Tax Journal/2003/Issue 695, 2 June/Articles/Tax And Corporate Responsibility - Tax Journal, Issue 695, 2
The non-taxing weekly for top practitioners
Tax Journal, Issue 695, 2
2 June 2003
Tax And Corporate Responsibility
In My Opinion
KPMG Professor of Taxation Law, University of Oxford
© Reed Elsevier (UK) Ltd 2003
Judith Freedman, KPMG Professor of Taxation Law, University of Oxford, argues that the debate on tax avoidance needs to be re-cast
What are company directors and their advisers to think about tax avoidance? Where does it fit, if at all, within the debate on corporate social responsibility (CSR)? So far, tax has figured surprisingly little in this debate. As the CSR concept grows in strength, however, and given the role of tax schemes in the case of Enron, tax is now firmly on the agenda as an issue for company managers and audit committees. The concept of enlightened shareholder value has shifted the emphasis from immediate profit maximisation. Longer term and sustainable profit maximisation is not only consistent with paying regard to wider stakeholder interests but is actually dependent upon doing so. The principle in the Duke of Westminster's case (that taxpayers may organise their affairs to pay the least tax possible under the law) now has to be read in this context. Tax minimisation may not always be in the corporate taxpayer's best interests, widely construed. But on what principles other than that in the Duke of Westminster's case can corporate directors rely when performing the balancing act required of them by CSR?
The need for a legislative framework
The tax debate does not appear to have kept pace with that on CSR. In 1979, the Institute of Economic Affairs published a book entitled Tax Avoision -- The economic, legal and moral inter-relationships between avoidance and evasion
. Twenty-four years later the debate about evasion and unacceptable and unacceptable avoidance seems to have moved on very little, despite all the legislative, judicial and administrative developments that have taken place. At its core the debate seems to remain one about boundaries and a search for certainty in an uncertain world. Morality continues to be a word much used but rarely defined. There is a danger that recent statements are over-emphasising the adversarial nature of the relationship between taxpayers and the revenue authorities. What is needed is a search for sound principles covering common ground, which can guide taxpayers and revenue authorities alike. At the moment the focus is too heavily on defining the boundaries
, rather than finding this common ground.
I shall argue that the evolution of such common ground would be assisted by the introduction of a General Anti Avoidance Principle. (I shall call this principle GANTIP, to distinguish it both from a General Anti-Avoidance Rule or GAAR and from the GAAP of accounting standards.) The judiciary has shown itself unwilling to develop a coherent anti-avoidance principle, with Lord Hoffmann denying that the courts have authority to impose an overriding legal principle superimposed upon the whole of revenue law (MacNiven v Westmoreland Investments Ltd  STC 237). GANTIP could not eliminate uncertainty but would create a legislative principles based framework to guide the actions of taxpayers and revenue authorities and give the courts authority to develop a legitimate principle through case law.
A principle not a rule
I am arguing here for a principle, not a rule. As in the now familiar debate on accounting standards
, so in the context of taxation we can see that detailed rules can result in the phenomenon of 'creative compliance' (McBarnet and Whelan (1991) 54 MLR
848). The form of the rules is followed but the spirit of the legislation is deliberately subverted. The more complexity is introduced into the tax system, the more hooks there are on which to hang avoidance schemes. What is more, as is demonstrated both in the academic literature and in practice, complex legislation does not result in certainty for the taxpayer. This leads writers on regulation and on tax to propose the use of principles rather than rules. But John Avery Jones ( BTR
580) has pointed out the danger that when people write of principles they sometimes just mean less detailed rules. A true principle is not just a less specific rule but is qualitatively different from rules. Rules are all or nothing but principles are not. Principles do not aim at certainty, though they may not decrease it, given the difficulties experienced with detailed complex legislation. Principles can be weighed against each other and fill in gaps in the rules. A GANTIP could be weighed against the principle in the Duke of Westminster's case to assist directors in reaching a balanced view of their duties under the law on tax avoidance.
The Inland Revenue's proposal for a GAAR in 1998, a response to the Tax Law Review Committee's (TLRC's) paper, was rejected largely for lack of certainty (see, for example,  BTR 1-11, Troup; Gillett). Gillett, whilst keen to stop the 'extreme scheme merchants', thought that unacceptable avoidance was a matter for the courts to determine, 'in accordance with the social and political mores of the time'. This, like references to 'morality' in this context, raises difficult philosophical questions about where these mores or morals are to be found, but the immediate problem is that the UK courts have not shown themselves willing to evolve a general anti-avoidance principle on this basis. The confusion engendered by the succession of cases decided has actually made a sensible debate about acceptability and social mores hard to achieve.
The problem with the proposed GAAR was that in response to the justified fear that it would be rejected for uncertainty, both the TLRC's original proposal and that of the Inland Revenue made the principle too detailed and complex. It began to look like another rule, instead of a true principle. Since by definition, principles are not certain, to criticise a GANTIP for uncertainty would be to apply the wrong test.
The spirit of the legislation
It is easy to talk about the spirit of the legislation but more difficult to ascertain what that is. In some of the more extreme tax avoidance schemes it is obvious to all that subversion of the legislation is under way, but this is not always so. The tax system is so complex anyway and is used for so many purposes that it is sometimes not possible to say what is intended by Parliament. Expressing the general intention of the legislature in the case of tax legislation is never going to be simple. In some cases it is purely to raise revenue but often it is mixed. Such expression is particularly important where tax law is being used to provide incentives. This argues for principles-based preambles to statutes in order to aid with interpretation, as suggested by Avery Jones.
The ideal might be for Governments to abandon the use of the tax system to provide incentives but this seems unlikely to happen. The resulting problem is that reality is not a useful test. Government itself has created a climate of artificiality. So we find fiscal incentives available to small companies but not sole traders, and capital allowances based on legal rather than economic ownership. This is legal, but not economic or commercial, reality. It then ill behoves the Government to plead that these incentives were not intended. It is no accident that the recent Barclays Mercantile v Mawson case, which split the views of some of our leading judges, was about capital allowances. Carnwath LJ held that the fact that the essential purpose of the arrangement was to obtain a tax advantage in the form of capital allowances could hardly be an objection to the scheme when capital allowances were the bed-rock of the finance leasing trade ( STC 66, 88). The court looked at parliamentary intention but found that it was to create this artificial world. Within this culture of artificiality, on what principle should the company's directors should have acted any differently from the way they did? And yet the scheme has been attacked by the Inland Revenue and by the media, and it is still very hard to be sure what the final outcome will be.
Lord Hoffmann's 'solution' in Westmoreland is that the only principle of construction is to ascertain what Parliament meant by using the language of the statute. But his distinction of commercial from juristic meanings has not given us certainty. It is part of an ingenious analysis that is ultimately seriously short of predictive qualities. A combination of principles-based drafting on a case by case basis, plus a GANTIP would assist here. The GANTIP would fill in the gaps left by Parliament explicitly and legitimately and would be a genuine principle going beyond pure interpretation. The aim of the GANTIP would not be certainty, but the provision of a framework in which uncertainty could be managed.
Do we need clear boundaries?
The argument for a GANTIP is that we should focus on common ground and not boundaries. Sometimes we may need to recognise the inevitability of uncertainty. What matters then is how we deal with the lack of clarity. A GANTIP would not only give the judiciary the tools and legitimacy it needed to develop a sensible principle, but would also provide a structure for development of the common ground of understanding which is needed in order for such a principle to operate. This would involve procedures, clearance systems and codes of conduct, which could be worked on together by the tax community. There might be 'fuzzy edges' but provided the areas of uncertainty did not affect day to day transactions and were not governed by arbitrariness but by procedures that attracted the support of the compliant members of the tax community, there would be no deficit of the rule of law. (Weisbach (1999) 66 University of Chicago Law Review, 860); Endicott, (2000) Vagueness in Law.)
I am not suggesting the removal of all boundaries. There are circumstances in which certainty is a paramount objective, which overrides the need for inclusivity. Any failure to draw a clear line between criminal evasion and legal (albeit ineffective) avoidance is a deficit of the rule of law because of the consequences for the taxpayer. We need strict rules about what constitutes criminal evasion, even if the need for clarity means that the rules do not cover all the behaviour the authorities would like to catch.
Corporate taxpayers have been unsettled recently because the revenue authorities, concerned by criticism that they are not doing enough to combat evasion
, are arguing that failed avoidance schemes could become evasion. Of course, this is possible. As John Gribbon argued in The Tax Journal
in 1997 (22 September; Issue No.420), 'If an 'avoidance' scheme relies on misrepresentation ... or concealment of the full facts, then avoidance is a misnomer
; the scheme would be more accurately described as fraud'. This we can all agree, but there have been suggestions, not least in the pages of this journal, that recent pronouncements from the revenue authorities are trying to muddy the waters. The Inland Revenue has indicated that there should be no criminal offence where there is no trace of any concealment of the true facts of arrangements for which there is a respectable technical case
. (Tax Bulletin No 49
). This leaves open the problem of deciding what level of disclosure is necessary to ensure honesty. At what point does being economical with the facts or not volunteering information become concealment? Is presentation of the relevant information amid a large volume of detail adequate or must the points at issue be spelt out to the Inland Revenue and highlighted for them? Who is to decide whether there is a respectable technical case? The complexity of the tax system is such that
, as discussed above, there may well be reasonable different views on whether a scheme will work. How definite must advisers be that there is a reasonable case?
Not all evasion is criminal, but it always involves concealment. Whether innocent or dishonest, evasion will lead to re-assessment for tax purposes. Only dishonesty should result in criminal prosecution (CIR v Challenge Corporation  STC 548 at 554). Whether the prosecution is for fraudulent evasion or cheating the public revenue, dishonesty must be proved. The problem with relying on dishonesty is that it depends on a combination of findings of fact about what the defendant knew and believed and an application of the current standards of ordinary decent people (Ghosh  QB 1053). Professor Ashworth suggests that this test derogates from the principle of maximum certainty in the criminal law (Principles of Criminal Law, 2003). He acknowledges that juries may recognise dishonesty easily in some situations, but suggests it is far more difficult in situations with which a jury or magistrates are unfamiliar, of which complex tax schemes are a prime example. Here, imprecision may lean in favour of the defendant by applying the standards of ordinary people, since polls show mixed attitudes to tax evasion. On the other hand, complex schemes undertaken by wealthy individuals and corporate firms may be seen differently by juries from their own activities. A jury may often need more to go on in terms of what is required of a taxpayer than it has at present if it is to be able to take an informed view about the circumstances necessary for dishonesty in complex tax scheme cases.
It is right that the authorities should seek to counteract dishonest behaviour and to combat tax fraud but taxpayers need to know what will be considered dishonest. Juries and judges need to know what the norms of disclosure are and how to assess whether there was a respectable technical case. This is not something to be left to fuzzy principles. This may be an area where indications to taxpayers need to be strengthened by new rules about disclosure.
In the UK, if a tax scheme fails to be effective but falls short of evasion, there is usually no specific penalty. This can encourage taxpayers to try out tax schemes even if there is a doubt about their efficacy. In the United States, by contrast, there are newly strengthened regimes, backed by penalties, designed to flush out early disclosure, which could give pause for thought. In New Zealand, there are penalties for abusive avoidance and for unacceptable interpretations, designed to 'send a clear message from government that the manipulative and aggressive interpretation of tax laws is unacceptable behaviour'. Despite the problems about judging the standards imposed, these systems give a steer to taxpayers and to professional advisers. In a corporate setting they give directors a standard of behaviour by which to assess whether they have taken sufficient advice and what are the risks to the company's reputation. Breach of these requirements could be useful evidence of dishonesty to a jury. Developments of this nature could assist taxpayers and revenue authorities alike in the United Kingdom by clarifying responsibilities and thus the avoidance/evasion border.
Corporate citizenship and taxpaying
Companies cannot be expected to pay voluntary taxes over and above the amounts imposed by law. Unlike other areas of CSR and other discussions about law and morality, this debate is not about whether the citizen should disobey a bad law or go further than the law requires. Despite the Inland Revenue Chairman's recent remarks on the link between taxes and social goods (what he called the 'foundation of tax morality'), the tax avoidance debate is not the place for a discussion of redistributive justice or levels of taxation and expenditure. On these policy issues there will be genuine moral disagreement that cannot be resolved by appeal to generally agreed values. The CSR requirement is that citizens should pay their share of the tax lawfully collected by governments in whatever country they are operating. What is their fair share takes us back full circle to the context and way in which the law should be interpreted and applied. Company directors are increasingly concerned with reputational risk and will not wish to be seen to be poor citizens; but they are also aware of their duty to profit maximise if there is no counter-balancing principle which requires them to pay regard to the objectives of the legislation.
A GANTIP would not be a certain rule but a framework for defining common ground. It would need to be backed by appropriate procedures and codes and, preferably, clearer indications in legislation about objectives. But the GANTIP would provide a balance to the unqualified profit maximisation principle we have at present, bringing tax law into line with 21st century corporate governance.
This article is extracted from the author's inaugural lecture at Oxford University. It states the author's independent views and should not be taken to represent the views of KPMG or any other body.