General Anti-Abuse Rule Berwin Leighton Paisner LLP's comments on draft legislation and guidance published 11 December 2012

Transcription

1 Introduction In our response to the consultation on the proposed general anti-abuse rule ( GAAR ) that ran to 14 September 2012 we highlighted a number of serious constitutional problems with the GAAR. Although some improvements have been made since then (e.g. HMRC not sitting on the advisory panel), most of the problems have been exacerbated by the draft guidance and the process for its approval by an interim advisory panel. We set out below: a summary of our key concerns; comments on the draft legislation and Part A of the guidance; comments on some of the examples in Part B of the guidance; and some further inheritance tax examples for possible inclusion in the guidance. Key points Our key concerns are: the concept of exploiting shortcomings in legislation, which remains emotional and poorly defined; loosely worded and unsubstantiated statements in the guidance about legislative purpose or policy objectives; the risk of taxpayers being taxed on an entirely hypothetical transaction and an amount which is significantly greater than the return that they actually receive; consequential relieving adjustments should be available when the GAAR is self-assessed; whether a court will be able to take into account established practice that has not been expressly accepted by HMRC; loose wording in examples of where the GAAR does not apply, which imply similar fact patterns where it might apply; the lack of any discussion of reasonableness in the examples in Part B of the guidance; the guidance needs to explain what a just and reasonable counteraction would be; the appointment and role of the interim panel; the procedure for the panel approving the guidance; and HMRC s refusal to be bound by a unanimous panel opinion in the taxpayer s favour. Draft legislation and Part A Clause 2 Double reasonableness None of the examples in the guidance actually discuss reasonableness. They all look solely at the factors in clauses 2(2)(a) to (c). Are these intended to form the heart of the test? Some further guidance on the weight to be put on these factors would be helpful. Page 01 Berwin Leighton Paisner

2 Shortcomings We remain particularly concerned about clause 2(2)(c). It is not clear what constitutes a shortcoming and there seem to be at least four different interpretations in the examples given in the guidance: some of the examples treat as a shortcoming the fact that the legislation allows a tax result contrary to the purpose/principles/policy (e.g. the SDLT sub-sale example), which essentially just repeats clause 2(2)(a); some examples say there is a shortcoming because the legislation is being applied mechanistically (e.g. the working wheels example); some simply say there is a shortcoming because a particular provision is being exploited but without identifying any particular defect (e.g. the Blumenthal and Huitson examples), which is no more than saying there is a shortcoming because the legislation allows a result that (in someone s view) should not be allowed; and others do point to a specific drafting defect (e.g. the shares as debt and capital allowances examples) but again this seems to amount to nothing more than the legislation allowing something contrary to its intended purpose. The difficulty of identifying a shortcoming will in turn create problems when trying to determine whether one has been exploited. As we said in our original response to the consultation, the concept of exploiting a defect in legislation is an emotional one and not fit for legislative purpose. It becomes even more objectionable if a taxpayer is to be regarded as exploiting a shortcoming even when no specific defect can be identified or simply because the tax result is claimed to be inconsistent with an implied underlying principle or policy objective. Principles and policy objectives As policy and principles will be central to reasonableness test (including identifying any shortcomings ) there is a risk of legislative history being re-written in the guidance. Any statement in the guidance about principles or policy objectives should, therefore, be properly substantiated with evidence of Parliament s intention (e.g. by quoting from explanatory notes) and not be based solely on common sense or HMRC s view. We expect that consultation documents will feature heavily in the materials used to determine principles and policy objectives. With that in mind, we would welcome guidance on how a decision not to take forward a particular proposal after consultation or to legislate in a different way should be taken into account when determining the principles or policy objectives behind legislation that is enacted. This can provide useful evidence of Parliament s intention. Clause 4 The guidance on identifying a tax advantage (4.3 of Part A) suggests the appropriate comparator will usually derive from the arrangements that would have occurred absent the relevant tax purpose. We are concerned that this formulation allows scope for a taxpayer to be taxed on a greater economic return than they have actually received (see, for example, the Australian case Federal Commissioner of Taxation v Spotless Services Ltd [1996] HCA 34). It should be made clear that the GAAR cannot have this effect. Page 02 Berwin Leighton Paisner

3 Clause 5 Although self-assessment of the GAAR may be relatively rare, we can see no reason why a consequential relieving adjustment should not be available in those circumstances. It is inequitable if adjustments cannot be made following self-assessment of the GAAR. Clause 6 Please confirm that clause 6(3)(b) is not intended to prevent a court from taking into account evidence of established practice that has not been expressly accepted by HMRC. Further guidance is needed on what is meant by HMRC indicating its acceptance. Guidance Part B General points The guidance needs to be more carefully worded and comments that might cause uncertainty, such as caveats in examples where GAAR does not apply, need to be avoided. For example, the implication of paragraph of Part B is that if the shareholder had asked for loan notes at a time when he had an intention to leave the UK then the transaction may be caught by the GAAR. Comments such as this, which hint that a slightly different fact pattern might be caught, severely limit the usefulness of the examples. Taxpayers will be concerned that a similar fact pattern that does not fall squarely within the example could be caught. As we have noted above, any statement in the guidance about principles or policy objectives needs to be properly substantiated with evidence of Parliament s intention. There needs to be some discussion of reasonableness beyond working through the factors set out in clauses 2(2)(a) to (c). The examples where the GAAR applies should set out what the counteraction would be and explain why it is just and reasonable. Some of the examples conclude that the GAAR applies (e.g. paragraph of Part B). Most conclude that the arrangements are abusive arrangements to which HMRC would seek to apply the GAAR (e.g. paragraph 4.2.6). The latter would appear to be a less forceful conclusion as it is just a statement of HMRC s view. Is there intended to be a difference (bearing in mind that the court must have regard to this guidance)? Capital allowances double dip What is the authority for the statement that part of the purpose of the capital allowances code is only to relieve expenditure once (see paragraph )? That may be a common sense point of view. However, given that principles and policy objectives are at the heart of the double reasonableness test, allowing the creation of new statements of principle for the purposes of the GAAR guidance will create huge uncertainty. In the case of capital allowances, the long funding lease rules have layered deeming provisions on top of other deeming provision in the main capital allowances code (e.g. section 13 Capital Allowances Act 2001, which specifically provides for allowances to be given for notional expenditure). The underlying policy of some of these provisions is not as obvious as the guidance suggests. Page 03 Berwin Leighton Paisner

4 Any statement about principles or policy objective will have far reaching effects beyond the immediate example in which it is made. A court will have to have regard to, for example, the statements about the capital allowances code when considering the application of the GAAR to any claim for capital allowances. Paragraph suggests the arrangements are intended to exploit a shortcoming because no regard is had to the purpose. Is the shortcoming, therefore, that the legislation does not properly reflect the purpose? If so, how is this different from applying clause 2(2)(a)? Blumenthal QCB/Non-QCB Paragraph suggests that the scheme exploits a shortcoming because it relies upon an artificial reduction in value. Is the artificiality not relevant to clause 2(2)(b) rather than 2(2)(c)? Paragraph goes on to say that the scheme seeks to exploit the market value and frozen gain provisions. This may be true. However, what shortcoming in those provisions does it exploit? No apparent defect is identified. If the shortcoming is that the taxpayer can utilise the interaction of two tax provisions to achieve a result that is contrary to the principle underlying one or both provisions then this should be explained more clearly in the guidance. However, again, how is this different from applying clause 2(2)(a)? The central point with this scheme is that part of the purpose and policy objective of roll-over relief for QCBs is that the gain is eventually taxed (unless, presumably, it is relieved by another relief). Clause 2(2)(a) is, therefore, key to the analysis (along with 2(2)(b)) not 2(2)(c). SDLT sub-sales Paragraph states that the intention of the sub-sale rules is that there should be a charge on C. However, what if C a charity or qualifies for some other relief? This demonstrates the problem of including loosely drafted and unsubstantiated statements about policy or purpose. In this example, would it be more accurate to say that the intention is that C pays SDLT unless it qualifies for a relief and both A to B and B to C are bona fide commercial transactions? Paragraph states that the arrangements exploit shortcomings because section 45 is in place to assist bona fide commercial transactions. Again, this is just to repeat the test in paragraph 2(2)(a). It would be extremely helpful to the functioning of the commercial real estate market if the introduction of the GAAR were to lead to a significant simplification of the SDLT code. Section 75A should be repealed, along with the redundant anti-avoidance provisions in paragraph 17A of Schedule 15. We also encourage HMRC to look carefully at whether the excessive amount of drafting and complexity in the new transfer of rights provisions is necessary in light of the GAAR. Late paid interest Paragraph states that the tax result is consistent with the purpose of the legislation. How does this square with HMRC s view, quoted in paragraph , that such arrangements are contrary to the spirit of the law and intentions of Parliament? Simple QCB/NQCB Although it is intended to be helpful, for the reasons given in our general comments on the guidance above, we further this example may end up causing more problems for genuine commercial transactions. Page 04 Berwin Leighton Paisner

5 Given the intended scope of the GAAR, we would welcome a more explicit statement in the guidance that the use of non-qcbs or QCBs as consideration on genuine commercial transactions should not be caught by the GAAR. We note from paragraph that HMRC accepts that chargeable gains might be deferred. However, in this example the tax is avoided completed. Advisory Panel Approval of guidance We understand that the guidance will be approved by all members of the panel. The legislation does not set out what constitutes approval for these purposes. Will it require unanimity or a simple majority? Given the likely size of the full panel the former would obviously raise some practical difficulties. However, given the spread of interests to be represented on the panel, allowing approval by a majority would raise concerns about whether that spread of interests is properly represented in the majority. We understand that matters such as this will be dealt with in the panel s terms of reference. Given the status of the guidance it is important that these are published as soon as possible and are consulted on properly. Interim panel The constitutional objections to the advisory panel that we set out in our original response to the consultation have been exacerbated by the use of an interim panel to approve the guidance before Royal Assent. Power to draft and approve guidance with quasi-legislative effect has been conferred on an unaccountable and unelected group of professionals, some of whom have a direct professional interest in the application of the GAAR. Furthermore, this has been done in an entirely opaque manner with no explanation of the selection process or criteria or of the interim panel s terms of reference. This must be rectified. HMRC should publish all correspondence relating to the appointment of Graham Aaronson QC and the other members of the interim panel and to their terms of reference. Parliament will shortly be asked to give statutory force to the guidance approved by this interim panel. It must know all the facts before doing so. We acknowledge that Gary Richards of this firm is a member of the interim panel. He has been appointed in a personal capacity and has had no hand in drafting these representations. Panel opinions We understand the decision to ask the panel a single reasonableness question as this will distinguish the role of the panel from that of the courts. However, it makes HMRC s refusal to be bound by a unanimous decision that arrangements are reasonable even more untenable. If all the panel members conclude that arrangements are reasonable then for HMRC to win in court it would effectively need to show that they were all being unreasonable (because it could not reasonably be regarded as a reasonable course of action). That would surely make the positions of the panel members untenable and undermine the existence and role of the panel generally. On the other hand, if a panel concludes that arrangements are unreasonable then the taxpayer need only show in court that it could be a reasonably held view that they are reasonable. Hence taxpayers should not be bound by a panel decision that arrangements are unreasonable. At the very least HMRC should confirm in the guidance that only in exceptional circumstances will it seek to apply the GAAR after a unanimous panel decision that arrangements are reasonable. Page 05 Berwin Leighton Paisner

6 We are concerned about the suggestion that panel members will be added ad-hoc if a person of appropriate experience to consider a particular case is missing. Panel appointments must be made transparently and within carefully defined parameters. Inheritance tax examples The application of the GAAR to inheritance tax is bound to be problematic. This is a highly technical tax. It applies to transfers in lifetime and on death but is subject to many exceptions and reliefs. Every well advised wealthy individual will normally undertake some form or other of estate planning. It will often, therefore, be very difficult to discern whether arrangements cannot reasonably be regarded as a reasonable course of action. The indicia in section 2(4) are entirely inapplicable to a tax on gifts and estates. Part B of the draft guidance gives three IHT examples. At paragraph 8.1 an example is given of a taxpayer realising assets, investing the proceeds in a farm and having held the farm for a sufficiently long period that it qualifies for agricultural property relief gifting the farm into a trust. This is such a mild example of estate planning that it is inconceivable that anyone would consider this to be abusive arrangement. However, the example in paragraph 4.2 (property excluded from charge) is extremely artificial and would likely fall within the GAAR. On the other hand, the example in paragraph 4.1 where the taxpayer gifts a reversionary lease is more questionable but we believe that most practitioners would regard this arrangement as a reasonable course of action. It was recognised from the earliest days of estate duty, which contained gift with reservation provisions, that the legislation only applied where the donor enjoyed a benefit in the property interest gifted. The legislature has therefore had many opportunities to broaden the circumstances in which the gift with reservation provisions apply but has only done so in limited circumstances, such as by section 104 Finance Act 1999 (enacted to counteract Ingram type lease arrangements). There are many routine estate planning arrangements in relation to which practitioners would wish to have confirmation that HMRC would not seek to apply the GAAR provisions. Some examples: 1. Mr X owns all the shares in a company. The share capital of the company is reorganised, so as to create a new class of growth share, all of which are issued to Mr X as bonus shares. Mr X gives these away to his children or into a trust, but continues to retain his original shares upon which he receives dividends. Mr X dies more than seven years later, so that only the original shares, now of limited value, are comprised in his estate and liable to IHT. 2. Mr X establishes with a nominal sum a settlement for the benefit of his children. Mr X lends a large sum, free of interest and repayable on demand, to the trustee. This sum is invested and the investment returns arise outside Mr X s estate. Mr X from time to time receives loan repayments. On Mr X s death only the balance of the loan outstanding is comprised in his estate and liable to IHT. 3. If a person is not domiciled in the UK then in principle a charge to IHT only applies to a gift of UK situate assets and on death to those assets comprised in his estate which are situate in the UK. However, under section 267 IHTA 1984 a person becomes deemed domiciled in the UK for IHT purposes when he has been resident in the UK for 17 years of tax assessment. So Mr X, just before he becomes deemed domiciled, settles foreign situate assets in a trust under which he is a beneficiary. On his death, domiciled in the UK, the foreign assets held in trust are considered to be excluded property and free of IHT because he was not domiciled in the UK when he established the settlement (section 48(3) IHTA 1984). Berwin Leighton Paisner LLP 6 February Page 06 Berwin Leighton Paisner