Righting wrongs

KEY POINTS

• The difficulty of defining unacceptable tax avoidance.
• Choosing suitable candidates for a GAAR advisory panel.
• Is a general anti-abuse rule actually needed?
• Do the courts have sufficient power to override tax avoidance schemes?
• The political forces behind the introduction of a GAAR.

Graham Aaronson QC’s proposed general anti-abuse rule (GAAR) has generated a fair amount of comment since it was published late last year. It remains to be seen whether the government will be persuaded to take the matter forward.

So far as I see it, there are essentially two questions: first, should the country adopt a statutory provision that will have the effect of neutralising any unacceptable tax avoidance that is undertaken and, second whether such a provision should take the form proposed by Mr Aaronson?

This article will focus on the first of those questions, but it is also worth addressing the second, albeit more briefly.

On the whole, when I read the draft prepared by Mr Aaronson, I considered it to represent fairly the kind of rule that had been advocated in the Aaronson report.

In short, it would enable HMRC, when faced with a particular tax-planning arrangement, to counteract the tax advantages arising from the arrangement, but only after jumping through a number of hoops to be laid down by the statute.

However, as I frequently experienced during my involvement with the Tax Law Rewrite Project, there is a fundamental distinction between the statutory codification of a particular policy and a proper discussion as to whether the policy itself is the right way forward. A GAAR would be no exception.

Would the proposal work?

At its heart (both in the policy and the draft) is the problem of defining what constitutes unacceptable tax avoidance. The draft proposes that this question be delegated to a quasi-judicial committee (the advisory panel) which would decide what is and what is not ‘reasonable tax planning’.

However, to my mind, this is essentially the fallacy of the entire proposal.

It is always difficult to define clearly what constitutes tax avoidance, let alone acceptable avoidance: there will inevitably be arrangements that most individuals (even tax practitioners) will accept represent avoidance.

However, the difficulty is identifying precisely where the boundary lies. Similarly, the difficulty will be identifying precisely what does and what does not constitute ‘reasonable tax-planning’.

There will be some individuals who consider the ‘egregious’ schemes that are meant to be targeted by the proposed rule to be perfectly reasonable responses to a particular legislative régime: otherwise, there would not be any taxpayers entering into those schemes, nor would there be any promoters of such schemes.

So the proposed rule will not have the intended effect if such individuals (or their views) are represented on the advisory panel. This could have quite a distortive effect.

Conversely, there are individuals who will assert that any tax-prompted arrangement would constitute (unacceptable) avoidance, even though the courts continue to emphasise that taxpayers are perfectly entitled to take a course of action that reduces their tax liabilities.

For similar reasons, those individuals are also unlikely to be of any real benefit to the advisory panel.

Thus, when one starts to chip away at the potential pool of individuals who are meant to determine what constitutes reasonable tax planning, the exercise becomes rather self-defeating.

However, assuming that one can identify a suitably balanced pool of individuals that can be trusted to debate fairly what constitutes reasonable tax planning, how can one be sure that a particular arrangement would still fall on the right side of the line?

And would that line ever move (either in favour of or against HMRC) over time?

Schemes and scenarios

One does not even need to look at the sophisticated tax schemes that have been litigated over the years to imagine scenarios where people might legitimately disagree.

There are instead plenty of examples of straightforward planning exercises, which I could recommend to clients with a totally clear conscience on the basis that I would personally regard them as a reasonable exercise of choices of conduct open to taxpayers.

See examples    1   ,   2  ,  3 , 4   and 5.

And I expect most readers would share my views. Indeed, some readers might consider my examples to be rather too conservative.

But, and this is where the problem lies, I cannot be sure that everyone would agree with my classification.

In fact, I am almost certain that there would be some readers who would disagree with my classification of what constitutes reasonable tax planning.

And it is the very existence of this uncertainty which, in my mind, militates against any statutory rule.

The difficulty with all this is that there is no single right answer. What is acceptable to one person might be unacceptable to the next. And vice versa.

Exercising caution

Furthermore, I remember that there was one leading practitioner who actively discouraged clients from going down the Arctic Systems route (see example 3): not because he thought it unacceptable tax planning, but because he was sure that it would be subject to a Revenue challenge in due course: and history shows that he was right.

How does one factor in a practitioner’s caution when determining what is reasonable tax planning?

Even if one can overlook that, consider a trainee in a cautious firm. The trainee’s formative tax years would be spent in an environment where something ‘just isn’t done’ and that trainee will be part of the next generation of members of the advisory panel.

How does that trainee subsequently distinguish between client care and reasonableness?

Indeed, I recall my own years in accountancy practice. I qualified with what is now a Big 4 firm where ‘use of home’ allowances were just not heard of; they were not relevant. Nor did they feature in any of my training courses or exams.

But such allowances represented the final entry in the detailed profit and loss account of virtually every unincorporated client at the smaller firm I moved to after qualification.

What one considers to be reasonable is partly a product of what one has seen in practice.

In short, my concern about the proposal is that it would not necessarily give us any certainty. Furthermore, the reasonableness of a particular arrangement will depend on an individual’s own prejudices as well as external factors such as the political and economic climate when the question is asked.

It should, of course, be noted that the proposed boundary has some broad similarities with the threshold of ‘abuse’ as adopted by the CJEU in VAT cases (e.g. Halifax plc v CCE (and related appeals) [2006] STC 919) and I suspect that this is no coincidence.

However, just because the approach has been imposed upon the UK for VAT purposes, it does not mean that the UK must adopt it for direct taxes.

Do we actually need it?

Having concluded that the proposal would not necessarily provide the right approach, I will now address what is strictly the prior question as to whether the UK actually needs a statutory anti-avoidance rule.

This is something that I have been considering for some time, and I have now reached the firm view that no such rule is needed.

And my conclusion is not because of any question of fairness; nor is it to do with the unworkability of any statutory rule. I firmly believe that, contrary to the views expressed by some, the state of the current law is, in fact, as good as we need.

Ever since the landmark Ramsay case 30 years ago (WT Ramsay Ltd v CIR [1981] STC 174), the courts have moved towards a more purposive interpretation of tax statues, so that they would be interpreted using the same principles as other statutory provisions.

As Lord Steyn so pithily put it in  CIR v McGuckian [1997] STC 908: before Ramsay, ‘tax law was by and large left behind as some island of literal interpretation’.

The standard approach now taken when determining whether a particular statutory provision applies to a particular set of circumstances was set out by Ribeiro PJ in the Hong Kong case, Collector of Stamp Revenue v Arrowtown Assets Ltd [2003] HKCFA 46 at paragraph 35:

‘The ultimate question is whether the relevant statutory provisions, construed purposively, were intended to apply to the transaction, viewed realistically.’

Purposive approach

This approach has been fully endorsed as applicable in the UK both by the House of Lords and the Supreme Court.

In my view, the Court of Appeal’s decision in PA Holdings Ltd v HMRC [2008] SSCD 1185 is entirely consistent with this approach.

That case concerned an attempt to turn employment income into dividend income; the Court of Appeal considered the true nature of the payment to be employment income and held that it should therefore be taxed as such.

Moses LJ (with whom the other two judges agreed) did state that the court did not need to apply any ‘post-Ramsay prophylactic against tax avoidance’ in order to reach that conclusion.

However, in my view, the so-called Ramsay doctrine has been sufficiently refined over the past three decades that it is difficult to identify any material differences between the approach that the court did take in PA Holdings and any alternative approach which expressly invokes the Ramsay-doctrine as understood post-Arrowtown.

In other words, the courts rarely take a literalistic approach to the legislation when a more holistic view of a transaction would give a different tax result, particularly in cases where the different tax result is less favourable to the individual taxpayer concerned.

The one that got away

Of course, much of the recent debate about a GAAR has focused on HMRC v D Mayes [2011] STC 1269 – the one where the taxpayer got away.

At the ICAEW’s Wyman Symposium last summer, it was the Mayes case which seemed to typify the kind of egregious scheme that was thought to be real target of, and the justification for, any future GAAR. In my view, there are three possible responses.

First, Mayes is very much the exception and should not be considered as the norm.

Second, in the wake of the PA Holdings case, it is possible that a court might now feel more empowered than before to override the legal form of any transaction to ensure that any tax charge (or relief) reflects the commercial reality.

Indeed, it has been suggested by some that that same approach could have been adopted by the courts (or could have been pressed harder by HMRC before the judges) in Mayes which would have resulted in the taxpayer losing.

However, there is a third response for such cases.

As held by the High Court and the Court of Appeal, the Mayes case concerned a very tightly prescribed tax code where ordinarily commercial transactions would be taxed in a way which was inconsistent with the underlying commercial substance of what was going on.

As the court found, it would have been contrary to any legal doctrine to suspend the legislation just because it gave rise to a result which HMRC did not like.

On that basis, a sensible approach might be just to patch the perceived gap in the legislation as identified in such cases, rather than add a new layer of complexity and uncertainty to the entirety of the tax code.

Will politics intervene?

The difficulty with addressing the question as to the necessity of an anti-avoidance rule is that the answer from a practitioner’s perspective is only one part of the equation.

From a political perspective, the government needs to be seen (by the general public and the lay press) to be taking decisive action against what is considered to be unacceptable tax avoidance.

To some extent, it does not matter that many members of the public do not really understand all the issues: they see that there is a problem and they turn to the politicians to implement a solution.

It would not be the first time that a tax statute has had to wear a sticking plaster that has been applied purely for political reasons.

Therefore, assuming that political expediency dictates that an anti-avoidance rule is to be introduced, then I think that the current proposal is as good as one could expect. In other words, it is not ideal, but it is probably better than any alternative.

I suspect that Graham Aaronson has devised his proposal with this partly in mind and to that extent he definitely deserves our congratulations.

And finally

As an aside, if one is to have a rule that is intended to ensure that no taxpayer escapes paying the ‘fair’ amount of tax (because of a gap in the legislation), should there be an equivalent rule to ensure that no taxpayer pays an excessive amount of tax, simply because the legislation contains an unexpected catch, which is unjustified from any rational policy perspective?

In particular, I recall the case of Allcock v King [2004] SSCD 122 which was decided by a Special Commissioner in March 2003.

The case concerned the question as to whether or not Mrs Allcock was deemed to have annual income of at least £8,500 so as to be what the legislation then called a ‘higher-paid employee’ for the purposes of the benefits code.

Mrs Allcock received a salary, the use of a car and the provision of petrol for the car. If one added together the salary, the car benefit and the fuel benefit, the total would have been below £8,500.

However, what the Special Commissioner had to grapple with was whether, when ascertaining whether or not Mrs Allcock was a higher-paid employee, one also added the value of the reimbursed fuel (because it was funded by a company credit card) even though that figure would be excluded from any subsequent assessment.

The Special Commissioner concluded that it should be so added for those purposes and I do not think it would be possible to doubt the correctness of the decision from a statutory viewpoint.

However, it is the comments along the way that are particularly relevant for the present.

In his opening sentence, the Special Commissioner stated that the case brought to light ‘an anomaly – and a very unfair one’.

Even the tax inspector ‘accepted that there is no justification as a matter of policy for such a curious result’.

However, he added: ‘[because] this is what the law currently requires, … that is the end of it so far as this case is concerned’.

The Special Commissioner considered the taxpayer to have been caught by a legislative ‘trap’ and invited the then Inland Revenue to offer a concession to the taxpayer and/or to suggest a change in the law which would remove the anomaly.

(In due course, the problem was avoided by the introduction of what became extra-statutory concession A104, which was withdrawn when ITEPA 2003, s 219(5) and s 219(6) were repealed.)

The driving force of the GAAR (whatever one calls it and however it might be enacted) is to ensure that taxpayers pay a fair amount of tax based upon the legislative scheme in place.

If one is to be introduced, should there not also be one to rescue taxpayers from the situation faced by Mrs Allcock?

By the way, the Special Commissioner who heard Mrs Allcock’s case was one Graham Aaronson QC.

READ MORE ABOUT THE CASES CITED IN THIS ARTICLE

Keith M Gordon is a barrister, chartered accountant and tax adviser. He practises from Atlas Chambers (tel: 020 7269 7980), and won the chartered tax adviser category at the 2009 LexisNexis Taxation awards. He can be contacted by email
 

Issue: 4345 / Categories: Comment & Analysis , GAAR , Admin