Tax avoidance is now a headline-grabbing and career-threatening subject. But over seventy years ago the highest court in the land told us that taxpayers are entitled to do anything legal to reduce their tax bill. How we have got to here from there is a journey driven by strong personalities reacting to major economic events, changes in tax rates and what until very recently was an ever growing tax avoidance industry.
Tax avoidance is an impressionistic, indefinite and hugely controversial topic. So this article is just my own personal view of tax avoidance in the UK over the past 100 years. Despite what my grandchildren may think, I have been involved with tax for only about half of that period and, of course, my views are strongly coloured by what I have experienced while I have been involved. As for the earlier half of the 100 years, this is what I have come to understand from the case law and legislation which developed in the first half century, and the social and economic context in which they developed. To make this more focused, I have confined the topic to the main direct taxes.
Looking back over 100 years of tax avoidance is a bit like looking back on a 100 years of communications. It is hard for us oldsters to remember, and for youngsters even to imagine, what life was like without mobile phones and the internet. So it is with tax avoidance: who can remember or imagine what life was like when Panama was just some distant place where a canal had been cut between the Atlantic and Pacific Oceans, and not the detonator of explosive headlines about tax dodging revealed in the ‘Panama papers’?
I use the expression ‘tax dodging’ deliberately. Of course, the tax profession will always highlight the distinction between tax avoidance (legal, however unpopular) and tax evasion (illegal, however widespread). But to today’s wider public, ‘tax dodging’ sums up pretty accurately its disapproval of people or companies failing to pay tax which they ‘ought’ to pay. Jimmy Carr’s statement that ‘I pay what I have to, not a penny more’ may have resonated with tax purists, seeing his use of the ‘K2’ scheme as simply an example of legitimate tax avoidance; but it was met with derision by the media and the wider public, who saw him simply as failing to pay tax on his considerable earnings as a very successful comedian.
The reader might have expected this summary of 100 years of tax avoidance to begin at the beginning, in the middle of the First World War, and then move forward. But I prefer to look backwards from where we are, because I think that this makes it easier to see how, and whose, opinions have shaped the landscape of tax avoidance, and in turn to see what forces have shaped those opinions. So let me start with the most recent period.
The bankruptcy of Lehman Brothers
I am beginning with the collapse of Lehman Brothers in September 2008 because it wreaked destruction on the banking industry, made ‘banker’ a dirty word, and by association severely tarnished the reputation of the whole financial services sector.
Over the following years its longer term economic effects became ever more apparent, with an unprecedented decline in GDP, a substantial fall-off of tax revenues, a large increase in public debt, and in consequence the darkening reality of austerity across the population. ‘Occupy Wall Street’ spread to this side of the Atlantic, spawning ‘Occupied’ protestors in London who vented their anger at the ‘fat cats’ and at what they assumed to be the antisocial business and tax practices which helped build up the fat. Many more passively shared this view, and few were prepared to argue against it. Even fewer were interested in hearing about the distinction between ‘tax avoidance’ and ‘tax evasion’.
Public and political opinion
Against this background, on 19 June 2012 Alexi Mostrous wrote the first of his series of articles in The Times dealing with tax avoidance. The hapless target of this first front page headline was Jimmy Carr and his K2 tax scheme, and the story spread at lightning speed into the main news items on radio, TV and social media. A day later, prime minister David Cameron was describing the comedian’s tax avoidance as ‘morally wrong’. Within 12 months, David Cameron put tax compliance high on the agenda at the G8 which he chaired in Northern Ireland; and a few months later the G20 finance ministers called on the OECD to lead a comprehensive attack on international tax avoidance. The OECD’s response was the base erosion and profit shifting (BEPS) action plan, with its 15 separate measures to be put into effect within a very demanding timescale.
Here in the UK, Margaret Hodge MP used her position as chair of the House of Commons’ Public Accounts Committee (PAC) to make strident attacks on ‘the tax avoidance activities of multinational companies’, on HMRC for not acting more effectively against them, and, in her most poisonous barb, on ‘the big accounting firms…increasingly seen as being part of the problem of corporate tax avoidance, rather than the solution’ (the PAC’s 9th report, June 2013).
Legislative attack on avoidance
Depending on your viewpoint, Margaret Hodge’s and the PAC’s harsh criticism of HMRC was either excessive or wholly justified. In my opinion it was in fact both: it was excessive because it failed to do justice to the radical steps which HMRC was already taking to deter and counteract tax avoidance; but it also responded to and harnessed public intolerance of tax avoidance and so created a climate where HMRC could equip itself with the means needed to combat avoidance more effectively.
So, like a surf-boarder riding a giant wave off Hawaii, the Treasury moved to introduce a succession of anti-avoidance measures having the aim of counteracting and deterring tax avoidance strategies across a range of taxes. Prominent amongst these were three measures which took an unprecedented approach to tax avoidance:
The judges: litigation in tax avoidance cases
As a country we take justifiable pride in the standards and impartiality of our judges. Here, unlike the USA for example, our judges are not appointed by public election or subject to the approval of legislatures. But many of us who have worked in the courts for decades sense that prevailing public opinion does subtly influence the attitudes and approaches of judges – or at least where tax avoidance is concerned. So it is no coincidence that in this period HMRC has won an extraordinarily high proportion of cases challenging what it categorises as tax avoidance. Of the 38 reported cases of this sort heard in the Upper Tribunal or above between June 2012 and May 2016, only seven were won by the taxpayer. The judges’ willingness to decide so overwhelmingly in favour of HMRC has reflected the current political and public intolerance of tax avoidance.
Critically, the judicial tools which enabled them to do so had been developed over the previous three decades in a series of cases where the most influential judges agonised and argued, but eventually agreed, on the right approach to take to tax avoidance.
This mix of public opinion, tough legislation, and the new judicial approach to tax avoidance has drastically reduced the extent of abusive tax planning, and it has also dulled the appetite of many corporations and high net worth individuals to take advantage of more moderate forms of tax planning. Very few, if any, mainstream tax advisers would now recommend any sort of aggressive tax planning. So the decades of clever tax avoidance planning have, like Lehman Brothers itself, crashed to a halt.
Rossminster turns tax avoidance into a full-scale industry
I have taken these years together because it was during this period that the tax avoidance landscape was transformed beyond recognition. And I have begun it in 1973 because that was the year when Ron Plummer set up a company with one of the most notorious names in modern UK tax history – Rossminster. Before this he was at the financial conglomerate Slater Walker, where he collaborated with Roy Tucker, an ingenious deviser of tax avoidance schemes designed for high income earners. Very soon Rossminster’s main activity was working with Roy Tucker in developing and promoting these schemes; and these often complex arrangements were turned out like cars on a production line, to be known either as ‘Rossminster schemes’ or ‘Tucker schemes’, for high earners willing to pay a substantial slice of the intended tax saving for the ability to hold onto most of their earnings.
Before rushing to moral judgment, it is important to see the context at this time. Denis Healey, the shadow chancellor of the exchequer, had made it clear that if Labour came to power he intended to tax the rich heavily. At the Labour Party conference in the autumn of 1973, he warned ‘that there are going to be howls of anguish from those rich enough to pay over 75% on their last slice of earnings’. A few months later he promised that he would ‘squeeze property speculators until the pips squeak’. In 1974/75, the year when Denis Healey became chancellor in Harold Wilson’s second government, tax on the highest slice of income was raised from 75% to 83%, and with a continuing 15% surcharge for investment income. So high earning taxpayers could be left with only 2% of their investment income and only 17% of their highest slice of earned income (which started at £20,000 per year). With the benefit of hindsight, therefore, it is hardly surprising that many high earners of that era thought it could be more productive to call for help from Roy Tucker and Ron Plummer rather than howl with anguish or wait until their pips squeaked.
Margaret Thatcher and tax-rate climate change
On 4 May 1979, Margaret Thatcher entered 10 Downing Street as prime minister. In the first Budget under her premiership the top marginal rate of income tax dropped from 83% to 60%, and the 15% investment income surcharge was abolished in 1984. These substantial tax reductions ended the era when high income earners might expect some sympathy for resorting to tax avoidance in the hope of retaining a reasonable proportion of their earnings; but in the new climate of lower tax rates there was no place for such sympathy.
There was one other critical factor in the mix. Ever since the landmark decision of the House of Lords in the Duke of Westminster case [1936] AC 1 nearly 40 years earlier, it had become accepted in the tax profession, in the Inland Revenue and in the courts that tax statutes were to be interpreted literally and strictly, and that the tax provisions so construed had to be applied to transactions which were themselves to be analysed in a very legalistic way. Armed with these legal principles, the Rossminster schemes sought out and exploited loopholes in the tax statutes and created evermore elaborate and artificial transactions, some having no purpose at all except to create notional losses to reduce their clients’ taxable income.
I have highlighted the role played by Roy Tucker and Ron Plummer because their Rossminster scheme factory was probably the most prolific of the tax avoidance merchants. But the environment created by the punitive tax rates of the pre-Thatcher era was a fertile breeding ground for many other creators and purveyors of tailored tax avoidance schemes spanning the whole spectrum from subtle and sophisticated to brazen and reckless.
As with all good parties this heady mix of legalistic tax law and clever tax schemes risked leading to a hangover. And so it was: as the scale and artificiality of these schemes increased, powerful voices started to question, then to criticise and ultimately to overturn, the legal principles on which these schemes depended.
Several barons and a dead duke
Until the Rossminster era, the most active field of tax avoidance involved what was known as ‘dividend-stripping’. Briefly, the combination of the statutory tax rules dealing with trading losses (Income Tax Act 1952 s 341) and case law (for example, the House of Lords decision in JP Harrison (Watford) Ltd v Griffiths 40 TC 281 in March 1962) allowed traders in shares to exclude the receipt of dividends from those shares in computing the profits or losses of their trade for tax purposes. So, to take a very simple example, if a trader in shares bought a company for £100, extracted £40 worth of dividends from the company and then sold the company for £60, it would be treated as making a loss of £40 for tax purposes even though it had in fact made no loss at all. This simple, and to modern eyes, ridiculous, proposition was exploited by promoters of dividend-stripping schemes of varying degrees of sophistication. But some were so blatantly artificial that some judges felt able to uphold the Inland Revenue’s argument that the activities were not trading in shares at all but were instead elaborate tax avoidance arrangements.
One such case was Lupton v FA & AB Ltd 47 TC 580. This 1971 case is worth dwelling on for just a moment because it shows the conflicting currents which were starting to flow at that time. The majority of the Court of Appeal upheld the Inland Revenue’s arguments, deciding that the transactions in that particular case did not qualify as trading. In the words of the Master of the Rolls, Lord Denning (47 TC 580 at 605C): ‘I decline to elevate dividend-stripping into a trade. It is dividend-stripping and nothing else.’
But Sachs LJ dissented, taking the traditional line that taxpayers are entitled to rely on the strict language of the statute (47 TC 580 at 611E): ‘If the legislature chooses to leave a category of dividend-stripping untouched by the Finance Act, it does not seem to me that it is for the courts to interfere.’
A particularly noteworthy feature of this case was the identity and attitude of the QC leading the taxpayers’ team. This was Sydney Templeman QC. He relied on the established case law, and the record of the case shows him as saying (47 TC 580 at 614E):
‘The logic of that decision [FS Securities v CIR 41 TC 688] is that a trader in shares can for tax purposes show a loss when in fact he is not a penny the worse off, and the loss he shows is the amount of dividend he has received. He can therefore claim tax on that dividend under section 341 and he gets back the tax which has been paid, and rightly paid, by someone else.’
Those of us who encountered Sydney Templeman quickly became aware that he was a man with very strong views. Although he was appearing for the taxpayer in the FA & AB Ltd case, it was pretty clear from the quote that he was no lover of artificial tax avoidance schemes. So to those of us who knew him, it came as no surprise that as soon as he was appointed as a judge, he would take the lead in castigating such schemes and do his utmost to change the judicial ground rules which allowed them to flourish.
1975: Templeman J opens fire
His first salvo as a judge was fired in the case of Black Nominees Ltd v Nicol [1975] STC 372. This case dealt with an elaborate scheme involving three trusts, various loans and options, more than 20 inter-related documents and over 40 individuals and entities. The objective was to turn the considerable earnings of the actress Julie Christie into capital receipts for tax purposes, thus bearing a far smaller tax charge.
Templeman J tore into the scheme, using the language of ‘trickery’ and ‘performance’, which set the tone for his persistent efforts to change judicial attitudes ([1975] STC 372 at 416F):
‘The trick in the present case is that the £475,000 disappeared. It did not belong to anybody when the performance began, and it did not belong to anyone when the performance ended. It was invented by Miss Christie’s advisers for the purpose of performing a circle of payments...
‘If it were not for the trick with £475,000, no-one would suggest that the moneys received by the taxpayer company were capital. Once the trick is exposed the moneys are seen to be what they are, namely annual profits or gains.’
1979: Ramsay as Houdini
Four years later Sydney Templeman, now elevated to the Court of Appeal as Templeman LJ, used his vitriolic wit in upholding the Inland Revenue’s appeal in what has become one of the most famous tax cases of all time. The case was WT Ramsay Ltd v IRC, involving a scheme to shield a capital profit of over £172,000 from capital gains tax. The scheme involved the creation of two loans with very weird repayment terms. One of the loans was intended to qualify as ‘a debt on a security’, which as such would be subject to capital gains tax; and the other loan was fashioned to be a debt which was not ‘on a security’, and which would therefore be outside the scope of the tax. The objective was to create a loss on the ‘debt on a security’, which would be treated as an allowable loss for capital gains tax purposes, with a corresponding gain on the debt which was not ‘on a security’, and which would therefore be ignored for capital gains tax purposes. So the net tax result would be a capital loss.
Templeman LJ rounded on the scheme, using language which has become legendary ([1979] STC 582 at 583H–584B):
‘The facts ... demonstrate yet another circular game in which the taxpayer and a few hired performers act out the play; nothing happens save that the Houdini taxpayer appears to escape from the manacles of tax ... The object of the performance is to create the illusion that something has happened, that Hamlet has been killed and that Bottom did don an ass’s head so that tax advantages could be claimed as if something had happened…The critics are mistakenly informed that the play is based on a classic masterpiece called “The Duke of Westminster”.’
When the taxpayer’s appeal reached the House of Lords in March 1981, Lord Wilberforce delivered a judgment that confronted the two elements of the judicial approach which had hitherto underpinned tax avoidance schemes.
The first was the proposition that taxation has to be imposed by clear words. This had been generally taken to mean that tax statutes must be given a literal interpretation. Lord Wilberforce said that this was not correct ([1981] STC 174 at 179J):
‘A subject is only to be taxed upon clear words, not upon “intendment”, or upon the “equity” of an Act ... What are ‘clear words’ is to be ascertained upon normal principles: these do not confine the courts to literal interpretation. There may, indeed should, be considered the context and scheme of the relevant Act as a whole, and its purpose may, indeed should, be regarded.’
The second proposition was the belief that complex transactions had to be analysed, for tax purposes, on a step-by-step basis. Lord Wilberforce dispelled this ([1981] STC 174 at 181J):
‘While the techniques of tax avoidance progress are technically improved, the courts are not obliged to stand still ... to force the courts to adopt, in relation to closely integrated situations, a step-by-step, dissecting, approach which the parties themselves may have negated, would be a denial rather than an affirmation of the true judicial process ... The capital gains tax was created to operate in the real world, not that of make-belief.’
1981: The warnings
The Ramsay case involved a self-cancelling and wholly artificial arrangement that served no purpose whatsoever apart from creating a tax loss. Although Lord Wilberforce did not explicitly question the correctness of the Duke of Westminster decision, his judgment raised doubts as to its continuing relevance. A few months later, in IRC v Burmah Oil Co Ltd [1982] STC 30, Lord Diplock warned (at 32E):
‘It would be disingenuous to suggest, and dangerous on the part of those who advise on elaborate tax avoidance schemes to assume, that Ramsay’s case did not mark a significant change in the approach adopted by this House in its judicial role to a pre-ordained series of transactions (whether or not they include the achievement of a legitimate commercial end) into which there are inserted steps that have no commercial purpose apart from the avoidance of a liability to tax.’
Like in Ramsay, the taxpayer in Burmah Oil carried out a wholly artificial transaction involving a loan and a rights issue which had the sole purpose of creating a tax loss. This time the House of Lords had no difficulty in applying the Ramsay approach, deciding that, because there was no real loss, there was no loss in the sense contemplated by the tax legislation.
Lord Diplock’s warning, in the passage set out above, had explicitly referred to ‘inserted steps that had no commercial purpose’. His warning was echoed in the same case by Lord Scarman but, significantly, in more general terms ([1982] STC 30 at 39D):
‘... it is of the utmost importance that the business community (and others, including their advisers) should appreciate, as my noble and learned friend Lord Diplock has emphasised, that Ramsay’s case marks “a significant change in the approach adopted by this House in its judicial role” towards tax avoidance schemes.’
Was Lord Scarman presaging a broader attack, to include schemes which did serve some commercial purpose in addition to their tax avoidance objectives? No case had yet arisen to test the point, which would certainly be a game changer.
1984: Tax avoidance with some commercial purpose
That point came to be tested in a highly controversial case called Furniss v Dawson [1983] STC 549. In the broadest of terms, this involved the sale by the taxpayers of all the shares in two family companies. Carried out in a straightforward way this would have given rise to a substantial and immediate capital gains tax charge. To avoid the charge (technically, to defer it), the sale was carried out in two stages using an intermediary company. So there was a real commercial purpose in selling the companies to an outside purchaser; but there were additional steps designed to avoid (or defer) the capital gains tax.
In the Court of Appeal Kerr LJ gave a very short judgment, concluding that the ‘classic statement’ of Lord Tomlin in the Duke of Westminster case remained ‘fully authoritative for the purpose of the present case’ (at 566D), notwithstanding the new approach to composite transactions initiated by Ramsay. Slade LJ gave a longer judgment in which he (at 574E): ‘respectfully agree[d] with, and would wholly endorse, the observations of Kerr LJ in relation to the “classic statement” of Lord Tomlin in the Duke of Westminster case’.
When the case reached the House of Lords the judgment of the Court of Appeal was overturned unanimously, with each of the five law lords delivering a separate judgment. Their decision was to elide the two steps and treat them as constituting a disposal of the operating companies by the taxpayers to the outside purchaser, thus triggering an immediate capital gains tax charge.
In his short judgment Lord Roskill referred to the Duke of Westminster case, with the ominous comment (at 157C): ‘the ghost of the Duke of Westminster ... has haunted the administration of this branch of the law for too long. I confess that I had hoped that that ghost might have found quietude with the decisions in Ramsay and in Burmah. Unhappily it has not. Perhaps the decision of this House in these appeals will now suffice as exorcism’.
Furniss v Dawson, like Ramsay and Burmah, dealt with composite tax avoidance transactions which were ‘pre-ordained’ in the sense that each step of the transaction would inevitably follow the previous one. This was given as the justification for treating them as being, in effect, a single transaction. What it did not address was whether this Ramsay driven approach should apply more broadly to all pre-planned tax avoidance schemes.
1988: Open warfare between the judges
This was the main question at issue in the case of Craven v White [1988] STC 476, which was heard by the House of Lords in 1988 and which gave rise to some of the angriest comments recorded in the generally dull pages of tax reports. The anger, unsurprisingly, came from Lord Templeman who, following normal practice, had been shown the judgments of the other law lords in draft. Three of their lordships (therefore forming the majority) took the line that the Ramsay line of reasoning was confined to cases where it was inevitable that every step in a composite arrangement designed to avoid tax would in fact be carried out. Lord Oliver devoted 15 pages to defending this reasoning. To Lord Oliver the contrary argument involved (at 505C):
‘Substituting a determination to prevent the avoidance of tax for which there is no statutory, moral or logical basis for a rational, factual and intellectually possible appraisal of what is the reality of the position at the time when the relevant transaction is undertaken. I cannot, for my part, derive this from Dawson and I am quite sure that this House was not seeking to construct so irrational a doctrine.’
Lord Templeman must have been sorely provoked by this depiction of his approach as ‘illogical’ and ‘irrational’. After noting that he had read the other judgments he said that the majority view accepted (at 498C):
‘... the extreme argument of the taxpayer that Dawson is limited to its own facts or is limited to a transaction which has reached an advanced stage of negotiation ... These limitations would distort the effect of Dawson, are not based on principle, are not to be derived from the speeches in Dawson, and if followed would only revive a surprised tax avoidance industry and cost the general body of taxpayers hundreds of millions of pounds by enabling artificial tax avoidance schemes to alter the incidence of taxation ... In my opinion, a knife-edged majority has no power to limit this principle which has been responsible for four decisions of this House approved by a large number of our predecessors’.
2004: A short burial service
The acrimonious case of Craven v White was followed by two more House of Lords cases dealing with complex tax avoidance schemes (IRC v McGuckian [1997] STC 908, and MacNiven v Westmoreland Investments Ltd [2001] STC 237). The various law lords in these cases gave separate judgments explaining or elaborating in their own words what had been decided in the previous House of Lords cases starting with Ramsay. Taking all these cases together, there were now no less than 24 House of Lords separate judgments that needed to be considered in dealing with tax avoidance schemes. Apart from the obvious cases where the judges openly disagreed with each other, there were less obvious, but nonetheless very significant, differences in nuance and emphasis even when they were in broad agreement with each other.
Faced with this, in the next major case which came along to test the nature and the scope of the Ramsay principle, the House of Lords was asked explicitly to consider giving a single and clear judgment as to what the Ramsay doctrine was and where it applied.
That case was Barclays Mercantile Business Finance (BMBF) Ltd v Mawson [2005] STC 1 in November 2004. The five law lords accepted the invitation to deliver a unified judgment. Moreover, they also acceded to the request for a clear statement of principle by adopting the remarkably succinct statement given by a judge in the Hong Kong Court of Final Appeal. That judge was Riberio PJ, and his summary of the Ramsay principle was simply this (Collector of Stamp Revenue v Arrowtown Assets Ltd [2003] HKCFA 46 at [35], (2004) ITLR 454 at [35] quoted in [2005] STC 1 at 13C):
‘The driving principle of the Ramsay line of cases continues to involve a general rule of statutory construction and an unblinkered approach to the analysis of the facts. The ultimate question is whether the relevant statutory provisions, construed purposively, were intended to apply to the transaction, viewed realistically.’
Thus, in these two short sentences, the House of Lords buried hundreds of pages of analysis and, with it, the remains of the Duke of Westminster. All that the courts now needed to do was to look at the facts ‘realistically’ and apply the statutory tax law ‘purposively’. From now on each case could be decided by the judges using these simple but flexible and powerful tools of analysis. And armed with these tools the courts have been able to counteract the aggressive tax avoidance schemes which became prevalent in the 1990s and the 2000s.
Tax legislation in this era
Although we can now see how the struggle to change the judicial approach to tax avoidance was destined to end, that is of course with the benefit of hindsight. While this was going on, however, the outcome was not so clear. It will be recalled that Lord Templeman’s more radical approach to tax avoidance was rejected by the majority of the judges in Craven v White in 1988; and it was not until 2004, with the decision in BMBF v Mawson, that his approach was finally vindicated.
So it is not surprising that during this period of uncertainty the Inland Revenue hedged its bets by enacting very detailed tax rules, protected by what became known as targeted anti-avoidance rules (TAARs).
Of the many examples of TAARs, one is the ‘sales of occupation income’ legislation introduced in 1969. It soon came to be known as ‘the Beatles clause’, and counteracts individuals selling their earnings to some other person or entity, in order to convert them from being highly taxed income into lower taxed capital receipts – a device which was particularly suited to entertainers. Many of the Tucker and Plummer schemes resulted in TAARs, such as the advance interest scheme, which involved companies borrowing a large amount, paying a correspondingly large amount of tax deductible interest upfront and then getting a financial institution to take over the loan at a discount. This was outlawed in Finance Act 1976 (now in ITA 2007 s 773), following exposure of the scheme in the newspapers earlier that year, which disallowed interest where the sole or main benefit expected from the transaction was the obtaining of the relief.
The intention, of course, was that the detailed tax rules would leave potential tax avoiders with little room to manoeuvre, and that the TAARs would block whatever room was left. The reality, though, was that tax professionals bent on avoidance would seek out and exploit inconsistencies or errors of drafting in the detail of the rules; and so the utility of these tax avoidance ploys turned to futility only with the completion of the transformation of the judicial climate.
The Duke of Westminster case
In 1933, just over 70 years before the House of Lords in BMBF Ltd v Mawson laid down its simple test – apply the legislation purposively to the facts viewed realistically – the Special Commissioners did just that (Duke of Westminster v CIR 19 TC 490). They construed the ‘true effect and substance’ of the deeds of covenant under which payments were made to the duke’s gardeners and other employees. These deeds granted them annuities, but only on the basis that they would not expect also to receive their salaries (at 493):
‘... the payments made under these deeds to persons who remain in the [duke’s] employ were, in substance, payments for continuing service ejusdem generis with wages or salaries so long as the recipients in fact remain in the [duke’s] service and as such were not annual payments which were a proper deduction from his assessment to Sur-tax’.
So they rejected the duke’s attempt to reduce his tax bill by deducting the amount of these covenanted payments.
On appeal, Finlay J upheld the Special Commissioner’s decision, concluding that ‘the real nature of the transaction’ was to pay remuneration so long as the gardeners and others remained in their employment.
The Court of Appeal overturned Finlay J’s judgment, focusing on the legal nature of the deeds of covenant.
In the House of Lords (19 TC 490), there was a stark division of opinion. Lord Atkin, who was in a minority of one, looked at the whole arrangement and agreed with the Special Commissioners and Finlay J ‘that the substance of the transaction was that what was being paid was remuneration’ (at 517). This he took to be the legal effect of the documents, construed in the context of all the surrounding circumstances.
Lord Tomlin, though, vehemently disagreed with that approach, using language which became the tax avoiders’ charter for the next 70 years. He raised the question whether in tax cases (at 520): ‘There is a doctrine that the court may ignore the legal position and regard what is called “the substance of the matter”.’
His response was: ‘This supposed doctrine ... seems to rest for its support upon a misunderstanding of language used in some earlier cases. The sooner this misunderstanding is dispelled and the supposed doctrine given its quietus the better it will be for all concerned ... Every man is entitled if he can to order his affairs so that the tax attaching under the appropriate Acts is less than it otherwise would be. If he succeeds in ordering them so as to secure this result, then, however unappreciative the Commissioners of the Inland Revenue or his fellow taxpayers may be of his ingenuity, he cannot be compelled to pay an increased tax. This so-called doctrine of “the substance” seems to me to be nothing more than an attempt to make a man pay notwithstanding that he has so ordered his affairs that the amount of tax sought from his is not legally claimable.’
Lord Russell said very much the same thing about ‘the substance’ of the transaction, but added the additional stricture that the tax had to be imposed by the plain words of the statute (at 524):
‘I confess that I view with disfavour the doctrine that in taxation cases the subject is to be taxed if, in accordance with a court’s view of what it considers the substance of the transaction, the court thinks that the case falls within the contemplation or spirit of the statute. The subject is not taxable by inference or by analogy, but only by the plain words of the statute applicable to the facts and circumstances of this case.’
Lord Macmillan echoed this by adding that (at 727): ‘Your lordships are concerned only with the technical question whether the [duke] has brought himself within the language of the income tax rule as to contractual payments, and I think that he has succeeded in doing so.’
Looking back on this over 60 years later in McGuckian [1997] STC 908, Lord Steyn commented (at 915G) that:
‘In combination these two features – literal interpretation of tax statutes and the formalistic insistence on examining steps in a composite scheme separately – allowed tax avoidance schemes to flourish to the detriment of the general body of taxpayers. The result was that the court appeared to be relegated to the role of a spectator concentrating on the individual moves in a highly skilled game ... The courts became habituated to this narrow view of their role.’
If anything, Lord Steyn may have understated the impact of the Duke of Westminster case. More than just ‘allowing tax avoidance schemes to flourish’, Lord Tomlin’s comments could be taken to encourage any form of tax avoidance so long as it was legal.
On this basis, the Inland Revenue would have to introduce specific legislation to block particular types of tax avoidance as they came to its attention.
Sticky anti-avoidance rules
Given the habit of tax advisers to stare at any new piece of legislation looking for gaps, the Revenue used deliberately imprecise language to address avoidance so that advisers could not see exactly what it was meant to catch.
One of the best examples is the first major piece of anti-avoidance legislation after Duke of Westminster. The 1936 Finance Act (s 18; and still on the statute books as ITA 2007 s 721) attacked the avoidance of tax by the ‘transfer of assets’ abroad where the UK taxpayer still had ‘power to enjoy’ the income produced by those assets. The section made full use of the passive tense that rights are acquired ‘by means of’ transfers of assets, and introduced the concept of ‘associated operations’. So wide is the wording that it also introduced the now familiar let-off of a motive test – that it would not apply if the taxpayer could satisfy the Special Commissioners that the transfers ‘were effected mainly for some purpose other than the purpose of avoiding a liability to taxation’.
Needless to say, despite the wide and sticky wording of the section it did not take long for a taxpayer to have a go at side-stepping it. In the case of Congreve and Congreve v CIR 30 TC 163, the cunning plan was to make a transfer to a UK-resident company which promptly became non-resident. This did not make it past the House of Lords in 1948, who agreed with the Revenue that the section did not say that the transferee had to be non-resident at the time of transfer, and that the transfer did not need to be made by the taxpayer directly, and that in any event the whole lot of events were clearly ‘associated operations’; so the sticky language had caught its prey.
Statutory motive tests developed further: the dividend stripping avoidance legislation when introduced in Finance Act 1960 (s 28; now ITA 2007 s 682) had belt and braces, requiring not only the lack of a tax avoidance motive but a positive requirement that the transaction was ‘for bona fide commercial reasons or in the ordinary course of making and managing investments’ on top. But targeted anti-avoidance still suffered from the intrinsic weakness that it had to be attached to particular sections. So as soon as dividend stripping was plugged, Messrs Tucker and Plummer revised their planning with a structure using covenants to charity rather than dividends.
The dawn of high tax rates and the game of tax avoidance
This final section starts in the middle of the First World War. The war marked the beginning of sustained higher rates of taxation. The standard rate of income tax went from 5.8% at the beginning of the war to 30% by the end, with surtax increasing from 2.5% to a top rate of 22.5%. These increased rates meant that the balance of tax also shifted from indirect to direct tax, the indirect/direct tax ratio being 48/58 in 1912 and 18/82 by 1918 (‘Settlements and the avoidance of tax on income – the period to 1920’ (David Stopforth) BTR 1990 and ‘Just Taxes’ (Martin Daunton) CUP, p 46). Channelling the increases into direct tax in this way meant, no doubt intentionally, that more of the total increase was paid by the better off.
Unsurprisingly this led to some wealthier citizens looking at their tax bills more closely. Covenants to children, which were deductible from the parent’s taxable income, became popular. In turn, this led the Revenue to look more closely at covenants. The first major piece of anti-avoidance legislation followed, in Finance Act 1922 s 20. A covenant was to be ineffective for tax in three situations: if it was for six years or less (hence the Duke of Westminster’s seven years); if it was revocable without the consent of another person (other than the covenantor’s spouse); or if it was to the covenantor’s child for less than the life of the child.
The legislation spawned more cases than just Duke of Westminster. Finlay J, who found for the Revenue in Duke of Westminster, had already decided for the Revenue on similar reasoning in CIR v Clarkson-Webb 17 TC 451 in 1932, where two brothers attempted to circumvent the need for covenants to their own children to be for life by entering into admittedly reciprocal covenants for each other’s children. Finlay J said that (at 457): ‘one has ... to look nonetheless at the substance of the matter ... they were part of an arrangement which must be looked at as a whole’.
Even he, however, had to admit defeat on the wording on one scheme which was doing the rounds. Because the section specified that revocable covenants would still be effective if the consent of another was required, a practice developed of specifying that consent of someone – often the taxpayer’s accountant, or any one of several friends – was required, in the obvious expectation that the accountant would not want to lose his client and that if one friend proved difficult, another might not. The 1933 case of CIR v Firth 17 TC 603 took this to extremes, with a provision that the covenant could be revoked with the consent of the taxpayer’s accountant or some other person named by the taxpayer, other than his wife. Finlay J conceded: ‘the gentleman in question might go first to one person and then another until he got somebody to consent. All I can say is that I think is the result of the section’.
The third situation, covenants to a child for less than the life of the child, was also the subject of litigation which went all the way to the House of Lords, in Wiggins v Watson’s Trustees 17 TC 728 also in 1933: a covenant to the taxpayer’s child revocable with the consent of one of several friends was again found to be effective for tax. So, the first piece of clear and simple avoidance legislation resulted in Revenue losses in all three situations where taxpayers took avoidance steps.
The century of avoidance, followed by avoidance legislation and litigation, had begun.
The author thanks Catherine Gosh for her assistance with this article. This article forms a chapter of the book ‘Plucking the goose: a century of taxation from the Great War to the digital age’ (Tolley, September 2016), which is available to purchase for £9.99 here. Tolley is donating all profits from book sales to the ‘Bridge the gap’ charitable campaign, supported by the CIOT.
Tax avoidance is now a headline-grabbing and career-threatening subject. But over seventy years ago the highest court in the land told us that taxpayers are entitled to do anything legal to reduce their tax bill. How we have got to here from there is a journey driven by strong personalities reacting to major economic events, changes in tax rates and what until very recently was an ever growing tax avoidance industry.
Tax avoidance is an impressionistic, indefinite and hugely controversial topic. So this article is just my own personal view of tax avoidance in the UK over the past 100 years. Despite what my grandchildren may think, I have been involved with tax for only about half of that period and, of course, my views are strongly coloured by what I have experienced while I have been involved. As for the earlier half of the 100 years, this is what I have come to understand from the case law and legislation which developed in the first half century, and the social and economic context in which they developed. To make this more focused, I have confined the topic to the main direct taxes.
Looking back over 100 years of tax avoidance is a bit like looking back on a 100 years of communications. It is hard for us oldsters to remember, and for youngsters even to imagine, what life was like without mobile phones and the internet. So it is with tax avoidance: who can remember or imagine what life was like when Panama was just some distant place where a canal had been cut between the Atlantic and Pacific Oceans, and not the detonator of explosive headlines about tax dodging revealed in the ‘Panama papers’?
I use the expression ‘tax dodging’ deliberately. Of course, the tax profession will always highlight the distinction between tax avoidance (legal, however unpopular) and tax evasion (illegal, however widespread). But to today’s wider public, ‘tax dodging’ sums up pretty accurately its disapproval of people or companies failing to pay tax which they ‘ought’ to pay. Jimmy Carr’s statement that ‘I pay what I have to, not a penny more’ may have resonated with tax purists, seeing his use of the ‘K2’ scheme as simply an example of legitimate tax avoidance; but it was met with derision by the media and the wider public, who saw him simply as failing to pay tax on his considerable earnings as a very successful comedian.
The reader might have expected this summary of 100 years of tax avoidance to begin at the beginning, in the middle of the First World War, and then move forward. But I prefer to look backwards from where we are, because I think that this makes it easier to see how, and whose, opinions have shaped the landscape of tax avoidance, and in turn to see what forces have shaped those opinions. So let me start with the most recent period.
The bankruptcy of Lehman Brothers
I am beginning with the collapse of Lehman Brothers in September 2008 because it wreaked destruction on the banking industry, made ‘banker’ a dirty word, and by association severely tarnished the reputation of the whole financial services sector.
Over the following years its longer term economic effects became ever more apparent, with an unprecedented decline in GDP, a substantial fall-off of tax revenues, a large increase in public debt, and in consequence the darkening reality of austerity across the population. ‘Occupy Wall Street’ spread to this side of the Atlantic, spawning ‘Occupied’ protestors in London who vented their anger at the ‘fat cats’ and at what they assumed to be the antisocial business and tax practices which helped build up the fat. Many more passively shared this view, and few were prepared to argue against it. Even fewer were interested in hearing about the distinction between ‘tax avoidance’ and ‘tax evasion’.
Public and political opinion
Against this background, on 19 June 2012 Alexi Mostrous wrote the first of his series of articles in The Times dealing with tax avoidance. The hapless target of this first front page headline was Jimmy Carr and his K2 tax scheme, and the story spread at lightning speed into the main news items on radio, TV and social media. A day later, prime minister David Cameron was describing the comedian’s tax avoidance as ‘morally wrong’. Within 12 months, David Cameron put tax compliance high on the agenda at the G8 which he chaired in Northern Ireland; and a few months later the G20 finance ministers called on the OECD to lead a comprehensive attack on international tax avoidance. The OECD’s response was the base erosion and profit shifting (BEPS) action plan, with its 15 separate measures to be put into effect within a very demanding timescale.
Here in the UK, Margaret Hodge MP used her position as chair of the House of Commons’ Public Accounts Committee (PAC) to make strident attacks on ‘the tax avoidance activities of multinational companies’, on HMRC for not acting more effectively against them, and, in her most poisonous barb, on ‘the big accounting firms…increasingly seen as being part of the problem of corporate tax avoidance, rather than the solution’ (the PAC’s 9th report, June 2013).
Legislative attack on avoidance
Depending on your viewpoint, Margaret Hodge’s and the PAC’s harsh criticism of HMRC was either excessive or wholly justified. In my opinion it was in fact both: it was excessive because it failed to do justice to the radical steps which HMRC was already taking to deter and counteract tax avoidance; but it also responded to and harnessed public intolerance of tax avoidance and so created a climate where HMRC could equip itself with the means needed to combat avoidance more effectively.
So, like a surf-boarder riding a giant wave off Hawaii, the Treasury moved to introduce a succession of anti-avoidance measures having the aim of counteracting and deterring tax avoidance strategies across a range of taxes. Prominent amongst these were three measures which took an unprecedented approach to tax avoidance:
The judges: litigation in tax avoidance cases
As a country we take justifiable pride in the standards and impartiality of our judges. Here, unlike the USA for example, our judges are not appointed by public election or subject to the approval of legislatures. But many of us who have worked in the courts for decades sense that prevailing public opinion does subtly influence the attitudes and approaches of judges – or at least where tax avoidance is concerned. So it is no coincidence that in this period HMRC has won an extraordinarily high proportion of cases challenging what it categorises as tax avoidance. Of the 38 reported cases of this sort heard in the Upper Tribunal or above between June 2012 and May 2016, only seven were won by the taxpayer. The judges’ willingness to decide so overwhelmingly in favour of HMRC has reflected the current political and public intolerance of tax avoidance.
Critically, the judicial tools which enabled them to do so had been developed over the previous three decades in a series of cases where the most influential judges agonised and argued, but eventually agreed, on the right approach to take to tax avoidance.
This mix of public opinion, tough legislation, and the new judicial approach to tax avoidance has drastically reduced the extent of abusive tax planning, and it has also dulled the appetite of many corporations and high net worth individuals to take advantage of more moderate forms of tax planning. Very few, if any, mainstream tax advisers would now recommend any sort of aggressive tax planning. So the decades of clever tax avoidance planning have, like Lehman Brothers itself, crashed to a halt.
Rossminster turns tax avoidance into a full-scale industry
I have taken these years together because it was during this period that the tax avoidance landscape was transformed beyond recognition. And I have begun it in 1973 because that was the year when Ron Plummer set up a company with one of the most notorious names in modern UK tax history – Rossminster. Before this he was at the financial conglomerate Slater Walker, where he collaborated with Roy Tucker, an ingenious deviser of tax avoidance schemes designed for high income earners. Very soon Rossminster’s main activity was working with Roy Tucker in developing and promoting these schemes; and these often complex arrangements were turned out like cars on a production line, to be known either as ‘Rossminster schemes’ or ‘Tucker schemes’, for high earners willing to pay a substantial slice of the intended tax saving for the ability to hold onto most of their earnings.
Before rushing to moral judgment, it is important to see the context at this time. Denis Healey, the shadow chancellor of the exchequer, had made it clear that if Labour came to power he intended to tax the rich heavily. At the Labour Party conference in the autumn of 1973, he warned ‘that there are going to be howls of anguish from those rich enough to pay over 75% on their last slice of earnings’. A few months later he promised that he would ‘squeeze property speculators until the pips squeak’. In 1974/75, the year when Denis Healey became chancellor in Harold Wilson’s second government, tax on the highest slice of income was raised from 75% to 83%, and with a continuing 15% surcharge for investment income. So high earning taxpayers could be left with only 2% of their investment income and only 17% of their highest slice of earned income (which started at £20,000 per year). With the benefit of hindsight, therefore, it is hardly surprising that many high earners of that era thought it could be more productive to call for help from Roy Tucker and Ron Plummer rather than howl with anguish or wait until their pips squeaked.
Margaret Thatcher and tax-rate climate change
On 4 May 1979, Margaret Thatcher entered 10 Downing Street as prime minister. In the first Budget under her premiership the top marginal rate of income tax dropped from 83% to 60%, and the 15% investment income surcharge was abolished in 1984. These substantial tax reductions ended the era when high income earners might expect some sympathy for resorting to tax avoidance in the hope of retaining a reasonable proportion of their earnings; but in the new climate of lower tax rates there was no place for such sympathy.
There was one other critical factor in the mix. Ever since the landmark decision of the House of Lords in the Duke of Westminster case [1936] AC 1 nearly 40 years earlier, it had become accepted in the tax profession, in the Inland Revenue and in the courts that tax statutes were to be interpreted literally and strictly, and that the tax provisions so construed had to be applied to transactions which were themselves to be analysed in a very legalistic way. Armed with these legal principles, the Rossminster schemes sought out and exploited loopholes in the tax statutes and created evermore elaborate and artificial transactions, some having no purpose at all except to create notional losses to reduce their clients’ taxable income.
I have highlighted the role played by Roy Tucker and Ron Plummer because their Rossminster scheme factory was probably the most prolific of the tax avoidance merchants. But the environment created by the punitive tax rates of the pre-Thatcher era was a fertile breeding ground for many other creators and purveyors of tailored tax avoidance schemes spanning the whole spectrum from subtle and sophisticated to brazen and reckless.
As with all good parties this heady mix of legalistic tax law and clever tax schemes risked leading to a hangover. And so it was: as the scale and artificiality of these schemes increased, powerful voices started to question, then to criticise and ultimately to overturn, the legal principles on which these schemes depended.
Several barons and a dead duke
Until the Rossminster era, the most active field of tax avoidance involved what was known as ‘dividend-stripping’. Briefly, the combination of the statutory tax rules dealing with trading losses (Income Tax Act 1952 s 341) and case law (for example, the House of Lords decision in JP Harrison (Watford) Ltd v Griffiths 40 TC 281 in March 1962) allowed traders in shares to exclude the receipt of dividends from those shares in computing the profits or losses of their trade for tax purposes. So, to take a very simple example, if a trader in shares bought a company for £100, extracted £40 worth of dividends from the company and then sold the company for £60, it would be treated as making a loss of £40 for tax purposes even though it had in fact made no loss at all. This simple, and to modern eyes, ridiculous, proposition was exploited by promoters of dividend-stripping schemes of varying degrees of sophistication. But some were so blatantly artificial that some judges felt able to uphold the Inland Revenue’s argument that the activities were not trading in shares at all but were instead elaborate tax avoidance arrangements.
One such case was Lupton v FA & AB Ltd 47 TC 580. This 1971 case is worth dwelling on for just a moment because it shows the conflicting currents which were starting to flow at that time. The majority of the Court of Appeal upheld the Inland Revenue’s arguments, deciding that the transactions in that particular case did not qualify as trading. In the words of the Master of the Rolls, Lord Denning (47 TC 580 at 605C): ‘I decline to elevate dividend-stripping into a trade. It is dividend-stripping and nothing else.’
But Sachs LJ dissented, taking the traditional line that taxpayers are entitled to rely on the strict language of the statute (47 TC 580 at 611E): ‘If the legislature chooses to leave a category of dividend-stripping untouched by the Finance Act, it does not seem to me that it is for the courts to interfere.’
A particularly noteworthy feature of this case was the identity and attitude of the QC leading the taxpayers’ team. This was Sydney Templeman QC. He relied on the established case law, and the record of the case shows him as saying (47 TC 580 at 614E):
‘The logic of that decision [FS Securities v CIR 41 TC 688] is that a trader in shares can for tax purposes show a loss when in fact he is not a penny the worse off, and the loss he shows is the amount of dividend he has received. He can therefore claim tax on that dividend under section 341 and he gets back the tax which has been paid, and rightly paid, by someone else.’
Those of us who encountered Sydney Templeman quickly became aware that he was a man with very strong views. Although he was appearing for the taxpayer in the FA & AB Ltd case, it was pretty clear from the quote that he was no lover of artificial tax avoidance schemes. So to those of us who knew him, it came as no surprise that as soon as he was appointed as a judge, he would take the lead in castigating such schemes and do his utmost to change the judicial ground rules which allowed them to flourish.
1975: Templeman J opens fire
His first salvo as a judge was fired in the case of Black Nominees Ltd v Nicol [1975] STC 372. This case dealt with an elaborate scheme involving three trusts, various loans and options, more than 20 inter-related documents and over 40 individuals and entities. The objective was to turn the considerable earnings of the actress Julie Christie into capital receipts for tax purposes, thus bearing a far smaller tax charge.
Templeman J tore into the scheme, using the language of ‘trickery’ and ‘performance’, which set the tone for his persistent efforts to change judicial attitudes ([1975] STC 372 at 416F):
‘The trick in the present case is that the £475,000 disappeared. It did not belong to anybody when the performance began, and it did not belong to anyone when the performance ended. It was invented by Miss Christie’s advisers for the purpose of performing a circle of payments...
‘If it were not for the trick with £475,000, no-one would suggest that the moneys received by the taxpayer company were capital. Once the trick is exposed the moneys are seen to be what they are, namely annual profits or gains.’
1979: Ramsay as Houdini
Four years later Sydney Templeman, now elevated to the Court of Appeal as Templeman LJ, used his vitriolic wit in upholding the Inland Revenue’s appeal in what has become one of the most famous tax cases of all time. The case was WT Ramsay Ltd v IRC, involving a scheme to shield a capital profit of over £172,000 from capital gains tax. The scheme involved the creation of two loans with very weird repayment terms. One of the loans was intended to qualify as ‘a debt on a security’, which as such would be subject to capital gains tax; and the other loan was fashioned to be a debt which was not ‘on a security’, and which would therefore be outside the scope of the tax. The objective was to create a loss on the ‘debt on a security’, which would be treated as an allowable loss for capital gains tax purposes, with a corresponding gain on the debt which was not ‘on a security’, and which would therefore be ignored for capital gains tax purposes. So the net tax result would be a capital loss.
Templeman LJ rounded on the scheme, using language which has become legendary ([1979] STC 582 at 583H–584B):
‘The facts ... demonstrate yet another circular game in which the taxpayer and a few hired performers act out the play; nothing happens save that the Houdini taxpayer appears to escape from the manacles of tax ... The object of the performance is to create the illusion that something has happened, that Hamlet has been killed and that Bottom did don an ass’s head so that tax advantages could be claimed as if something had happened…The critics are mistakenly informed that the play is based on a classic masterpiece called “The Duke of Westminster”.’
When the taxpayer’s appeal reached the House of Lords in March 1981, Lord Wilberforce delivered a judgment that confronted the two elements of the judicial approach which had hitherto underpinned tax avoidance schemes.
The first was the proposition that taxation has to be imposed by clear words. This had been generally taken to mean that tax statutes must be given a literal interpretation. Lord Wilberforce said that this was not correct ([1981] STC 174 at 179J):
‘A subject is only to be taxed upon clear words, not upon “intendment”, or upon the “equity” of an Act ... What are ‘clear words’ is to be ascertained upon normal principles: these do not confine the courts to literal interpretation. There may, indeed should, be considered the context and scheme of the relevant Act as a whole, and its purpose may, indeed should, be regarded.’
The second proposition was the belief that complex transactions had to be analysed, for tax purposes, on a step-by-step basis. Lord Wilberforce dispelled this ([1981] STC 174 at 181J):
‘While the techniques of tax avoidance progress are technically improved, the courts are not obliged to stand still ... to force the courts to adopt, in relation to closely integrated situations, a step-by-step, dissecting, approach which the parties themselves may have negated, would be a denial rather than an affirmation of the true judicial process ... The capital gains tax was created to operate in the real world, not that of make-belief.’
1981: The warnings
The Ramsay case involved a self-cancelling and wholly artificial arrangement that served no purpose whatsoever apart from creating a tax loss. Although Lord Wilberforce did not explicitly question the correctness of the Duke of Westminster decision, his judgment raised doubts as to its continuing relevance. A few months later, in IRC v Burmah Oil Co Ltd [1982] STC 30, Lord Diplock warned (at 32E):
‘It would be disingenuous to suggest, and dangerous on the part of those who advise on elaborate tax avoidance schemes to assume, that Ramsay’s case did not mark a significant change in the approach adopted by this House in its judicial role to a pre-ordained series of transactions (whether or not they include the achievement of a legitimate commercial end) into which there are inserted steps that have no commercial purpose apart from the avoidance of a liability to tax.’
Like in Ramsay, the taxpayer in Burmah Oil carried out a wholly artificial transaction involving a loan and a rights issue which had the sole purpose of creating a tax loss. This time the House of Lords had no difficulty in applying the Ramsay approach, deciding that, because there was no real loss, there was no loss in the sense contemplated by the tax legislation.
Lord Diplock’s warning, in the passage set out above, had explicitly referred to ‘inserted steps that had no commercial purpose’. His warning was echoed in the same case by Lord Scarman but, significantly, in more general terms ([1982] STC 30 at 39D):
‘... it is of the utmost importance that the business community (and others, including their advisers) should appreciate, as my noble and learned friend Lord Diplock has emphasised, that Ramsay’s case marks “a significant change in the approach adopted by this House in its judicial role” towards tax avoidance schemes.’
Was Lord Scarman presaging a broader attack, to include schemes which did serve some commercial purpose in addition to their tax avoidance objectives? No case had yet arisen to test the point, which would certainly be a game changer.
1984: Tax avoidance with some commercial purpose
That point came to be tested in a highly controversial case called Furniss v Dawson [1983] STC 549. In the broadest of terms, this involved the sale by the taxpayers of all the shares in two family companies. Carried out in a straightforward way this would have given rise to a substantial and immediate capital gains tax charge. To avoid the charge (technically, to defer it), the sale was carried out in two stages using an intermediary company. So there was a real commercial purpose in selling the companies to an outside purchaser; but there were additional steps designed to avoid (or defer) the capital gains tax.
In the Court of Appeal Kerr LJ gave a very short judgment, concluding that the ‘classic statement’ of Lord Tomlin in the Duke of Westminster case remained ‘fully authoritative for the purpose of the present case’ (at 566D), notwithstanding the new approach to composite transactions initiated by Ramsay. Slade LJ gave a longer judgment in which he (at 574E): ‘respectfully agree[d] with, and would wholly endorse, the observations of Kerr LJ in relation to the “classic statement” of Lord Tomlin in the Duke of Westminster case’.
When the case reached the House of Lords the judgment of the Court of Appeal was overturned unanimously, with each of the five law lords delivering a separate judgment. Their decision was to elide the two steps and treat them as constituting a disposal of the operating companies by the taxpayers to the outside purchaser, thus triggering an immediate capital gains tax charge.
In his short judgment Lord Roskill referred to the Duke of Westminster case, with the ominous comment (at 157C): ‘the ghost of the Duke of Westminster ... has haunted the administration of this branch of the law for too long. I confess that I had hoped that that ghost might have found quietude with the decisions in Ramsay and in Burmah. Unhappily it has not. Perhaps the decision of this House in these appeals will now suffice as exorcism’.
Furniss v Dawson, like Ramsay and Burmah, dealt with composite tax avoidance transactions which were ‘pre-ordained’ in the sense that each step of the transaction would inevitably follow the previous one. This was given as the justification for treating them as being, in effect, a single transaction. What it did not address was whether this Ramsay driven approach should apply more broadly to all pre-planned tax avoidance schemes.
1988: Open warfare between the judges
This was the main question at issue in the case of Craven v White [1988] STC 476, which was heard by the House of Lords in 1988 and which gave rise to some of the angriest comments recorded in the generally dull pages of tax reports. The anger, unsurprisingly, came from Lord Templeman who, following normal practice, had been shown the judgments of the other law lords in draft. Three of their lordships (therefore forming the majority) took the line that the Ramsay line of reasoning was confined to cases where it was inevitable that every step in a composite arrangement designed to avoid tax would in fact be carried out. Lord Oliver devoted 15 pages to defending this reasoning. To Lord Oliver the contrary argument involved (at 505C):
‘Substituting a determination to prevent the avoidance of tax for which there is no statutory, moral or logical basis for a rational, factual and intellectually possible appraisal of what is the reality of the position at the time when the relevant transaction is undertaken. I cannot, for my part, derive this from Dawson and I am quite sure that this House was not seeking to construct so irrational a doctrine.’
Lord Templeman must have been sorely provoked by this depiction of his approach as ‘illogical’ and ‘irrational’. After noting that he had read the other judgments he said that the majority view accepted (at 498C):
‘... the extreme argument of the taxpayer that Dawson is limited to its own facts or is limited to a transaction which has reached an advanced stage of negotiation ... These limitations would distort the effect of Dawson, are not based on principle, are not to be derived from the speeches in Dawson, and if followed would only revive a surprised tax avoidance industry and cost the general body of taxpayers hundreds of millions of pounds by enabling artificial tax avoidance schemes to alter the incidence of taxation ... In my opinion, a knife-edged majority has no power to limit this principle which has been responsible for four decisions of this House approved by a large number of our predecessors’.
2004: A short burial service
The acrimonious case of Craven v White was followed by two more House of Lords cases dealing with complex tax avoidance schemes (IRC v McGuckian [1997] STC 908, and MacNiven v Westmoreland Investments Ltd [2001] STC 237). The various law lords in these cases gave separate judgments explaining or elaborating in their own words what had been decided in the previous House of Lords cases starting with Ramsay. Taking all these cases together, there were now no less than 24 House of Lords separate judgments that needed to be considered in dealing with tax avoidance schemes. Apart from the obvious cases where the judges openly disagreed with each other, there were less obvious, but nonetheless very significant, differences in nuance and emphasis even when they were in broad agreement with each other.
Faced with this, in the next major case which came along to test the nature and the scope of the Ramsay principle, the House of Lords was asked explicitly to consider giving a single and clear judgment as to what the Ramsay doctrine was and where it applied.
That case was Barclays Mercantile Business Finance (BMBF) Ltd v Mawson [2005] STC 1 in November 2004. The five law lords accepted the invitation to deliver a unified judgment. Moreover, they also acceded to the request for a clear statement of principle by adopting the remarkably succinct statement given by a judge in the Hong Kong Court of Final Appeal. That judge was Riberio PJ, and his summary of the Ramsay principle was simply this (Collector of Stamp Revenue v Arrowtown Assets Ltd [2003] HKCFA 46 at [35], (2004) ITLR 454 at [35] quoted in [2005] STC 1 at 13C):
‘The driving principle of the Ramsay line of cases continues to involve a general rule of statutory construction and an unblinkered approach to the analysis of the facts. The ultimate question is whether the relevant statutory provisions, construed purposively, were intended to apply to the transaction, viewed realistically.’
Thus, in these two short sentences, the House of Lords buried hundreds of pages of analysis and, with it, the remains of the Duke of Westminster. All that the courts now needed to do was to look at the facts ‘realistically’ and apply the statutory tax law ‘purposively’. From now on each case could be decided by the judges using these simple but flexible and powerful tools of analysis. And armed with these tools the courts have been able to counteract the aggressive tax avoidance schemes which became prevalent in the 1990s and the 2000s.
Tax legislation in this era
Although we can now see how the struggle to change the judicial approach to tax avoidance was destined to end, that is of course with the benefit of hindsight. While this was going on, however, the outcome was not so clear. It will be recalled that Lord Templeman’s more radical approach to tax avoidance was rejected by the majority of the judges in Craven v White in 1988; and it was not until 2004, with the decision in BMBF v Mawson, that his approach was finally vindicated.
So it is not surprising that during this period of uncertainty the Inland Revenue hedged its bets by enacting very detailed tax rules, protected by what became known as targeted anti-avoidance rules (TAARs).
Of the many examples of TAARs, one is the ‘sales of occupation income’ legislation introduced in 1969. It soon came to be known as ‘the Beatles clause’, and counteracts individuals selling their earnings to some other person or entity, in order to convert them from being highly taxed income into lower taxed capital receipts – a device which was particularly suited to entertainers. Many of the Tucker and Plummer schemes resulted in TAARs, such as the advance interest scheme, which involved companies borrowing a large amount, paying a correspondingly large amount of tax deductible interest upfront and then getting a financial institution to take over the loan at a discount. This was outlawed in Finance Act 1976 (now in ITA 2007 s 773), following exposure of the scheme in the newspapers earlier that year, which disallowed interest where the sole or main benefit expected from the transaction was the obtaining of the relief.
The intention, of course, was that the detailed tax rules would leave potential tax avoiders with little room to manoeuvre, and that the TAARs would block whatever room was left. The reality, though, was that tax professionals bent on avoidance would seek out and exploit inconsistencies or errors of drafting in the detail of the rules; and so the utility of these tax avoidance ploys turned to futility only with the completion of the transformation of the judicial climate.
The Duke of Westminster case
In 1933, just over 70 years before the House of Lords in BMBF Ltd v Mawson laid down its simple test – apply the legislation purposively to the facts viewed realistically – the Special Commissioners did just that (Duke of Westminster v CIR 19 TC 490). They construed the ‘true effect and substance’ of the deeds of covenant under which payments were made to the duke’s gardeners and other employees. These deeds granted them annuities, but only on the basis that they would not expect also to receive their salaries (at 493):
‘... the payments made under these deeds to persons who remain in the [duke’s] employ were, in substance, payments for continuing service ejusdem generis with wages or salaries so long as the recipients in fact remain in the [duke’s] service and as such were not annual payments which were a proper deduction from his assessment to Sur-tax’.
So they rejected the duke’s attempt to reduce his tax bill by deducting the amount of these covenanted payments.
On appeal, Finlay J upheld the Special Commissioner’s decision, concluding that ‘the real nature of the transaction’ was to pay remuneration so long as the gardeners and others remained in their employment.
The Court of Appeal overturned Finlay J’s judgment, focusing on the legal nature of the deeds of covenant.
In the House of Lords (19 TC 490), there was a stark division of opinion. Lord Atkin, who was in a minority of one, looked at the whole arrangement and agreed with the Special Commissioners and Finlay J ‘that the substance of the transaction was that what was being paid was remuneration’ (at 517). This he took to be the legal effect of the documents, construed in the context of all the surrounding circumstances.
Lord Tomlin, though, vehemently disagreed with that approach, using language which became the tax avoiders’ charter for the next 70 years. He raised the question whether in tax cases (at 520): ‘There is a doctrine that the court may ignore the legal position and regard what is called “the substance of the matter”.’
His response was: ‘This supposed doctrine ... seems to rest for its support upon a misunderstanding of language used in some earlier cases. The sooner this misunderstanding is dispelled and the supposed doctrine given its quietus the better it will be for all concerned ... Every man is entitled if he can to order his affairs so that the tax attaching under the appropriate Acts is less than it otherwise would be. If he succeeds in ordering them so as to secure this result, then, however unappreciative the Commissioners of the Inland Revenue or his fellow taxpayers may be of his ingenuity, he cannot be compelled to pay an increased tax. This so-called doctrine of “the substance” seems to me to be nothing more than an attempt to make a man pay notwithstanding that he has so ordered his affairs that the amount of tax sought from his is not legally claimable.’
Lord Russell said very much the same thing about ‘the substance’ of the transaction, but added the additional stricture that the tax had to be imposed by the plain words of the statute (at 524):
‘I confess that I view with disfavour the doctrine that in taxation cases the subject is to be taxed if, in accordance with a court’s view of what it considers the substance of the transaction, the court thinks that the case falls within the contemplation or spirit of the statute. The subject is not taxable by inference or by analogy, but only by the plain words of the statute applicable to the facts and circumstances of this case.’
Lord Macmillan echoed this by adding that (at 727): ‘Your lordships are concerned only with the technical question whether the [duke] has brought himself within the language of the income tax rule as to contractual payments, and I think that he has succeeded in doing so.’
Looking back on this over 60 years later in McGuckian [1997] STC 908, Lord Steyn commented (at 915G) that:
‘In combination these two features – literal interpretation of tax statutes and the formalistic insistence on examining steps in a composite scheme separately – allowed tax avoidance schemes to flourish to the detriment of the general body of taxpayers. The result was that the court appeared to be relegated to the role of a spectator concentrating on the individual moves in a highly skilled game ... The courts became habituated to this narrow view of their role.’
If anything, Lord Steyn may have understated the impact of the Duke of Westminster case. More than just ‘allowing tax avoidance schemes to flourish’, Lord Tomlin’s comments could be taken to encourage any form of tax avoidance so long as it was legal.
On this basis, the Inland Revenue would have to introduce specific legislation to block particular types of tax avoidance as they came to its attention.
Sticky anti-avoidance rules
Given the habit of tax advisers to stare at any new piece of legislation looking for gaps, the Revenue used deliberately imprecise language to address avoidance so that advisers could not see exactly what it was meant to catch.
One of the best examples is the first major piece of anti-avoidance legislation after Duke of Westminster. The 1936 Finance Act (s 18; and still on the statute books as ITA 2007 s 721) attacked the avoidance of tax by the ‘transfer of assets’ abroad where the UK taxpayer still had ‘power to enjoy’ the income produced by those assets. The section made full use of the passive tense that rights are acquired ‘by means of’ transfers of assets, and introduced the concept of ‘associated operations’. So wide is the wording that it also introduced the now familiar let-off of a motive test – that it would not apply if the taxpayer could satisfy the Special Commissioners that the transfers ‘were effected mainly for some purpose other than the purpose of avoiding a liability to taxation’.
Needless to say, despite the wide and sticky wording of the section it did not take long for a taxpayer to have a go at side-stepping it. In the case of Congreve and Congreve v CIR 30 TC 163, the cunning plan was to make a transfer to a UK-resident company which promptly became non-resident. This did not make it past the House of Lords in 1948, who agreed with the Revenue that the section did not say that the transferee had to be non-resident at the time of transfer, and that the transfer did not need to be made by the taxpayer directly, and that in any event the whole lot of events were clearly ‘associated operations’; so the sticky language had caught its prey.
Statutory motive tests developed further: the dividend stripping avoidance legislation when introduced in Finance Act 1960 (s 28; now ITA 2007 s 682) had belt and braces, requiring not only the lack of a tax avoidance motive but a positive requirement that the transaction was ‘for bona fide commercial reasons or in the ordinary course of making and managing investments’ on top. But targeted anti-avoidance still suffered from the intrinsic weakness that it had to be attached to particular sections. So as soon as dividend stripping was plugged, Messrs Tucker and Plummer revised their planning with a structure using covenants to charity rather than dividends.
The dawn of high tax rates and the game of tax avoidance
This final section starts in the middle of the First World War. The war marked the beginning of sustained higher rates of taxation. The standard rate of income tax went from 5.8% at the beginning of the war to 30% by the end, with surtax increasing from 2.5% to a top rate of 22.5%. These increased rates meant that the balance of tax also shifted from indirect to direct tax, the indirect/direct tax ratio being 48/58 in 1912 and 18/82 by 1918 (‘Settlements and the avoidance of tax on income – the period to 1920’ (David Stopforth) BTR 1990 and ‘Just Taxes’ (Martin Daunton) CUP, p 46). Channelling the increases into direct tax in this way meant, no doubt intentionally, that more of the total increase was paid by the better off.
Unsurprisingly this led to some wealthier citizens looking at their tax bills more closely. Covenants to children, which were deductible from the parent’s taxable income, became popular. In turn, this led the Revenue to look more closely at covenants. The first major piece of anti-avoidance legislation followed, in Finance Act 1922 s 20. A covenant was to be ineffective for tax in three situations: if it was for six years or less (hence the Duke of Westminster’s seven years); if it was revocable without the consent of another person (other than the covenantor’s spouse); or if it was to the covenantor’s child for less than the life of the child.
The legislation spawned more cases than just Duke of Westminster. Finlay J, who found for the Revenue in Duke of Westminster, had already decided for the Revenue on similar reasoning in CIR v Clarkson-Webb 17 TC 451 in 1932, where two brothers attempted to circumvent the need for covenants to their own children to be for life by entering into admittedly reciprocal covenants for each other’s children. Finlay J said that (at 457): ‘one has ... to look nonetheless at the substance of the matter ... they were part of an arrangement which must be looked at as a whole’.
Even he, however, had to admit defeat on the wording on one scheme which was doing the rounds. Because the section specified that revocable covenants would still be effective if the consent of another was required, a practice developed of specifying that consent of someone – often the taxpayer’s accountant, or any one of several friends – was required, in the obvious expectation that the accountant would not want to lose his client and that if one friend proved difficult, another might not. The 1933 case of CIR v Firth 17 TC 603 took this to extremes, with a provision that the covenant could be revoked with the consent of the taxpayer’s accountant or some other person named by the taxpayer, other than his wife. Finlay J conceded: ‘the gentleman in question might go first to one person and then another until he got somebody to consent. All I can say is that I think is the result of the section’.
The third situation, covenants to a child for less than the life of the child, was also the subject of litigation which went all the way to the House of Lords, in Wiggins v Watson’s Trustees 17 TC 728 also in 1933: a covenant to the taxpayer’s child revocable with the consent of one of several friends was again found to be effective for tax. So, the first piece of clear and simple avoidance legislation resulted in Revenue losses in all three situations where taxpayers took avoidance steps.
The century of avoidance, followed by avoidance legislation and litigation, had begun.
The author thanks Catherine Gosh for her assistance with this article. This article forms a chapter of the book ‘Plucking the goose: a century of taxation from the Great War to the digital age’ (Tolley, September 2016), which is available to purchase for £9.99 here. Tolley is donating all profits from book sales to the ‘Bridge the gap’ charitable campaign, supported by the CIOT.
