.png)
.png)
ONE OF THE FIRST articles I wrote for Taxation in 1992 was called 'The Brightman Line'. Written after the House of Lords decision in Ensign Tankers (Leasing) Ltd v Stokes [1992] STC 226, the article looked at how the law was developing on tax avoidance. Twelve years later, in my first comment article as editor, I find myself writing about the same subject. So what has changed?
One noticeable difference is that practitioners do not talk about tax avoidance anymore; or if they do it is to distinguish the virtuous and ethical advice that they give — tax planning, tax mitigation — from the evil and disreputable industry of tax avoidance. Whether this is anything more than just a fiscal version of political correctness is a question I will return to later.
But perhaps the biggest difference is that all the attention is now focused on the legislature, whereas in 1992 it was focused on the courts. For the first time, a Chancellor chose to address the general issue of tax avoidance in a Finance Act. It was a brave move, and he deserves full marks for effort. Attainment, however, is a different matter …
The most obvious statutory attack would be a general anti-avoidance rule. A transaction entered into for avoidance purposes would be deemed ineffective for tax purposes. But how do you define a tax-avoidance transaction? Lord Templeman, that redoubtable opponent of tax avoidance, was prepared in Ensign Tankers to accept the validity of a pre-1998 bed and breakfast share transaction, even though some commentators felt it was a clear example of a pre-ordained and circular transaction.
A cunning plan …
To require, instead, that tax avoidance schemes simply be disclosed seemed to be a clever trick. By the time that the Inland Revenue gets to know about the most effective tax-avoidance schemes, millions of pounds have been lost, or saved, depending on your point of view. Forcing very early disclosure would mean that the Inland Revenue could react quickly and announce that legislation would be introduced in the next Finance Bill to block the scheme, backdated to the day of the announcement. The chances of the promoters making any significant profits from the sale of such schemes would be severely reduced. Deprived of the cashflow which sustained it, the tax avoidance weed would wither and die. But the weedkiller could be selective: if acceptable tax planning was inadvertently caught by the disclosure rules, no action need be taken against it.
It was, as Baldrick might have said, 'a cunning plan'. Unfortunately, as viewers of Blackadder know, cunning plans do not always work as expected. If the regulations which identified the arrangements that had to be disclosed were too widely drawn, the Inland Revenue would be flooded with unwanted disclosures. On the other hand, if they were drawn too narrowly, the tax-avoidance industry would get round them.
Financial securities
The area which caused the most difficulty was financial securities. Here, there was a simple way of seeing whether the regulations were too strict. If they forced the disclosure of individual savings accounts as tax avoidance schemes, the new régime was going to be unworkable.
The original draft regulations have been covered twice in Taxation. They were explained in practical terms by Ashley Greenbank and Stephanie Tidball (' Disclosure Rules', 3 June 2004 at pages 239 to 241), and blown apart in a fine piece of polemic by John Tallon QC ('Is Anyone Out There Listening?', 24 June 2004 at pages 319 to 321). The definition centred on a formula, which provided that the tax benefits plus the economic benefits had to be greater than or equal to the costs. That is not a definition of a tax-avoidance scheme; it is a definition of a successful investment! The guidance notes tied themselves in knots trying unsuccessfully to prove that individual savings accounts were not disclosable, and only succeeded in proving to most commentators that they were.
On 22 June, the Government announced that the draft regulations would be changed in the light of the representations made during the consultation period. Dawn Primarolo MP, the Paymaster General, assured us that:
'These changes will ensure that disclosure will only be required of those schemes and arrangements which pose the greatest threat to the Exchequer.'
Specifically we were told that the formula test would be removed, and replaced by a series of filters to differentiate between acceptable and unacceptable schemes. There would also be a list of arrangements that were automatically deemed to be acceptable. But when the new, and supposedly final, regulations came out on 9 July, both the filters and the approved list were disappointing.
As I said, the touchstone of successful regulations on financial securities was that they were drawn in such a way as to exclude individual savings accounts, but the only way that such accounts are excluded in the final regulations is because they are mentioned specifically as an excluded product in paragraph 7(2) to the Schedule. That is not the point. If the other tests are so widely drawn as to include individual savings accounts unless there is a specific exclusion, what other entirely innocent transactions do they cover?
After all, the definition of a financial product is extremely wide, and was widened still further in the final regulations. All loans were now included, and all shares, including ordinary shares that had previously been excluded. Derivatives are quite rightly included, but so are contracts which are, in substance, 'the making of a loan or the advancing or depositing of money'. This, in turn, is defined by reference to section 43A(1), Taxes Act 1988, and appears to be wide enough to include most collective investments such as unit trusts and insurance policies.
The filters
But what about the much-vaunted filters? Would these exclude normal tax planning through financial products from being disclosed? At first sight they looked promising. Paragraph 8 of the Schedule says that at least two, and possibly three, tests need to be met in order for the product to be excluded:
- the premium fee test;
- the confidentiality test; and where appropriate
- the open market test.
The confidentiality test is the easiest to understand. Paragraph 8(3) says that the tax advantage must not arise from an element of the arrangements that a promoter might want to keep confidential from other promoters. Since the first thing that you will be asked to do if you approach an adviser offering a whizzo tax-avoidance scheme is to sign a confidentiality agreement, this is quite an effective and subtle test.
The open market test only applies where the promoter is a party to the product. The test is set out in subparagraph 4, and could have done with rather more punctuation than four commas in a single sentence of 61 words. However, the test appears to be that the financial product which is part of the arrangements must be offered on similar terms to those that could be obtained in the open market for a product which is 'the same or substantially similar'.
The Inland Revenue issued guidance on the application of the regulations on 29 July. The example given for the open market test is that if interest rate swaps were being offered to investors at five per cent fixed, an investor entering into a swap that charged six per cent fixed would expect to receive a premium. If, however, a bank had created an interest rate swap that had a particular tax advantage, the investor would be prepared to accept the swap without a premium. This would fail the open market test.
Why is that? There is no reference in the test to the 'similar product' being one that lacked the tax advantages. Assuming that the advantages are inherent in the product itself, and not in any other part of the arrangements, the test is surely what the investor would accept in a product that had the same tax advantage being offered by another bank? The fact that the product is confidential and not marketed elsewhere is probably irrelevant; the test seems to deem an open market, deem a product to exist which is the same as or substantially similar to the product in question, and then ask what the terms would be. If the products are the same or substantially similar, then the terms are, by definition, surely also the same or substantially similar …?
Premium fee
However, it is the premium fee test which has proved the most controversial; sufficiently controversial for the wording to have been changed slightly from the supposedly final regulations announced on 9 July. The regulations as actually laid before Parliament omit paragraph 8(5)( c ) from the definition of a premium fee, and tidy up some of the wording in other subparagraphs, but still leave a fundamental flaw in the premium fee test — it does not test for a premium fee at all.
The idea of a premium fee test is quite attractive: any scheme as yet unknown to the Inland Revenue which saves a lot of tax is likely to be sold for a lot of money. Paragraph 8(2) imposes an objective test which says that the arrangements must be such that no promoter would expect to be able to charge an experienced client a premium fee for the product, so there is no way of escaping the test by an unrealistic apportionment of fees.
The guidance gives several examples of fees which will not be treated as 'premium fees' simply because they are high — London versus provincial firms, urgency of the advice, etc . They are all entirely irrelevant. The term 'premium fee' does not carry its normal dictionary definition, it has its own definition in subparagraph 5. This says that a premium fee is one chargeable by virtue of any element of the arrangements from which the tax advantage arises, and which is:
- a fee to a significant extent attributable to that tax advantage; and
- a fee to any extent contingent upon the obtaining of that tax advantage.
There is no reference anywhere in the definition to the size of the fee. This is not a definition of a premium fee; it is a definition of a tax planning, or tax saving, fee.
Premium bond …?
Let us consider an example. A client comes into your office complaining that she used to get age allowance, but for some reason this year it has been withdrawn. On investigation, it becomes apparent that her income now exceeds the income limit for age allowance. However, a significant part of her income comes from building society interest. By putting her savings into a life insurance bond invested in a cash deposit fund she can get the same return with no more risk, but provided she is taking out no more than five per cent of her investment each year, it will not be counted as income in calculating her tax liability. You calculate the details for her and introduce her to a financial adviser who will set up the investment (taking a commission). Of course, you bill her for the time spent advising her. It is not a large bill, she is only just over the income limit after all, but still it is a fee for the tax advice that you have given her.
The arrangements, taken together, will result in an income tax saving and involve a financial product as defined by the regulations. They are therefore disclosable unless you meet the conditions for exclusion. The open market test is not relevant, and the confidentiality test does not apply; this is basic tax planning advice. But what about the premium fee? You have not arranged the investment, and therefore have not charged her for it. If she had known about the tax advantage, she could have gone to the financial adviser and bought the insurance bond without incurring any more than the normal commission. The fee you have charged is entirely attributable to the advice and calculations concerning the tax advantages for her in relation to her age allowance, which is the element of the arrangements that gives rise to the tax advantage. You fail the premium fee test, and the transaction is disclosable.
Tertiary legislation
The Inland Revenue does not, of course, want you to disclose it. The guidance notes say that it is the amount of the fee which must be attributable to the tax advantage, not the mere existence of it. As an example, it says that a fee which significantly exceeded the fee that would normally be charged by an adviser for that sort of advice would be a premium fee.
This may be what the Inland Revenue wishes the regulations said, but it is not what they do actually say. John Whiting, giving evidence to the House of Lords committee investigating the Finance Bill, referred to the growing importance of guidance notes in making sense of secondary legislation, and said the guidance was beginning to be interpreted as 'tertiary legislation, needed to make the secondary rules work'. This seems to be a prime example: the regulations are drafted far too widely, and we are dependent on the guidance to limit them. This is simply unacceptable: we should be taxed (or penalised) by legislation, not exempted by concession. But the failure of the intensive negotiations between the professional bodies and the Inland Revenue to produce workable legislation suggests that it is impossible to meet the concerns of both sides. Perhaps it is time to take a step back.
The Brightman Line
Why are we in this mess? Tax avoidance schemes have been around ever since taxes were first charged, but the number of schemes offered has multiplied dramatically in the past few years. In the time between the Budget and Royal Assent the Inland Revenue has announced changes to block six new schemes related to unapproved share schemes, which is why the regulations requiring disclosure of arrangements connected with employment have attracted far less criticism.
The whole point about the Brightman Line in Furniss v Dawson [1984] STC 153 is that it attacks artificial elements introduced into composite transactions purely for a tax advantage. If the intention is to get from A to D, but transactions B and C are introduced purely for a tax advantage, they can effectively be ignored. From the mid-1980s to the end of the century, this was an effective curb on the worst excesses of tax avoidance.
Everything changed with the House of Lords decision in MacNiven v Westmoreland [2001] STC 237. Lord Hoffmann drew a distinction between construing terms that have a commercial meaning, such as 'payment' or 'loss', and construing terms which have a purely legal or juristic meaning. In broad outline, he said that the Brightman Line was confined to the former, and would not normally apply to the latter. Since most tax avoidance schemes concern highly technical parts of tax legislation, it seems likely that they would come within the exclusion outlined by Lord Hoffmann. This must have been a significant factor in the increased tax avoidance activity since the case.
Right at the beginning of this article I questioned whether the demonising of the expression tax avoidance, as opposed to tax planning or tax mitigation, was anything more than a form of political correctness. I think that it is. Tax mitigation is advising a pensioner that one form of investment will result in a lower tax bill than another, or advising a sole trader that operating through a company will reduce his tax bill — real transactions with real consequences. When employees receive securities with special conditions that they do not understand, sign elections that make no sense, but do so because they are assured that the result will be the same amount of cash in their bank account with less tax to pay, that is avoidance. It is perfectly acceptable for the Government to make avoidance impracticable, provided that it does not at the same time attack ordinary tax planning.
Turning back the clock
How could the clock be turned back to before the Westmoreland decision? There are two possibilities. First, it is fair to say that the decision has come in for some criticism, most notably in Collector of Stamp Revenue v Arrowtown Assets Ltd FACV No 4 of 2003, where Lord Millett (sitting as a non-permanent judge in the Hong Kong courts) disagreed with the argument of Lord Hoffmann and refused to follow it as persuasive authority. He also commented that the Court of Appeal judges in the Barclays Mercantile case had found the decision hard to interpret.
Barclays Mercantile [2003] STC 66 comes to the House of Lords soon, probably before the end of this year. Lord Millett is no longer a Lord of Appeal in Ordinary, but is still eligible to hear appeals. Whether he sits on this appeal or not, the House of Lords, and Lord Hoffmann himself, may be able to give further guidance on the new direction that judicial interpretation of tax avoidance has taken.
If Barclays Mercantile confirms the distinction between juristic and commercial meanings, then it is open to the Government to decide that it wants to legislate — which was exactly the conclusion I came to in my previous article, twelve years ago. Then I was commenting on the uncertainty created by differing approaches to the Furniss v Dawson principle in subsequent cases. That, however, is nothing compared to the uncertainty created by the new regulations. Rather than coping with a badly-drawn paper-chase for disclosing tax-avoidance schemes, would it not be more appropriate to simply enact Lord Brightman's famous statement, together with a clarification that this applies regardless of whether the term being construed is commercial or juristic? We have had nearly 20 years of living with the Brightman Line, and on the whole it has been a positive force. Better the devil we know …
