GLOs can involve hundreds or thousands of different parties. In such a situation any step which each of the many parties has to take can cumulatively so effect the total costs, as to make them disproportionate both to the means of the parties to the action and the issues at stake. For this reason it is important that such steps generate the least possible costs.
This may not appear to be such a problem in this case, as the actions are being brought against a government department over very substantial sums of money. However, where this is avoidable, having one rule for one set of litigants who have ample means and a different rule for litigants of lesser means, cannot be justified. All litigants are entitled to be protected from incurring unnecessary costs. This is the objective of the GLO regime. Primarily, it seeks to achieve its objective, so far as this is possible, by reducing the number of steps litigants, who have a common interest, have to take individually to establish their rights and instead enables them to be taken collectively as part of a GLO Group. This means that irrespective of the number of individuals in the group each procedural step in the actions need only be taken once. This is of benefit not only to members of the group, but also those against whom proceedings are brought. In a system such as ours based on cost shifting this is of benefit to all parties to the proceedings.
Before a GLO can be made it is necessary for each individual potential member who wishes to join the GLO to make an individual claim under CPR Part 7 or Part 8. This in conjunction with the application to register enables the court to determine whether the respective litigants qualify to be a member of the GLO. It also prevents time continuing to run for purposes of limitation of actions. None the less the claim once made will usually almost immediately be of only limited historic interest because what matters is the application to register and the register of the GLO on which all proceedings subject to the GLO are registered. The purpose of a GLO is then "to provide for the case management of claims which give rise to common or related issues of fact or law (the GLO issues") (CPR Part 19.10). Section III of Part 19 contains additional case management powers for GLOs which include directing that there shall be a group particulars of claim and specifying the details to be included in a statement of case (Part 19.13 (d)). Directions may also provide "for one or more claims on the group register to proceed as test claims" as happened in the cases the subject of these proceedings (Part 19.13(b)). Where judgment is given on an issue on the group register in relation to a GLO, that judgment or order is binding on the parties to all other claims that are on the group register at the time the judgment is given, unless the court orders otherwise. (Part19.12 (1) (a)).
In considering the leave to amend issue the Court of Appeal do not appear to have had submissions addressed to them as to the significance of the proceedings being the subject of a GLO. No doubt it was as a result of this, that they failed to consider the question of whether an amendment was necessary and if so whether permission could be given in the context of the GLO. In the context of a GLO, a claim form need be no more than the simplest of documents. It needs to be read together with the application to register and the register bearing in mind its place in the GLO process and the need to limit pre registration costs so far as this is possible. In this case the suggested deficiency in the claim forms are that they did not sufficiently identify the basis for the revenue being under an obligation to repay the tax paid assuming this should not have been claimed by the revenue. This is an area of the law the parameters of which are still evolving. In my judgment it would be wholly inconsistent with the objective of the GLO to require the nature of the remedy claimed to be spelt out in detail in the claim forms of the taxpayers. The Revenue knew perfectly well the basis of the claims once the issues had been defined for the purpose of the GLO. For each of the parties to have to spell out details of the manner in which they would advance their claim at the outset would have caused substantial extra costs to be incurred in researching the law. Cumulatively this would have been grossly wasteful. The decision of the Court Appeal should not be treated as requiring a claim to set out more than an outline of the claim. Furthermore for limitation purposes in the case of a GLO the individual claims should be construed in conjunction with the applications for the claims to be registered and, from the time of registration, the register.
Lord Walker of Gestingthorpe
I am in full agreement with the opinion of my noble and learned friend Lord Hoffmann, which I have had the privilege of reading in draft. For the reasons given by Lord Hoffmann I too would dismiss this appeal.
I have had the benefit of reading in draft the opinion prepared by my noble and learned friend, Lord Hoffmann, with which I find myself in full agreement on the first issue, whether section 247 of the 1988 Act imposed discrimination on the appellant subsidiaries of US and Japanese companies, contrary to the respective Double Taxation Conventions. I therefore agree that this appeal should be dismissed.
I have also had the benefit of reading in draft the opinion of my noble and learned friend, Lord Neuberger of Abbotsbury, and I agree with what he says on the second issue: had the first issue been answered as the court below answered it (viz. that there was discrimination), it would, in my view and as the courts below held, have been discrimination in respect of which the appellants had no domestic law remedy because, properly construed, the words "corporation tax in respect of income or chargeable gains" in section 788 (3)(a) of the 1988 Act do not cover advance corporation tax.
Lord Neuberger of Abbotsbury
I have had the great privilege of reading in draft the opinion of my noble and learned friend, Lord Hoffmann, and I agree with his views. Accordingly, I consider that this appeal should be dismissed on the first issue that has been argued, namely that identified in paragraphs 6 and 7 above. I agree that the denial of the right of election under section 247 of the Income and Corporation Taxes Act 1988 ("the 1988 Act") to groups with foreign parents did not infringe the non-discrimination articles of the Double Taxation Convention ("DTC") between the United Kingdom and Japan or of the DTC between the UK and the United States.
I would also dismiss the appeal on the second issue that has been argued, namely that identified in paragraph 8 of Lord Hoffmann's opinion. I consider that the provisions of section 788 of the 1988 Act would not have given effect to the non-discrimination articles, even if (as the Court of Appeal and Park J thought) the UK was required by those articles to give a UK subsidiary with a Japanese (or US) parent the right of election under section 247. Although this second issue does not necessarily arise in the light of our conclusion on the first issue, I propose to consider it, as it involves a fairly short point of statutory construction, which was the primary basis upon which both Park J and the Court of Appeal would have dismissed the present claims, and it has been fully argued.
The resolution of this second issue turns on whether ACT is within the expression "corporation tax in respect of income or chargeable gains" in section 788(3)(a), quoted in paragraph 8 above. There seems to me to be no doubt but that ACT is "corporation tax". That is clear not only from its name, but also (and primarily) from section 14(1) of the 1988 Act, which appears to be a statutory definition, in that it identifies the ACT which falls to be paid pursuant to the section as "an amount of corporation tax". The issue therefore turns on whether ACT is corporation tax "in respect of income or chargeable gains". The argument on behalf of the Revenue is that, unlike so-called mainstream corporation tax, which is payable on income and chargeable gains, ACT is payable on the making of a distribution.
Mr Aaronson QC for the claimants placed great reliance on section 6 of the 1988 Act, which is the first section of the Act which mentions corporation tax, and, indeed, is the first section under the heading "Corporation tax". Section 6(1) provides that "corporation tax shall be charged on profits of companies", and "profits" is defined in section 6(4)(a) as meaning "income and chargeable gains". However, that only assists the claimants if section 6(1) is, in effect, an exclusive definition of "corporation tax". But it does not appear to me to be a definition at all: it is simply a charging provision. The fact that corporation tax is chargeable on the income and chargeable gains of a company does not mean that corporation tax is not chargeable on anything else.
Section 6, as already mentioned, and indeed section 8, of the 1988 Act provide for corporation tax (i.e. "mainstream corporation tax") to be charged on a company's profits, and say so in terms – in sections 6(1) and 8(1), (2) and (3). Section 14(1) of the 1988 Act, which introduces ACT, makes no such provision. Indeed, section 14(4) states that the "provisions" of the 1988 Act "as to the charge, calculation and payment of corporation tax" should "not be construed as affecting the charge, calculation or payment of" ACT. Thus, nothing in section 14(1) or (4) suggests that ACT payable by a company is payable in respect of, or charged on, the profits of the company.
Sections 14(2) and (3) do not seem to me to call that conclusion into question. Section 14(2) is not of relevance for present purposes. As for section 14(3), both sides relied on the fact that it refers to ACT being "payable on an amount equal to the amount or value of the distribution". Mr Aaronson suggested that the apparently rather cumbersome dual reference to "amount" supports the notion that ACT is being treated as payable on the profits of the relevant company. I do not see how that follows as a matter of logic. I think that Mr Glick QC, for the Revenue, made a better point when he said that section 14(3) is expressed in this way to make it clear that ACT is payable on an amount calculated by reference to the value of the distribution and not on the distribution itself, and that therefore, even where the distribution is made out of profits, any ACT due would not be payable on the distributed profits of the company.
Accordingly, section 14, which is an important section in the present context as it introduces ACT, appears to me to indicate that (unlike mainstream corporation tax) ACT is not a tax which is charged on, or payable in respect of, or which is therefore "corporation tax in respect of", "income or chargeable gains". That conclusion seems to be reinforced when one looks elsewhere in the 1988 Act.
The provisions of Chapter V of Part VI of the 1988 Act, which contain sections 238 to 246, are concerned with "Advance Corporation Tax and Franked Investment Income". It seems to me that, in terms of both their effect and the way in which they are expressed, those provisions support the Revenue's contention that ACT is not corporation tax "in respect of income or chargeable gains". The important point about their effect in the present connection is that, while payments of ACT can be set off against liability for mainstream corporation tax (and hence could be said to be indirectly "in respect of income or chargeable gains"), it is clear that there will be cases where that is not so – for instance, where the company concerned does not have "profits" in the past, present or future against which the ACT can be set off.
As to the terms in which the sections in Chapter V of Part VI of the 1988 Act are expressed, section 239(1) refers to mainstream corporation tax as "liability to corporation tax on any profits charged to corporation tax", consistently with sections 6 and 8. By contrast, section 239(1) refers to ACT as "paid .... in respect of any distribution", an expression which is consistent with section 14, and is substantially repeated in subsections (2), (3) and (4), and in sections 241(1) and 246(1). Similarly, section 240(1) and (2) refer to ACT as having been paid "in respect of a [or one] dividend".
In other words, the 1988 Act, when dealing with the interrelationship of ACT and mainstream corporation tax, first makes it clear that there will be circumstances in which ACT will not, even indirectly through set-off against mainstream corporation tax, be connected with profits, and secondly acknowledges that mainstream corporation tax is charged on profits (i.e. on income and chargeable gains), whereas ACT is payable in respect of distributions.
The Revenue's case on this point appears to be further supported by the contrast between the reference to "corporation tax" in the opening part of section 788(3) and the reference to "corporation tax in respect of income or chargeable gains" in section 788(3)(a), and indeed in section 788(1)(b). As Lloyd LJ put it in paragraph 50 of his judgment in the Court of Appeal, it appears "that 'corporation tax' is used as a general category and 'corporation tax in respect of income or chargeable gains' as a subset". If that is right, then, so far as I am aware, the only candidate for a type of corporation tax which is in the "general category" but is excluded from the "subset" would be ACT.
Mr Aaronson sought to meet that point by relying on the legislative history of section 788 of the 1988 Act, namely section 347(1) of the Income Tax Act 1952, sections 39(1) and 64(1) of the Finance Act 1965, section 497(1) of the Income and Corporation Taxes Act 1970, and section 100(1) of the Finance Act 1972. In effect, he said that the reference in section 788(1)(b), and therefore in section 788(3)(a), of the 1988 Act to income or chargeable gains is attributable to the fact that, when the concept of chargeable gains was first introduced, they were originally dealt with separately from income in the legislation. Accordingly, he argued, the centrally relevant words "in respect of income or chargeable gains" are included either as an evolutionary survival or, rather than with a view to limiting the ambit of the reference to corporation tax, in order to emphasise that both forms of profit are covered. I would not myself question the propriety of invoking the legislative history of an unclear or ambiguous provision in a consolidating statute such as the 1988 Act. However, it does not appear to me that the legislative history in this case justifies our disregarding the natural meaning of the centrally relevant words, particularly as the claimants' case means, at best, that they have no meaning, or, at worst, that they are positively misleading.
That opinion is reinforced in the present case by considering the White Paper which gave rise to the introduction of ACT (Reform of Corporation Tax (1972) (Cmnd 4955). As Mr Glick said, it is made clear in paragraphs 22 to 25 and 32 of that Paper that, while the proposed introduction of ACT included the rights of a UK subsidiary to elect not to pay ACT in respect of dividends paid to a UK parent and, if it did not so elect, to surrender its ACT credit to its UK parent, these rights would not apply in the case of a foreign parent. This was reflected in clauses 93 and 95 of the draft Bill attached to the White Paper, which in due course became sections 98 and 100 of the Finance Act 1972 (now substantially re-enacted in the 1988 Act). If one is to consider the legislative history, therefore, it seems to me that it is very unlikely that the draftsman of the 1972 Act, introducing ACT substantially in accordance with the proposals in the White Paper, could have intended section 98 or 100 to have the effect of enabling UK subsidiaries and their foreign parents, even based in countries with whom there was a DTC in standard form, to elect or surrender as if the parents were UK-resident, or to be compensated if they could not do so.
Mr Aaronson pointed out that, if the Revenue were right as to the effect of section 788, it would be open to the legislature to impose more penal rates of ACT on subsidiaries of foreign companies than on subsidiaries of UK companies, and that this would be a breach of articles 25(3) and (5) of the DTC with Japan (and article 24(5) read together with article 2(4) of the DTC with the US), which prohibit discrimination in respect of "taxes of every kind and description" in such cases. In my view, there is nothing in this point. The legislature can give effect to such articles in two ways (which are not mutually exclusive), one positive and the other negative. The first is specifically to enact a stipulation such as section 788 to ensure that the articles have effect; the second is not to enact any legislation which infringes the articles. In relation to discriminatory provisions such as those described by Mr Aaronson, the legislature has simply taken the latter course. In the unlikely event that the legislature did enact such provisions, appropriate steps would no doubt be taken under the mutual agreement procedure contained in article 26 of the DTC with Japan (and article 25 of the DTC with the US).
Mr Aaronson also relied on what Diplock LJ said in Salomon v Comrs of Customs and Excise  2 QB 116, 143, to the effect that, where a statutory provision was "not clear" and was "reasonably capable of [bearing] more than one meaning", the court should favour "the meaning which is consonant" with the UK's treaty obligations. This principle is of less weight in a case such as the present where there is no question, as in Salomon, of the legislative provision in issue, namely section 788 of the 1988 Act, having been enacted to give effect to a specific treaty obligation. In this case, section 788, while enacted to enable the UK's treaty obligations under DTCs generally to have effect in domestic law, was plainly not designed to give effect to any specific obligation or even any specific Convention. Nonetheless, the point would have some force here on the counter-factual hypothesis that the claimants succeed on the first issue. Nonetheless, it would, in my view, have been quite insufficient to overcome the Revenue's arguments as to why the claimants cannot succeed on the second issue, namely that, on a fair reading of the 1988 Act as a whole ACT is not within the ambit of section 788(3)(a).
Metallgesellschaft Ltd v Inland Revenue Comrs (Joined Cases C-397 and 410/98)  Ch 620
Pirelli Cable Holding NV v Inland Revenue Comrs  1 WLR 400
Test Claimants in FII Group Litigation v Comrs of Inland Revenue (Case C-446/04) (unreported), 12 Dec 2006
Test Claimants in the Thin Cap Group Litigation v Comrs of Inland Revenue (Case C-524/04) (unreported) 13 March 2007
Baars v Inspecteur der Belastingen Particulieren/Ondernemingen Gorinchem (Case C-251/98)  ECR I-2787
Salomon v Comrs of Customs and Excise  2 QB 116
Income and Corporation Taxes Act 1988: s.6, s.8, s.14, s.247, s.788
Double Taxation Relief (Taxes on Income) (Japan) Order 1970 (SI 1970 No 1948): Art.26, Schedule
Double Taxation Relief (Taxes on Income) (The United States of America) Order 1980 (SI 1980 No 568): Art.24, Schedule
EC Treaty: Art.56, Art.57
Authors and other references
White Paper, Reform of Corporation Tax (1972) (Cmnd 4955)
Graham Aaronson QC, David Cavender & Hugh Mercer (instructed by Dorsey & Whitney) for the appellants.
Ian Glick QC, David Ewart QC & Kelyn Bacon (instructed by Solicitor to Her Majesty's Revenue and Customs) for the respondents.
all rights reserved