.png)
.png)
I HAVE WRITTEN articles for Taxation on many topics, from TA 1988, s 660A and IR35 to tax credits and interventions. But this is the first time that I have taken up my pen to defend HMRC. Regular readers may be wondering what has happened. Have I lost my crusading zeal? Am I hoping for a sinecure, a peerage, or merely a ministerial Christmas card? Read on …
Disclosure of tax avoidance schemes
In two Taxation articles towards the end of last year, 'Suggestio falsi', (Taxation 19 October 2006, page 53) and 'A costly burden' (Taxation 9 November 2006, page 144), Simon McKie argued forcefully that the tax avoidance disclosure rules are so broad that most tax advice is disclosable. He asked whether advisers who do not disclose normal tax planning are, in consequence, breaking the ethical guidance of their professional bodies, as well as risking the imposition of penalties. If Simon is right, this is a very important issue for practitioners, professional institutes and HMRC, as well as for taxpayers generally.
Further impetus to consider these questions was provided by the Pre-Budget Report consultation document, Ensuring Compliance with the Tax Avoidance Disclosure Regime. This seeks to increase HMRC powers 'where there are reasonable grounds to believe that a promoter has failed to comply with the tax avoidance disclosure regime'. If Simon is correct in his views, most tax advisers would fall within the scope of this new harsher regime. Conversely, if Simon is wrong, but people believe him to be right, there is a risk that the disclosure rules will fall into disrepute. As a result, the regime may fail in its underlying purpose of discouraging structured tax avoidance.
The governmental response to a failure of the disclosure regime is likely to involve sweeping anti-avoidance legislation, some of which will be retroactive. The pre-owned asset rules are a good (or bad) example of such broad-spectrum legislation. The disclosure regime does not, of course, preclude the introduction of poorly targeted anti-avoidance provisions, but because it limits structured artificial avoidance, it reduces the likelihood of draconian legislation. If Simon's analysis is correct, it will thus impact not only HMRC and advisers, but ordinary taxpayers and the scope of future avoidance legislation.
Regular readers of this magazine will know that Simon is an experienced professional with a detailed knowledge of the law. I share the general respect for his intelligence and integrity, even though, as will become obvious, I disagree with his views on disclosure.
Summary of the articles
On 1 August 2006, the new Tax Avoidance Schemes (Prescribed Descriptions of Arrangements) Regulations (SI 2006 No 1543) took effect. Simon's first article states that in consequence 'almost all advice in relation to income tax, capital gains tax and corporation tax … is disclosable'. His second concludes with the words:
'Virtually all arrangements relating to income, capital gains and corporation tax will be hallmarked schemes as they are defined in the regulations.'
In 'A Costly Burden', Simon considers the governing legislation in FA 2004, s 306. This section sets out three conditions for an arrangement to be notifiable under the disclosure rules. One of the conditions is that the arrangement falls within one or more of the 'hallmarks' set out in SI 2006 No 1543.
Simon then moves on to consider these hallmarks, using an extensive worked example ('the Cordelia structure') in which an adviser recommends a well-known trust structure to mitigate capital gains tax on a gift. The article's conclusion is that this 'entirely routine private client tax planning, no doubt already well known to HMRC, is disclosable under the disclosure rules because it falls under no less than three of the seven hallmarks'.
For simplicity, my article follows Simon's pattern, looking first at the three conditions and then at the hallmarks.
The three conditions
FA 2004, s 306 defines 'notifiable arrangements' as those which satisfy all three of the following conditions. They must:
- fall within any description described by the Treasury by regulations (Condition (a));
- enable, or might be expected to enable, any person to obtain an advantage in relation to any tax that is so prescribed in relation to arrangements of that description (Condition (b)); and
- are such that the main benefit, or one of the main benefits, that might be expected to arise from the arrangements is the obtaining of that advantage (Condition (c).
I agree with Simon that the word 'arrangements' has a very wide meaning; I also accept that Condition (b) is largely ineffective, because the definition of tax advantage set out in TA 1988, s 709 is so wide, covering as it does 'a relief or increased relief from, or repayment or increased repayment of, tax, or the avoidance or reduction of a charge to tax or an assessment to tax or the avoidance of a possible assessment thereto …'.
When considering Condition (c), Simon acknowledges that 'if a transaction has a preponderant non-fiscal benefit the fact that it also confers a significant tax advantage will not be sufficient to bring it within (c)'.
In Example 1 (my numbering) below, he demonstrates how this condition works:
Example 1
'There is often a choice between selling the shares in a company which owns a valuable asset or the company selling the asset. The former route might be chosen because the shares have a high base cost whereas the asset does not. One might argue that in that case although a tax advantage has been obtained (because there is an alternative transaction which would have resulted in a higher tax charge) that tax advantage is not the main benefit of the arrangement. The main benefit of the arrangement is the receipt of the sale proceeds rather than the avoidance of the capital gain.'
Again, I agree with this analysis. From it, I would draw the conclusion that this condition does, in fact, exclude a significant number of transactions from the disclosure regime.
However, Simon then seeks to ascertain whether the taxpayers in Furniss v Dawson [1984] STC 153 would have fallen within this condition. He spends some time trying to establish which steps taken by the taxpayers constitute the 'arrangements' for the purposes of the main benefit test, and finally concludes that it is unclear whether one of the main benefits of the Furniss structure was the obtaining of a tax advantage.
But since Furniss v Dawson was a structured tax avoidance scheme, it is hardly an appropriate benchmark against which to test ordinary tax planning. Even if this were otherwise, the answer to the main benefit test seems pretty clear: Simon begins his analysis of the case by saying (my italics) that the Furniss v Dawson case:
'involved the sale of a company. In order to avoid capital gains tax on the sale the shareholders sold their shares in exchange for shares in a Manx company which in turn sold them to third party purchasers.'
Simon subsequently considers whether Condition (c) applies to his Cordelia tax planning structure, focusing on the question of whose benefit is to be considered — the giver, the recipient or both together. He concludes, again, that it 'is far from clear' whether the structure falls within the condition.
However, I did not find the question so difficult. The legislation directs us to look at the 'arrangements'. The fact pattern provided demonstrates that the trust structure (as distinct from the gift) was only implemented in order to achieve a tax advantage. In my view, the planning thus falls within Condition (c). But, as Example 1 makes clear, this will not always be the case. However, Simon says this (again, my italics):
'Whatever is the scope of Condition (c) it is clear that a very large amount of ordinary tax planning will constitute notifiable arrangements for the purposes of the disclosure rules unless they fall outside … Condition (a).'
He thus undermines the general, and valid, conclusion which I believe should have been derived from our Example 1 above, namely that this condition is a useful limitation on the scope of the disclosure provisions.
Hallmarks
Simon finally looks at Condition (a), which naturally leads him to the seven hallmarks set out in the August regulations (SI 2006 No 1543). He concludes that the Cordelia structure fails three: confidentiality, premium fee, and standardised tax products. I consider each in turn.
Confidentiality from other promoters
There are two parts to the confidentiality test: confidentiality from other promoters, and confidentiality from HMRC (Condition 1 is merely a link necessary to the structure of this paragraph as a whole).
The first leg of the test applies if:
'it might reasonably be expected that a promoter would wish the way in which that element of those arrangements secures a tax advantage to be kept confidential from any other promoter at any time in the period beginning with the opening date and ending with the appropriate date.' (SI 2006 No 1543, para 6(2)).
Simon rightly points out that the scope of this section relates to 'a promoter' not to 'the promoter' and that it also refers to 'any other promoter'. It is thus wider than if it referred only to the actual promoter of the scheme and/or to known competitors of that promoter. He then states that:
'The condition will be satisfied if any one of that massive population of possible promoters might reasonably be expected to wish the way in which that element of those arrangements secures a tax advantage be kept confidential from any other promoter … [if] any two firms of accountants operating in competition in any place would wish to prevent the other from knowing how an effective gift of an asset may be made without triggering a capital gains tax charge then the condition is satisfied … there are many tax advisers who are not aware of all the available standard planning techniques and if another adviser could ensure that that adviser was kept in his state of ignorance he would maintain his competitive advantage.'
There is, however, a difference between 'kept confidential' and 'ignorance'. Firm A, (which is familiar with the Cordelia structure) has a competitive advantage over Firm B, which is not. But if Firm B chooses to find out how the Cordelia structure works, it can do so, because the information is not 'kept confidential': it can be found by phoning a friend, attending a lecture, or buying a book.
Thus the condition is not satisfied merely because Firm A uses a planning technique of which Firm B is ignorant: the 'kept confidential' test is a much higher hurdle.
Secondly, the paragraph is hypothetical. It asks whether an independent third party might reasonably expect that someone selling, say, the Cordelia structure, would want to keep it confidential from any other promoter. It would be pure fantasy for an adviser to wish that a well-known piece of tax planning was known only to him and not to anyone else. Of course, an adviser could wish for this — just as he could wish to be reincarnated as Elvis in his next life — but it would not be reasonable for a third party to expect the adviser to be harbouring this secret, unfulfillable desire.
Confidentiality from HMRC
The second (and alternative) leg of the test is that: 'the promoter would, but for the requirements of these regulations, wish to keep the way in which that element secures that advantage confidential from HMRC for some or all of the period beginning with the opening date and ending with the appropriate date, and a reason for doing so is to facilitate repeated or continued use of the same element, or substantially the same element, in the future'. (SI 2006 No 1543, para 2.)
This subparagraph does not refer to 'a promoter', but 'the promoter' — it is thus a question of fact as to whether the promoter of the arrangements under consideration would have wished to keep the details of the scheme confidential from HMRC if the disclosure rules did not exist. The HMRC guidance states that 'the relevant question(s) must be consciously answered at the time the relevant trigger for disclosure arises'.
Simon believes that the Cordelia structure also falls within this hallmark, saying:
'Any adviser recommending a tax strategy to a number of clients is likely to wish, if it were possible, that the way in which the tax planning works would not become known to HMRC for a period of time.'
Of course, if a scheme is innovative, the adviser will wish that he didn't have to disclose it: this is why the test exists in the first place. But is it true of schemes with which HMRC are already familiar? Why would an adviser have such a wish when HMRC already know about the scheme?
Simon defends his stance by saying that 'The fact that one thinks that one cannot obtain a thing does not prevent one wishing for it'. However, it seems to me prima facie unlikely that a real-life adviser would have such a wish. Furthermore, the subparagraph specifies that, to fall within this hallmark, the promoter must have this wish in order 'to facilitate repeated or continued use of the same element, or substantially the same element, in the future'. It seems bizarre for the promoter to wish HMRC didn't know about the scheme, so that he can continue to use it — when he has been applying the methodology for many years with HMRC's full knowledge.
In my view, the Cordelia structure does not fall within this hallmark as:
(a) the promoter would have no reason to wish for it to be kept confidential from HMRC, who already know how it works; and
(b) would have no reason to have such a wish in order to facilitate its repeated use.
Premium fee
A 'premium fee' is defined as: 'A fee chargeable by virtue of any element of the arrangements (including the way in which they are structured) from which the tax advantage expected to be obtained arises, and which is to a significant extent attributable to that tax advantage'. (SI 2006
No 1543, para 8.)
Simon considers that the Cordelia structure falls within the premium fee hallmark saying that 'The fee was to a significant extent attributable to the tax advantage because it is in order to obtain advice leading to the tax advantage that [the promoter's] services were engaged'.
This is later generalised to:
'If the skill or reputation of the adviser which enables him to charge more for his advice than others might, is a skill in taxation planning it will be attributable to the advantage which the client expects to obtain from the advice. Not every professional adviser has substantial taxation expertise and clients who engage advisers with such substantial expertise do so because they expect the adviser's advice to be valuable. If advice on taxation is valuable, almost inevitably it will confer a tax advantage.'
However, this analysis takes the definition of premium fee out of its legislative context. A premium fee is one which the promoter could reasonably be expected to receive 'but for the requirements to disclose information under the regulations' from 'a person experienced in receiving services of the type being provided' (SI 2006 No 1543, para 8). Thus, this hallmark is not about the tax skill levels of the adviser. It is about the fee that could have been charged to the client if the disclosure rules did not exist. In other words, it does not catch ordinary tax planning, because the fees for such planning would be the same both before and after the introduction of the disclosure rules. It is designed to catch the sort of tax planning arrangements for which advisers would have been able to bill premium fees had there been no requirement to disclose.
Standardised tax products
Finally, Simon considers that the Cordelia structure is a standardised tax product. I agree that the structure, and most ordinary tax planning, falls within the definitional part of this hallmark. However, there is an exception from the hallmark's application for arrangements 'which are of the same, or substantially the same, description as arrangements which were first made available for implementation before 1 August 2006' (SI 2006 No 1543, para 11). This means that the Cordelia structure, and other commonplace tax planning, falls outside the disclosure rules providing they were known about before 1 August 2006. Further guidance on this is given at paragraph 6.6.6 of the HMRC booklet.
Over time, of course, as new 'commonplace structures' arise and this timeline retreats, this hallmark may become more problematic, but for the present it ensures that ordinary planning (such as the Cordelia structure) which would otherwise fall within this hallmark is excluded.
Ethics and commerce
In his first article, Simon asked whether — given that the disclosure legislation is too broad — the institutes' ethical guidance (that tax advisers must abide by the law, including that on disclosure) should be relaxed. He supported this suggestion by pointing out that HMRC do not want thousands of unnecessary disclosures.
Since I do not share his view of the disclosure regulations, I have not considered this question further. However, closely juxtaposed with this point — that the rules could be ignored because they are too broad — is a second, more worrying question. This is whether professionals may take the view that their 'commercial interests would be better served by accepting the risk of incurring a civil penalty for non-compliance'.
This is expanded as follows: 'the penalties imposed for failures to make returns, however, are civil penalties not criminal penalties … those who commit crimes are subject to the opprobrium of society. Civil wrongs are not morally equivalent to crimes'.
It is clearly unethical for an adviser not to disclose tax planning arrangements just because the commercial benefits of so doing outweigh the financial penalty of disclosure, whether or not this behaviour attracts 'the opprobrium of society'.
Simon concludes his first article by saying 'Professionals in the UK continue to recognise their duty to make disclosure in accordance with the law … I hope that they will continue to do so'. Naturally, I share that hope.
Conclusion
When the disclosure rules were introduced in 2004, HMRC announced that they would be closely targeted. Simon believes that this statement was untrue, and this belief reflected in the title of his first article, 'Suggestio falsi'. For the benefit of readers lucky enough not to have studied Latin to A level, this comes from the phrase, 'suppressio veri, suggestio falsi', which means 'Suppressing the truth is equivalent to creating a lie'.
In his second piece, Simon says that HMRC's assumption that the hallmarks are effective contains 'a pinch of truth and a tablespoon of falsehood': hence the name of this article.
It is good that the scope of new legislation is closely examined to ensure that it is fully understood and complied with; however, as you will probably have gathered by now, I do not share Simon's view of the scope of the disclosure rules. If correct, that interpretation would either result in a deluge of disclosures to HMRC or the possibility that large swathes of the tax profession would be ignoring their legal responsibilities; naturally, this would be in no-one's best interests. The rules were, of course, subject to significant consultation and are, I believe, both proportionate and workable.
Anne Redston CTA (Fellow), FCA is a Visiting Professor in Law at King's College London and Chair of Personal Taxes at the Chartered Institute of Taxation. The views expressed in this article are her own. Anne can be reached via aredston@ciot.org.uk.
